LS. Dena rtment 'El'f of Marine Pregame, I'Jg?rr; afJ'IIvem'L-g rice and Hanan I I DISPRUPDRTIUHATE CONTACT DISPROPORTIONATE MINORITY CONTACT TECHNICAL ASSISTANCE MANUAL Fourth Edition July 2009 U.S. Department of Justice Office of Justice Programs 810 Seventh Street NW. Washington, DC 20531 Eric H. Holder, Jr. Attorney General Laurie O. Robinson Acting Assistant Attorney General Jeff Slowikowski Acting Administrator Office of Juvenile Justice and Delinquency Prevention Office of Justice Programs Partnerships for Safer Communities www.ojp.usdoj.gov Office of Juvenile Justice and Delinquency Prevention www.ojp.usdoj.gov/ojjdp The Office of Juvenile Justice and Delinquency Prevention is a component of the Office of Justice Programs, which also includes the Bureau of Justice Assistance, the Bureau of Justice Statistics, the Community Capacity Development Office, National Institute of Justice, Office for Victims of Crimes, and Sex Offender Sentencing, Monitoring, Apprehending, Registering, and Tracking. Contents Introduction ............................................................................................................. Intro-1 Background and Lessons Learned ......................................................................... Intro-1 Lesson 1: Disproportionality can exist not only in detention and corrections but also in other contact points of the juvenile justice system ...................... Intro-1 Lesson 2: Many factors contribute to DMC at different juvenile justice system contact points, and a multipronged intervention is necessary to reduce DMC ............................................................................................... Intro-2 Lesson 3: Data are powerful tools, and DMC intevention strategies need to be data based .................................................................................................. Intro-2 Lesson 4: DMC reduction requires support from the top .................................. Intro-2 Lesson 5: DMC reduction needs to occur at the local level .............................. Intro-3 Lesson 6: DMC reduction requires strong partnerships .................................... Intro-3 Lesson 7: DMC reduction demands sustained efforts ....................................... Intro-3 Lesson 8: Evidence-based DMC reduction efforts are scarce ........................... Intro-4 Overview of This Manual ...................................................................................... Intro-4 Chapter 1: Identification and Monitoring ................................................................... 1-1 Identification ................................................................................................................ 1-1 The Relative Rate Index Method ............................................................................. 1-2 Implementing the RRI Calculation: Step by Step .................................................... 1-4 Collecting DMC Data To Implement the RRI Tool .............................................. 1-15 Common Issues in DMC Data Collection To Obtain RRI Values ........................ 1-17 Systematic Analysis of DMC RRI Values: Five Steps in Interpreting and Analyzing RRI Values To Drive Decisionmaking ............................................ 1-20 Continued Monitoring of DMC ................................................................................. 1-29 Purpose .................................................................................................................. 1-29 Using RRI Values for Monitoring ......................................................................... 1-29 Interpreting the Results .......................................................................................... 1-32 Important Caveats .................................................................................................. 1-33 Endnotes..................................................................................................................... 1-34 Appendix A: Data Required To Populate the Cells of the DMC Relative Rate Index Matrix .......................................................................................................... 1-35 Appendix B: State UCR Reporting Agencies, 2004 Listing ..................................... 1-45 Chapter 2: Assessment .................................................................................................. 2-1 An Overview of Assessment........................................................................................ 2-1 Stage 1: Generating Possible Explanations: Mechanisms Leading to DMC ............... 2-2 Differential Behavior ............................................................................................... 2-2 Mobility Effects: Importation/Displacement ........................................................... 2-3 Indirect Effects ......................................................................................................... 2-4 Differential Opportunities for Prevention and Treatment ....................................... 2-5 Differential Processing or Inappropriate Decisionmaking Criteria ......................... 2-7 Justice by Geography ............................................................................................... 2-7 Legislation, Policies, and Legal Factors With Disproportionate Impact ................. 2-8 DMC Technical Assistance Manual, 4th Edition • Contents i Contents (continued) Accumulated Disadvantage ................................................................................... 2-10 Stage 2: Identifying Data Types and Expected Results ............................................. 2-10 Stage 3: Obtaining the Data ....................................................................................... 2-15 Planning ................................................................................................................. 2-15 Methods ................................................................................................................. 2-17 Stage 4: Analyzing the Data and Identifying the Most Likely Mechanism(s) Creating DMC ....................................................................................................... 2-25 Conclusion ................................................................................................................. 2-28 References .................................................................................................................. 2-30 Appendix A: Example—Youth Interview (Spectrum Associates) ............................ 2-33 Appendix B: Example—Juvenile Court Coding Instrument for Case Records, Iowa, 2005 ............................................................................................................. 2-41 Chapter 3: Preparation at the Local Level .................................................................. 3-1 An Initial Issue: Financial Support .............................................................................. 3-1 Steps in Local Preparation ........................................................................................... 3-2 Establishing a Steering Committee .......................................................................... 3-2 Identifying Leadership ............................................................................................. 3-2 Reaching Consensus ................................................................................................ 3-2 Conveying a Sense of Urgency................................................................................ 3-3 Setting Priorities ...................................................................................................... 3-3 Organizing the Work ............................................................................................... 3-3 Basic Tasks for the Steering Committee: Examples From the Baltimore City DMC Reduction Initiative ................................................................................................. 3-4 Articulating Local DMC Goals and Objectives....................................................... 3-4 Mediating Discussions To Acknowledge and Respect Differences of Opinion Without Compromising Progress ........................................................................ 3-4 Keeping the DMC Agenda at the Forefront of All Juvenile Justice Activities in the Jurisdiction ..................................................................................................... 3-5 Moving Ahead With Intervention Strategies ........................................................... 3-5 The Burns Institute Process: An Example of Building Community Momentum ........ 3-5 The Juvenile Detention Alternatives Initiative Framework: An Example of Changing the System ............................................................................................... 3-6 Appendix: JDAI Core Strategies Matrix ..................................................................... 3-8 Chapter 4: Intervention ................................................................................................. 4-1 Overview of the Intervention Phase............................................................................. 4-1 Intervention Guidelines................................................................................................ 4-2 Designing a Comprehensive, Multimodal Approach .............................................. 4-2 Prioritizing Strategies To Focus on Critical Decision Points .................................. 4-3 Choosing Interventions That the Community Is Ready To Implement ................... 4-3 Using Evidence-Based Strategies and Drawing on the Successful Experiences of Current DMC Initiatives .................................................................................. 4-3 Planning To Evaluate Effectiveness ........................................................................ 4-4 DMC Technical Assistance Manual, 4th Edition • Contents ii Contents (continued) Conceptual Framework for Selecting DMC Strategies ............................................... 4-4 Types of Strategies for Reducing DMC ...................................................................... 4-6 Direct Services ......................................................................................................... 4-6 Training and Technical Assistance ........................................................................ 4-12 System Change ...................................................................................................... 4-20 Case Studies of Jurisdictional Initiatives ................................................................... 4-27 Bernalillo County, New Mexico ............................................................................ 4-28 Cook County, Illinois............................................................................................. 4-34 Hillsborough County, Florida ................................................................................ 4-39 King County, Washington ..................................................................................... 4-44 Mesa County, Colorado ......................................................................................... 4-48 Multnomah County, Oregon .................................................................................. 4-54 Santa Cruz County, California ............................................................................... 4-59 Endnotes..................................................................................................................... 4-66 References .................................................................................................................. 4-67 Chapter 5: Evaluation ................................................................................................... 5-1 Overview ...................................................................................................................... 5-1 OJJDP’s Performance Measurement Requirements for DMC .................................... 5-2 Logic Models ........................................................................................................... 5-3 DMC Reporting Requirements and Measures ......................................................... 5-5 Performance Measurement Data Collection ............................................................ 5-8 Reporting Periods .................................................................................................. 5-13 Systems for Data Entry and Reporting .................................................................. 5-13 Performance Measurement Technical Assistance ................................................. 5-14 Conducting DMC Evaluations ................................................................................... 5-14 Developing an Evaluation Framework .................................................................. 5-15 Selecting a Research Design .................................................................................. 5-16 Developing a Plan for Data Collection ...................................................................... 5-18 General Considerations .......................................................................................... 5-18 Examples of Data Collection Planning .................................................................. 5-20 Developing a Plan for Data Analysis and Reporting ................................................. 5-32 Analysis ................................................................................................................. 5-32 Reporting ............................................................................................................... 5-32 Reassessing Intervention Logic ................................................................................. 5-33 Overcoming Obstacles to Evaluations ....................................................................... 5-33 Summary .................................................................................................................... 5-34 Appendix A: Disproportionate Contact Performance Measure Definitions and Reporting Format ................................................................................................... 5-35 Appendix B: Hiring an Outside Evaluator ................................................................. 5-42 Appendix C: Model Interagency Agreement—Juvenile Information Exchange ....... 5-43 Appendix D: Sample Consent Forms ........................................................................ 5-45 Appendix E: Sample Participant Locator Form ......................................................... 5-54 DMC Technical Assistance Manual, 4th Edition • Contents iii Contents (continued) Chapter 6: Federal, State, and Local Partnerships To Reduce Disproportionate Minority Contact........................................................................................................ 6-1 OJJDP’s Role ............................................................................................................... 6-1 Determining States’ Compliance With the DMC Core Requirement...................... 6-1 Holding States Accountable for Noncompliance .................................................... 6-2 Providing Financial and Technical Assistance Support to Annual DMC Conferences ......................................................................................................... 6-2 Providing Training and Technical Assistance to State and Local DMC Reduction Efforts ................................................................................................. 6-3 Developing and Implementing Performance Measures To Assess DMC Reduction Activities, Providing Technical Assistance/Tools To Increase State/Local Evaluation Capacity.......................................................................... 6-3 Establishing and Maintaining a DMC Coordinators’ Listserv ................................ 6-3 Establishing and Maintaining a DMC Web Page .................................................... 6-4 Developing a DMC Web-Based Data Entry System To Be Used With the DMC Technical Assistance Manual .............................................................................. 6-4 Publishing the DMC Technical Assistance Manual (3rd Edition) Electronically ... 6-4 Making Direct Awards To Promote Innovative Local DMC Reduction Projects ... 6-4 Instituting Conference Calls Among DMC Coordinators ....................................... 6-5 Planned Activities .................................................................................................... 6-5 North Carolina’s DMC Reduction Initiative ............................................................... 6-6 Partnerships at the State Level ................................................................................. 6-6 State’s Partnerships With Demonstration Counties ................................................. 6-6 Partnerships and Activities in Demonstration Counties .......................................... 6-7 Future Plans for North Carolina’s DMC Initiative ................................................ 6-12 Endnotes..................................................................................................................... 6-13 Appendix: Guilford County’s DMC Reduction Initiative ......................................... 6-14 Chapter 7: Strategies for Serving Hispanic Youth ..................................................... 7-1 Introduction .................................................................................................................. 7-1 Project Overview ......................................................................................................... 7-2 Washoe County, Nevada.............................................................................................. 7-3 Data Collection ........................................................................................................ 7-3 Findings ................................................................................................................... 7-5 Recommendations.................................................................................................... 7-7 Interventions ............................................................................................................ 7-9 Travis County, Texas ................................................................................................. 7-11 Data Collection ...................................................................................................... 7-12 Research, Analysis, and Findings on Violation of Probation ................................ 7-13 Strategic Intervention............................................................................................. 7-16 Recommendations for Additional Activity ............................................................ 7-19 Recommendations for Race and Ethnicity Data Collection .................................. 7-20 Lessons Learned ........................................................................................................ 7-21 Data Collection ...................................................................................................... 7-22 DMC Technical Assistance Manual, 4th Edition • Contents iv Contents (continued) Interventions To Reduce Racial and Ethnic Disparities ........................................ 7-23 Time and Resource Commitments......................................................................... 7-23 Endnotes..................................................................................................................... 7-24 References .................................................................................................................. 7-26 Appendix A: Washoe County Juvenile Services Limited English Proficiency Questionnaire Appendix B: ADOMS II Risk Assessment Tool Appendix C: NCJJ Guidelines for Collecting and Recording the Race and Ethnicity of Juveniles in Conjunction With Juvenile Delinquency Disposition Reporting to the Juvenile Court Judges’ Commission Chapter 8: The Role of State DMC Coordinators ...................................................... 8-1 Administrative Component .......................................................................................... 8-1 Staff Support to the State Advisory Group and the DMC Subcommittee ............... 8-2 DMC Staff Support Plan .......................................................................................... 8-3 DMC Report Card.................................................................................................... 8-4 Budget Oversight and Management ........................................................................ 8-5 DMC Compliance Plan Development and Implementation .................................... 8-5 Program Component .................................................................................................... 8-6 Grant Administration and Management .................................................................. 8-6 Dissemination of DMC Resource Materials .......................................................... 8-10 Training and Technical Assistance ........................................................................ 8-10 Conclusion ................................................................................................................. 8-11 Endnotes..................................................................................................................... 8-11 Appendix A: Sample State DMC Coordinator Job Description ................................ 8-12 Appendix B: Glossary of Terms ................................................................................ 8-14 Appendix C: Juvenile Justice and DMC Resources .................................................. 8-25 DMC Technical Assistance Manual, 4th Edition • Contents v Introduction Heidi Hsia* Background and Lessons Learned In 1988, in response to overwhelming evidence that minority youth were disproportionately confined in the nation’s secure facilities, Congress amended the Juvenile Justice and Delinquency Prevention (JJDP) Act of 1974 (Public Law 93-415, 42 U.S.C. 5601 et seq.). This amendment mandated that the Office of Juvenile Justice and Delinquency Prevention (OJJDP) require all states participating in the Formula Grants Program (Title II, Part B, of the Act) to address disproportionate minority confinement (DMC) in their state plans. Specifically, the amendment required the state, if the proportion of a given group of minority youth detained or confined in its secure detention facilities, secure correctional facilities, jails, and lockups exceeded the proportion that group represented in the general population, to develop and implement plans to reduce the disproportionate representation (Section 223(a)(23)). In its 1992 amendments to the JJDP Act, Congress elevated DMC to a core requirement, tying 25 percent of each state’s Formula Grant allocation for that year to compliance. Ten years later, Congress modified the DMC requirement of the JJDP Act of 2002 to require all states that participate in the Formula Grants Program to address “juvenile delinquency prevention efforts and system improvement efforts designed to reduce, without establishing or requiring numerical standards or quotas, the disproportionate number of juvenile members of minority groups who come into contact with the juvenile justice system.” This change broadens the DMC core requirement from disproportionate minority “confinement” to disproportionate minority “contact,” and it further requires the states to institute multipronged intervention strategies including not only juvenile delinquency prevention efforts but also system improvements to assure equal treatment of all youth. In sum, the broadening of the DMC core requirement in the JJDP Act of 2002 reflects two important lessons learned in the field of DMC in the preceding 12 years: Lesson 1 Disproportionality can exist not only in detention and corrections but also in other contact points of the juvenile justice system. As states have undertaken efforts to reduce disproportionate minority confinement for youth, they have found evidence that disproportionality occurs at every contact point within the juvenile justice system, from arrest to cases transferred to criminal court and not just at detention and correction. Moreover, what happens to youthful offenders during their initial contacts with the juvenile justice system influences their outcomes at the later * About the author: Heidi Hsia, Ph.D., is the Disproportionate Minority Contact Coordinator at OJJDP. DMC Technical Assistance Manual, 4th Edition • Introduction Intro-1 stages, leading to a commonly observed amplification phenomenon (i.e., the extent of minority overrepresentation amplifies as minority youth penetrate deeper into the juvenile justice system). Therefore, to both understand the mechanisms that lead to DMC (which hereafter stands for disproportionate minority contact) and design appropriate intervention strategies to address these specific contributing mechanisms, one must first examine all contact points throughout the juvenile justice system from arrest to transfer to adult court and then target intervention at the relevant and selected priority contact points. Lesson 2 Many factors contribute to DMC at different juvenile justice system contact points, and a multipronged intervention is necessary to reduce DMC. In the first 12 years of the disproportionate minority “confinement” initiative, most states undertook prevention and intervention strategies to reduce delinquency among minority youth. Few states implemented a more comprehensive approach to also make their juvenile justice system response consistently fair. (For a case example of such a comprehensive approach, see the OJJDP Summary, Disproportionate Minority Confinement: 2002 Update, pp. 19–38.) The disproportionate minority “contact” requirement in the JJDP Act of 2002 reaffirms the fact that DMC is the result of a number of complex decisions and events and that only through a comprehensive, balanced, and multidisciplinary approach can the states and localities reduce DMC. Additional important lessons learned in the field of DMC include the following: Lesson 3 Data are powerful tools, and DMC intervention strategies need to be data based. Data are essential to determine if minority youth come into contact at disproportionate rates with the juvenile justice system, at which decision points, to what extent, and for which racial or ethnic groups. Once states and localities have collected and utilized the above data, they must collect further quantitative and qualitative data to determine the factors/mechanisms that contribute to the observed disproportionality. Moreover, these data, collected over time, should allow jurisdictions to compare changes in DMC trends in a particular locale and to examine if specific DMC reduction strategies have led to the intended outcome. In sum, data are powerful tools in guiding every phase of DMC reduction efforts. Lesson 4 DMC reduction requires support from the top. The enactment of the JJDP Act, with the inclusion and the broadening of the DMC core requirement for all states participating in the Formula Grants Program, underscores the strong support for DMC reduction that exists in Congress. OJJDP must diligently enforce this core requirement by setting uniform standards in its annual determination of states’ DMC Technical Assistance Manual, 4th Edition • Introduction Intro-2 DMC compliance status and unfailingly administering the consequences of noncompliance as the JJDP Act specifies: i.e., by restricting the drawdown of 20 percent of that state’s Formula Grant allocation in the subsequent year. At the state level, support from Governors and directors of the state agencies designated to administer federal JJDP funds leads to statewide DMC conferences and significant financial investments in DMC reduction activities. Such support demonstrates the strong state-level leadership required for serious DMC reduction efforts. Lesson 5 DMC reduction needs to occur at the local level. Juvenile justice systems vary from state to state and from locality to locality. DMC reduction efforts must occur at the local level based on the data collected regarding the existence, extent, and nature of DMC; the resource availability versus resource gaps; and a resultant locally developed, comprehensive DMC reduction plan. The systematic execution of the local plan requires the top-down support from local agency directors and bottom-up support from all line workers and other staff throughout the agencies involved in juvenile justice. DMC reduction at the local level at multiple sites collectively reflects DMC reduction at the state level. Likewise, DMC reduction in multiple states leads to DMC reduction at the national level. Lesson 6 DMC reduction requires strong partnerships. As discussed earlier, DMC is the result of a number of complex decisions and events, and the reduction of DMC requires a comprehensive, balanced, and multidisciplinary approach. That multidisciplinary approach implies, horizontally, a partnership of all stakeholders, public and private, at the local, state, and federal levels. To further strengthen horizontal state and local DMC reduction partnerships, OJJDP’s enforcement of the DMC core requirement and provision of financial and technical assistance to states and localities, together with states’ financial and technical assistance to localities, help create the vertical partnerships that are required for successful DMC reduction efforts. Lesson 7 DMC reduction demands sustained efforts. DMC is a complex and longstanding phenomenon that demands steadfast and sustained efforts. OJJDP, states, and localities must sustain the top-down and bottom-up support and horizontal and vertical partnerships described above to enable continuous DMC reduction efforts that are succinctly depicted in the following diagram. DMC Technical Assistance Manual, 4th Edition • Introduction Intro-3 The DMC Reduction Cycle Lesson 8 Evidence-based DMC reduction efforts are scarce. The ultimate success of the DMC initiative is measured not only by the number of states in compliance with the DMC core requirement but also by the effectiveness of the DMC activities in actually reducing the minority overrepresentation at every decision point of the juvenile justice system. Therefore, measuring or evaluating outcomes must be an integral part of all DMC reduction activities. Despite the expansion of the DMC core requirement from “confinement” to “contact,” the purpose of the DMC core requirement remains the same: to ensure equal and fair treatment for every youth in the juvenile justice system, regardless of race and ethnicity. OJJDP has incorporated the lessons this Office and the field have learned over the years into this Disproportionate Minority Contact Technical Assistance Manual (3rd Edition) to provide the field with up-to-date guidance on continuing DMC reduction efforts across the country. States and localities should use this manual along with a companion tool that OJJDP also provides: the Web-based Data Entry System at www.dsgonline.com/dmc. Localities and states can enter raw data concerning the volume of activities by race and ethnicity at different juvenile justice decision points to calculate the existence and extent of DMC as expressed by the Relative Rate Indexes. Overview of This Manual Chapter 1, Identification and Monitoring, presents in detail this new and improved method to calculate and analyze disproportionality. Chapter 1 and chapter 2, Assessment, also discuss lessons learned about the power of data, specifically, how jurisdictions can use data to facilitate their decisionmaking regarding where and how they should focus their efforts to recognize and understand their specific DMC issues. Chapter 3, Preparation at the Local Level, builds on what DMC reduction efforts need to occur at DMC Technical Assistance Manual, 4th Edition • Introduction Intro-4 the local level and illustrates ways local communities can prepare to undertake a multidisciplinary and collaborative approach in this area. Chapter 4, Intervention, has been developed through an extensive literature search, which found few DMC reduction programs that have been proven effective. Nevertheless, this chapter endeavors to match intervention strategies with identified factors/mechanisms that contribute to disproportionality and to identify strategies that are promising or “developing” (i.e., are likely to become promising with further development and evaluation). Chapter 5, Evaluation, illustrates the differences between performance measurement and evaluation and how evaluation builds on performance measurement. To enhance the applicability of this chapter, ways to assess the performance and evaluate the effectiveness of some commonly used strategies to reduce disproportionality at detention are discussed in detail. Lastly, chapter 6, Federal, State, and Local Partnerships, describes the importance of strong horizontal and vertical partnerships at all levels in facilitating and sustaining DMC reduction activities and presents examples from North Carolina. This manual is published electronically to capitalize on three technological advantages: Wide distribution at low cost. In-text hyperlinks that allow readers to immediately access reference materials and sources under discussion. Annual updates at low cost as new knowledge and resources emerge for chapters 1 through 5 and new state examples emerge for chapter 6. For example, few DMC reduction strategies have been shown to be effective. Those that have are in the area of reducing disproportionality at detention, in part because of the more than 10 years’ focus on minority overrepresentation in confinement. OJJDP hopes to expand, over time, chapter 5 to include an increasing number of promising and/or effective approaches to reduce disproportionality at all system contact points. Similarly, as more states systematically invest in financial support of their local DMC reduction sites, aggressively seek and/or provide technical assistance support to them, and track progress made in these sites, OJJDP will feature these efforts in chapter 6. DMC Technical Assistance Manual, 4th Edition • Introduction Intro-5 Chapter 1: Identification and Monitoring William Feyerherm, Howard N. Snyder, and Francisco Villarruel* Identification When a jurisdiction enters into an effort to identify where disproportionate minority contact (DMC) may exist within its juvenile justice system, there are at least three reasons to do so: To describe the extent to which minority youth are overrepresented in that jurisdiction’s juvenile justice system. To begin to describe the nature of that overrepresentation. By collecting and examining data on the volumes of occurrence at major decision points in the juvenile justice system (e.g., arrest, referral, diversion, detention, petition/charges filed, delinquent findings, probation, confinement in secure correctional facilities, and transfer to adult court), one can determine whether overrepresentation exists, where it exists within the jurisdictions, and the degree of overrepresentation at those points within the juvenile justice system. To create a foundation for ongoing measurement of DMC and provide the basis for monitoring activity. This is an ongoing process that is repeated—preferably annually, but at a minimum of at least every 3 years. Although one may think of the identification phase as the first step in a jurisdiction’s DMC efforts, it is also an ongoing process. OJJDP requires all states to collect these data statewide and from their targeted local DMC reduction sites on a continuing basis (updated at least every 3 years with the submission of a new 3-year comprehensive juvenile justice and delinquency prevention plan).1 The primary purpose of the identification phase is descriptive—it provides a quantitative answer to the question, are there differences based on race and ethnicity in the contact that youth have with the juvenile justice system?2 In addition, this phase provides initial guidance on what questions to ask (assessment) about the mechanisms and reasons for such differences. These purposes are summarized by the following questions: Are there differences in the rates of contact (e.g., arrest) that are based on race or ethnicity? If so, at what stages of the justice system are these differences more pronounced? * About the authors: William Feyerherm, Ph.D., is Vice Provost of Research at Portland State University in Portland, Oregon. Howard N. Snyder, Ph.D., was the Chief of Systems Research at the National Center for Juvenile Justice in Pittsburgh, Pennsylvania. Francisco Villarruel, Ph.D., is a professor and fellow in the Department of Family and Child Ecology at Michigan State University in East Lansing, Michigan. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-1 Are there differences in the processing of juveniles within the justice system that are based on race or ethnicity? If so, at what stages of the justice system are these differences more pronounced? Are the racial or ethnic differences in contact and processing similar across jurisdictions within a state? If not, in which jurisdictions are these differences more pronounced? Are the differences in contact and processing similar across all racial and ethnic groups? If not, which groups seem to show the greatest differences? Are racial or ethnic differences in contact and processing changing over time? It is important to note what is not included at this stage: any attribution about the reasons for the differences. Therefore, the identification phase of information neither describes the reasons for any differences that occur nor creates strategies to reduce those differences. The identification stage, and the use of the Relative Rate Index in this stage, is designed to help narrow the field of inquiry for the assessment stage, which describes the process of identifying the likely mechanisms that create differences in juvenile justice system contacts for minority youth and which, in turn, leads to the intervention stage. Both are described in later chapters of this manual. The Relative Rate Index Method Overview The method that OJJDP has selected for the identification stage is termed the Relative Rate Index (RRI). This method involves comparing the relative volume (rate) of activity for each major stage of the juvenile justice system for minority youth with the volume of that activity for white (majority) youth. The RRI provides a single index number that indicates the extent to which the volume of that form of contact or activity differs for minority youth and white youth. In its simplest form, the RRI is simply the rate of activity involving minority youth divided by the rate of activity involving majority youth. The RRI method involves the following general components (a more detailed description of the specific steps is provided later): The number of events in various stages of the juvenile justice system is tallied for the minority groups of interest, generally those groups that the federal Office of Management and Budget specifies as necessary for data collection (Hispanic and non-Hispanic members of the following racial groups: African American, Asian American, Native Hawaiian and other Pacific Islanders, Native Alaskan, and American Indian). The number of events is translated into rates of activity by dividing the number of events in one stage by the number of events in a preceding stage. For example, one divides the number of probation placements by the number of adjudications— situations in which youth were found delinquent—to determine the rate of probation placement. This calculation is performed separately for each minority DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-2 group in which the size of that group’s youth population is at least 1 percent of the total youth population in the jurisdiction. The rates for minority groups are compared to the rates for white (majority) youth by dividing the rate for each minority group by the rate for white youth. This creates an RRI, which provides a numeric indicator of the extent to which the rate of contact for minority youth differs from the rate of contact for white youth. Each RRI is tested to determine if it is statistically significant, that is, whether it differs sufficiently from a neutral value (1.00) such that the differences in the rates are not likely to be the result of random chance processes. Characteristics The RRI method has a number of features or characteristics that one must understand to interpret the results. First, one must calculate the relative volume (rate) of activity involving minority youth and contrast it with the relative volume of activity involving white youth. By using rates of activity to reflect the relative volume of activity at each stage, the process can take into account the relative size of the white and minority populations and the relative amount of activity in preceding stages of the justice system. However, this method is not the same as calculating the odds of particular types of contact because one is not tracking individual youth across time; instead, one is comparing the relative volume of activity within a specific time period. That relative volume may be created by the rapid turnover (churning) of a few youth or may be the result of a lower level of involvement of a large number of youth. A second major feature of the RRI method is that it involves a stage-by-stage calculation of these relative rates or relative volume. This is important because it shows the incremental increase or decrease in contact levels as youth move through the justice system. It would be unrealistic to assume that differences in processing of minority and white youth are constant across the various decision stages of the justice system. Moreover, it would be unrealistic to assume that the same stages of the justice system account for disproportionate minority contact across all justice systems. By basing the rate calculation on the volume of activity in the preceding stage of the justice system, one can examine the changes in rates of contact as youth of a certain racial or ethnic group move through the system. A third major feature of the RRI method is that it minimizes the extent to which calculations of differences between groups depend on accurate census information. The previous method of calculating disproportionality for each contact stage—by dividing the percentage of minority juveniles represented at that stage by the percentage of minority juveniles in the jurisdiction’s total juvenile population at risk for juvenile court involvement—was based entirely on comparison with the percentage representation in the population. This created several types of problems; notably, in many instances it appeared that the general population census amounted to a significant undercount of minority populations. The effect of such an undercount was to dramatically increase the previously recommended index or measure of disproportionate contact— Disproportionate Representation Index (DRI)—in which all stages of the juvenile justice DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-3 system were compared with the percentage distribution of race and ethnicity in the general census numbers. With the use of the RRI, once one moves past the first stage (arrest) in the justice system, a significant problem in the census numbers will have no marked effect on the RRI values. A fourth useful feature of the RRI method is that it does not require a transactional data system that tracks youth throughout the juvenile justice system. Indeed, the method does not require that the data available to describe the justice system all come from a single data system. It is possible to mix multiple data sources, although doing so raises concerns about common definitions of race and ethnicity as well as concerns about the comparability of the counting and classifying rules used in multiple agencies. Fifth, one of the attributes of the RRI method is that as long as the data are counted in a consistent fashion for a particular stage within the jurisdictions being examined, the method can relatively easily accommodate differences from some standard definitions in the particular counting rules. For example, in some states it is possible to obtain a count of the number of youth who are subject to secure detention each year. In other states, detention data are maintained by counting the number of juvenile cases in which detention is used, and in still other states it is possible to count only the number of detention episodes in which a youth is checked into a detention facility. Each of these methods will, of course, yield a different number, and that difference in numbers will yield a rate that seems to have a very different scale (e.g., the rate of detention episodes is likely to be much higher than the rate of youth detained). However, as long as the method of counting is applied uniformly to youth of color and white youth, the index value—the ratio of the rates—will actually be comparable across the three examples used. It will represent the general degree to which the rate of detention activity (however measured) will differ between youth of color and white youth. Indeed, the RRI values for jurisdictions using these different definitions can still be roughly compared to determine the differential detention contact rates for minority youth, even though the absolute measures of detention contact may be on different scales. However, if at all possible, each jurisdiction should maintain the same definitions from year to year to reduce the possibility that changing definitions may appear to indicate that the DMC levels in that jurisdiction are changing. Implementing the RRI Calculation: Step by Step The following materials provide step-by-step instructions for completing the initial identification stage for examining disproportionate minority contact within a jurisdiction. These instructions should provide some guidance in the analysis process, by both specifying the steps to take (including data, data definitions, and basic descriptions of the juvenile justice system) and providing an example to follow, using a data tool developed for the purposes of this analysis. What follows is an example of an actual jurisdiction with a fairly typical juvenile justice system. As a first step in understanding the example, and the analysis process, we have created a general model of the juvenile justice system (figure 1). Cases flow between major stages in the justice system and are depicted in such a way that one can follow the major DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-4 components and can record the number of cases passing through each stage during a year. The number of cases is used to compute a rate of occurrence, and those rates are compared among racial and ethnic categories. So, for example, one may calculate arrest rates for white youth and for Hispanic youth, comparing these two rates to determine the extent to which Hispanic youth may have a higher arrest rate than white youth. The result of that comparison is the RRI. It must be emphasized that the RRI is a first step in examining disproportionate minority contact. The RRI points to areas for more intensive examination and provides an ongoing set of ―vital signs‖ or an ―early warning system‖ for the management of the juvenile justice system. The following sections discuss each step of the RRI calculation process. Figure 1: Relationship of Data Elements for Relative Rate Index Calculations DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-5 Step 1: Understanding System Elements The use of the RRI methodology begins with understanding the basic relationship of the elements in the juvenile justice system and then comparing those elements to the general model in figure 1. Figure 1 does not show all of the possible pathways that a case involving a juvenile might follow in the juvenile justice system. Rather, it shows the major flows and the major points at which data are likely to be available. Because much of the RRI model is based on the relationship of these elements, each jurisdiction should confirm that its juvenile justice system generally fits the model. If there is not a good fit, then the jurisdiction must modify the model, either by changing the location of some decision points or by adding others. A jurisdiction may have to change its model (e.g., if diversion occurs only after a juvenile has been found guilty/delinquent, or probation can be ordered without a finding of delinquency) or may add another decision point to its model if an important decision stage exists in the local justice system that consistently generates reliable data that can be used to calculate relative rates. In many instances represented in figure 1, there are double-headed arrows between the stages—for example, between referrals and diversion. This indicates that some cases are indeed returned from diversion to the legal/court process due to violation of conditions or other reasons. The important feature, however, is that the total number of diversions is counted, both those resulting in an exit from the system and those resulting in return to further processing. Step 2: Defining Data Elements Next, gather the definitions for each data element. This means gathering both the legal definitions for the action (e.g., the definition of an arrest for the jurisdiction, or the definition of diversion, probation, etc.) and the operational definition for that stage (e.g., what action actually creates the data to count the number of instances of diversion, arrest, or a sentence of probation?). Given the variety of forms of juvenile justice data collected across the nation, two issues, in particular, need to be addressed. For each data element, there is a preferred type of data based on the congressional mandate to address total contact of youth with the juvenile justice system. First, for those data elements that involve ―holding‖ a youth in a particular status, the preferred information is that which identifies the total number of youth in that status during the year, not just the number of new entries into that status during the year. For example, the preferred data element would be the total number of cases in which youth are subject to confinement during the year rather than a count of the new admissions to secure confinement over the year. Likewise, there is the issue of whether data elements reflect ―duplicated‖ or ―unduplicated‖ counts. For example, if a youth is arrested four times during a year, does this count as one youth arrested (unduplicated) or four arrests of a youth (duplicated)? Again, given the congressional mandate to address total contact with the juvenile justice system, the preferred type of data is the duplicated count, one reflecting the total number of youth contacts with the justice system. As part of implementing a national data collection system for DMC issues, OJJDP has created a set of standard definitions for each of the stages in the juvenile justice system depicted in figure 1. These definitions are provided in table 1. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-6 Table 1: Standard Definitions for Each Stage in the Juvenile Justice System Stage Definition Arrest Youth are considered to be arrested when law enforcement agencies apprehend, stop, or otherwise contact them and suspect them of having committed a delinquent act. Delinquent acts are those that, if an adult commits them, would be criminal, including crimes against persons, crimes against property, drug offenses, and crimes against the public order. Referral Referral is when a potentially delinquent youth is sent forward for legal processing and received by a juvenile or family court or juvenile intake agency, either as a result of law enforcement action or upon a complaint by a citizen or school. Diversion The diversion population includes all youth referred for legal processing but handled without the filing of formal charges. Youth referred to juvenile court for delinquent acts are often screened by an intake department (either within or outside the court). The intake department may decide to dismiss the case for lack of legal sufficiency, resolve the matter informally (without the filing of charges), or resolve it formally (with the filing of charges). Detention Detention refers to youth held in secure detention facilities at some point during court processing of delinquency cases (i.e., prior to disposition). In some jurisdictions, the detention population may also include youth held in secure detention to await placement following a court disposition. For the purposes of DMC, detention may also include youth held in jails and lockups. Detention should not include youth held in shelters, group homes, or other non-secure facilities. Petition/ charges filed Formally charged (petitioned) delinquency cases are those that appear on a court calendar in response to the filing of a petition, complaint, or other legal instrument requesting the court to adjudicate a youth as a delinquent or status offender or to waive jurisdiction and transfer a youth to criminal court. Petitioning occurs when a juvenile court intake officer, prosecutor, or other official determines that a case should be handled formally. In contrast, informal handling is voluntary and does not include the filing of charges. Delinquency findings Youth are judged or found to be delinquent during adjudicatory hearings in juvenile court. Being found (or adjudicated) delinquent is roughly equivalent to being convicted in criminal court. It is a formal legal finding of responsibility. If found to be delinquent, youth normally proceed to disposition hearings where they may be placed on probation, committed to residential facilities, ordered to perform community service, or receive various other sanctions. Probation Probation cases are those in which a youth is placed on formal or court-ordered supervision following a juvenile court disposition. Note: Youth on ―probation‖ under voluntary agreements without adjudication should not be counted here; they should be part of the diverted population instead. Confinement in secure correctional facilities Confined cases are those in which, following a court disposition, youth are placed in secure residential or correctional facilities for delinquent offenders. The confinement population should not include all youth placed in any form of out-of-home placement. Group homes, shelter homes, and mental health treatment facilities, for example, would usually not be considered confinement. Every jurisdiction collecting DMC data must specify which forms of placement do and do not qualify as confinement. Transferred to adult court Waived cases are those in which a youth is transferred to criminal court as a result of a judicial finding in juvenile court. During a waiver hearing, the juvenile court usually files a petition asking the juvenile court judge to waive jurisdiction over the case. The juvenile court judge decides whether the case merits criminal prosecution. When a waiver request is denied, the matter is usually scheduled for an adjudicatory hearing in the juvenile court. If the request is granted, the juvenile is judicially waived to criminal court for further action. Juveniles may be transferred to criminal court through a variety of other methods, but most of these methods are difficult or impossible to track from within the juvenile justice system, including prosecutor discretion or concurrent jurisdiction, legislative exclusion, and the various blended sentencing laws. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-7 In some instances, a jurisdiction may have access to the local data required to support these standard definitions for each stage of processing, using the preferred units of count (e.g., cases placed in confinement, or number of arrests). In other jurisdictions, the ideal data may not be available. In many instances, such jurisdictions may have alternative definitions that the available data may support. Such alternative definitions and data are accepted in the OJJDP DMC data entry system as long as they are carefully defined and consistent over time. Therefore, persons who construct a jurisdiction’s RRI must develop a comprehensive understanding of the types of available information about its juvenile justice system processing and select from among those available data the ones that best represent each processing stage. In other words, these researchers must become experts in data that can be harvested to fulfill the DMC goals that OJJDP has established. To assist in this process, this chapter includes an appendix (see appendix A) that serves as a primer of the nature and sources of available data that researchers may use to populate the RRI matrix. Although no single source can meet all user needs, this appendix provides a sound foundation for those faced with the task of quantifying DMC at the jurisdictional level. When one uses alternative definitions, he or she should note the definition and sources of data at appropriate locations in the data entry screens provided in the online data tool. Step 3: Determining Racial/Ethnic Categories The next step is to determine the available race and ethnicity categories for each data element. This means determining not only what groups are counted but also what the source is for that classification (e.g., self-identification, classification by officials, records from other sources). This will also involve determining whether the classification is a single label for each youth, a set of possibilities (e.g., Hispanic and Asian), or a ―check all that apply‖ format. When possible, determine whether the classification system can be converted to follow the U.S. Census Bureau classification as referenced in the OJJDP regulations. Step 4: Entering Information Into the Data Tool Once the racial/ethnic categories are determined, gather the counts of events involving youth in each of the various stages (A–J) classified in each racial/ethnic category and enter that information into the data entry module of the data tool (see table 2). The data tool analysis of DMC data is available on the Internet at http://www.ojjdp.dmcdata.org. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-8 Table 2: Sample State Data for Entry Into the DMC Model Black or African American Other/ Mixed White A. Population at risk (ages 10 through 17) 1,097,108 184,372 65,596 27,925 3,564 ** B. Juvenile arrests 69,759 34,754 7,975 845 39 ** C. Referrals to juvenile court 22,175 12,682 2,531 227 29 1,683 D. Cases diverted 3,588 1,121 275 32 3 222 E. Cases involving secure detention 6,541 5,596 1,378 43 7 115 F. Cases petitioned (charges filed) 14,904 9,273 1,898 165 21 916 G. Cases resulting in delinquent findings 10,373 5,778 1,380 109 12 538 H. Cases resulting in probation placement 5,239 2,792 710 64 5 313 148 153 58 1 0 6 91 84 13 0 0 9 J. Cases transferred to adult court Asian American Indian or Alaska Native Data element I. Cases resulting in confinement in secure juvenile correctional facilities Hispanic Native Hawaiian or other Pacific Islanders Note: Data elements correspond to figure 1. ** See the discussion of these two entries in step 5. Assuming that the jurisdiction has been set up for data entry, the first step is to find and enter population data for the age range that is potentially subject to the jurisdiction of the juvenile court. Although many resources are available in some states to obtain these data, OJJDP provides a consistent resource on the Easy Access to Juvenile Populations Web site (http://www.ojjdp.ncjrs.org/ojstatbb/ezapop/). After the population data are entered into the tool for a jurisdiction, the tool calculates whether a specific racial or ethnic group meets the 1 percent rule, at which point OJJDP requires that the jurisdiction examine this group separately. In this instance, examine DMC separately for Native American and other/mixed groups. Identify the jurisdiction (state, county, or other entity) and the dates DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-9 that the data cover, along with the relevant age range for youth at risk of contact with the juvenile justice system (in this instance, ages 10 through 17). The cells for entering this information, as well as the entry areas for the numeric data, are highlighted in yellow in the data tool. The only other information that is needed for the DMC data tool is the total state juvenile population for the age range under consideration. In this example, the age range is 10 to 17 years, and the total state population for this age range is 1,377,550. Step 5: Determining the Availability of Data for Racial and Ethnic Groups Next, determine which racial and ethnic groups are available for analysis. Ideally, a state will have the information available on each of the seven groups shown across the top of table 2. There are, however, several situations in which that may not be so. The numbers presented in table 2 are actual state data that present some of the difficulties a state may encounter. The two empty data cells for other/mixed-race youth, represented with two asterisks, are absent for specific reasons. With respect to the population entry, the estimation derived from the National Center for Juvenile Justice source provides no estimates for mixed- or multiple-race youth; these estimates are spread across the other groups. Second, the law enforcement systems in the state provide no arrest information on mixed-race youth; it simply is not in their set of categories. The juvenile court system, on the other hand, does report and record the categories (as shown in table 2). It is impossible, however, to know how to distribute the numbers of mixed race or other youth across the other categories of youth. This mixture of classification methods across the population estimates and across multiple juvenile justice data systems raises a quandary—there is no accurate way to make categories completely consistent across the data entry system. For example, one could estimate the number of cases involving mixed or ―other‖ youth at the arrest stage, but the basis for such an estimate would be questionable. It might be possible (but not easy) to go back to population counts of the 2000 census, but the population estimates available for more recent years do not have all categories—they estimate only the major groups. It might be possible to distribute the number of cases involving mixed-race youth across the other stages in the juvenile justice system (e.g., the referral, detention, and other stages) but that could leave the results open to some challenge. Keeping the category of ―other/mixed,‖ as in the example, permits examination of whether any particular issues occur later in the system (e.g., in transition from referral to detention or conviction). Keeping this category means that the degree of DMC for some groups will probably be underestimated because other ―other/mixed‖ youth will be in the population estimates and in the arrest information for those other groups but not in the referral, detention, and other counts. As a result, the rates of activity will be somewhat lower than if one had better information, which, in turn, means that estimates of DMC will tend to be slightly lower than the actual extent of DMC. Step 6: Determining the Availability of Base Numbers Determining what base numbers are available for calculating the rates is necessary at this stage. In figure 1, those numbers that the authors recommend as base numbers on which to calculate a rate are in rectangular boxes down the center of the figure. For example, in DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-10 calculating the rate of secure confinement (oval I in figure 1), the authors suggest that the appropriate base number be the boxed count listed for the number of delinquent (guilty) findings. In this example, the rate of confinement for white youth is 1.43 per 100 delinquent findings, 100 (148/10,373), and for African American youth the rate is 2.65 per 100 delinquent findings, 100 (153/5,778). Given a situation in which that base number is not available, the authors recommend using the preceding boxed number—in this example, the number of petitions (charges) filed (see table 3). The data tool will automatically select the preceding base for the rate if the preferred base is unavailable (all zeros). Table 3: Identifying the Numerical Bases for Rate Calculations Stage / Contact point Preferred base for rates Arrest Rate per 1,000 population Referral to juvenile court Rate per 100 arrests Diversion (prior to adjudication) Rate per 100 referrals Detention Rate per 100 referrals Petition/charges filed Rate per 100 referrals Delinquency finding Rate per 100 petitions/charges filed Placement in probation Rate per 100 delinquency findings Placement in secure correctional facility Rate per 100 delinquency findings Transfer to adult court Rate per 100 petitions filed However, if data for any stage of the analysis is missing, then the rates will be calculated on the basis of the stage preceding the calculation for which data is available. For example, when examining referral data, the preferred base for referral rates is the number of arrests. If arrest numbers are not available, then the population numbers (the preceding stage with data) will be used. Those analyzing the RRI data must therefore identify the numerical base used for each rate calculation and then understand which stages of the juvenile justice system (figure 1) to use to calculate those rates. If data are missing from one or more stages of the justice system, you will need to identify the base for each rate calculation. At this point, analysis of the index values becomes more complex. For example, in table 2, assume for a moment that arrest information was not available. Although the preferred rate for calculation of court referral rates is the rate per 100 arrests, because arrest numbers are unavailable, the rates are calculated per 1,000 youth. If that were the case, the referral rate for white youth would be 20.2 referrals per 1,000 youth; for African American youth, the rate would be 68.8 referrals per 1,000 youth. The resulting RRI value would be 3.40, leading to the conclusion that the referral process is the source of greatest disparity in the contact experiences of African American youth. However, the full data show that, in this instance, the greatest disparity is in the processes DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-11 that lead to arrest—whether that means the behavior of youth, the community processes that lead to the involvement of law enforcement, or the actual processes of arrest. The point is that interpretation of incomplete data is more difficult, leads to even greater ambiguity in identifying stages for examination, and therefore underscores the importance of seeking more complete information. Step 7: Retrieving the Reports Once the information is entered in the appropriate analysis tool, two reports are available for the basic RRI analysis. Examples of each report are provided below. The first report is race specific; it provides the rates of contact for white youth and a specific minority group and then provides the RRI value, along with an indication of whether the value is statistically significant (discussed below). This report is valuable in that it provides both the RRI value and the rates that were used to calculate the RRI value. Those rates may be examined to determine if the jurisdiction has a level of contact that is higher or lower than other jurisdictions. The second report is a summary of all RRI values for all minority populations, which provides a snapshot of the level of disproportionate contact that all groups may have (or not) with the juvenile justice system. Although this report also indicates which values are statistically significant, it does not display the rates of contact. Step 8: Identifying Situations in Which Index Values Cannot Be Calculated Although it does not occur in this example, there may be situations (particularly for smaller counties and for stages toward the bottom of figure 1) in which no white youth were processed in a particular stage. For example, if no white youth were transferred to adult court, the rate of adult court transfer is zero, meaning that it is impossible to calculate an RRI for that stage (this would require division by zero, which is mathematically impossible). There are two additional situations in which one might calculate a value but in which its interpretation would be questionable. The first of these is when the volume of activity is extremely low (less than five events in the target stage for the group being examined, e.g., less than five instances of African American youth transferred to adult court). The second is when the base number for calculating the rate (the denominator of the rate) is less than 50. In both of those instances, a small fluke occurrence might lead to an abnormally high (or low) number of events (e.g., transfer to adult court) and, given a small base number for calculating rates, a small change in the number of transfers would translate into a large change in the rate of transfers. In other words, at some point it is no longer feasible to examine such data and believe that the examination really provides a pattern of systematic behavior within the justice system, as opposed to a number that might fluctuate greatly on the basis of relatively small actual changes in the justice system. In both of these situations, the data models that OJJDP uses in its data analysis system will not provide numerical answers but rather will indicate that there are insufficient numbers to produce reliable results. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-12 Report 1: Race-Specific RRI Calculations DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-13 Report 2: RRI Summary Values DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-14 Collecting DMC Data To Implement the RRI Tool First, it is essential to become familiar with appendix A: Data Required To Populate the Cells of the DMC Relative Rate Index Matrix. This document provides an excellent road map from which to harvest population data and volume of activity data for all nine juvenile justice system decision points. A number of situations exist in which the basic RRI model described above may be insufficient for the analytic needs of the identification stage. In addition to the calculations and issues of data manipulation, additional factors to consider include data availability, defining the minority groups to be studied, and pushing the RRI process so that it begins to point to areas to study further in the assessment process. Specifying System Stages To Be Examined Specifying the stages of the justice system to be examined is perhaps the most frequent situation in which jurisdictions modify the RRI process. This variation on a theme is played out in two directions. First, it may be the case that a jurisdiction lacks access to sufficient data to describe some of the stages outlined in figure 1. For example, some communities do not maintain sufficient records to adequately explore such stages as the diversion decision or the decision to refer a youth to the juvenile courts. As noted above in the discussion of the sample jurisdiction, when a stage is missing (court referral in the example above), the rate calculations for the stages following that missing stage (e.g., the cases in which a petition is filed) are based on the volume in the preceding stage (in this instance, the number of arrests). That substitution has several impacts that must not be overlooked. First, the RRI value that results from this calculation no longer represents simply the effect of one major decision, but the effects of two—both the referral to the juvenile court and the subsequent decision to file a petition of delinquency. Although the resulting RRI number for filing of petitions is labeled as ―filing of petitions,‖ it is likely to be a larger number than the comparable stages in other jurisdictions because it is the accumulated effect of two sets of decisions. Moreover, in terms of helping to focus attention on an appropriate stage for assessment, if the referral stage is missing, then one does not know whether to target the assessment study on that referral stage or on the subsequent stage of filing a petition. That will make the assessment study more difficult to design, more expensive to conduct, and less likely to actually pinpoint the areas in which intervention is most likely to be productive. Thus, the more missing stages that occur within the RRI analysis for a jurisdiction, the more problematic it will be to conduct an assessment and focus on changes within that system in a manner that will have maximum impact on reducing DMC. Although it is possible to calculate the RRI values with simply the population in a jurisdiction and one other set of numbers (e.g., the volume of admissions to secure confinement), such information would be of relatively little value in identifying areas of the justice system that might benefit from a variety of possible interventions. Beyond that, it would be unlikely that such a single set of numbers would be of much value in assessing the impact of changes in the justice system over time. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-15 Adding a Stage to the Analysis The second variation on this theme is in the opposite direction—what can be done when a jurisdiction believes it must add another stage to the analysis? In this instance, assuming that data of appropriate quality exist to describe such a decision stage, the difficulty is to add a stage to the analytic model in a way that augments the jurisdiction’s ability to make sense of the addition and also to compare this jurisdiction to others within the state or region. There are, of course, some very good policy reasons to add another stage or to subdivide cases into sets handled through a discretionary pathway as opposed to those prescribed by legislation or other agencies. The additional wrinkle in such an addition is that the analytic model that OJJDP tools use to calculate the RRI are relatively tightly integrated. It is not feasible to simply add a column or row to the models. As a result, those jurisdictions wishing to add a stage to their justice system model should contact their OJJDP state representatives to discuss and request technical assistance regarding that addition. In any event, one of the most critical elements of the state effort must be to ensure that all participating jurisdictions use consistent definitions of terms and data collection methods. This is especially important if some jurisdictions within a state are more likely to be home to the majority of minority youth. To obtain an accurate statewide picture of DMC issues, those jurisdictions with significant numbers of minority youth should record information using the same definitions and processes as other jurisdictions. If this is not the case, it is likely they will introduce some distortion into the measurement of DMC because of the differences in definitions and processes. Selecting Minority Groups to Be Examined Standards and guidelines. In addition to the stages of the justice system, the RRI process relies on identifying appropriate minority groups to be examined throughout the process for evidence of DMC issues. Several standards come into play in this selection. First, the basic selection of groups to be examined follows direction of the Office of Management and Budget (OMB), which has devised guidelines and groupings for addressing the issues of race and ethnicity and collecting such data. OMB’s guidance is available on the White House Web site, http://www.whitehouse.gov/omb/. In addition to the OMB information, a number of other fields, such as the study of health disparities (see the Health Research and Education Trust Disparities Toolkit at http://www.hretdisparities.org/), have gathered additional advice. Beyond the guidance of such general sources of information, jurisdictions may examine the census estimates for a particular state or jurisdiction. In general, as an OJJDP requirement, states should analyze information on each group that comprises 1 percent or more of the general youth population (e.g., the ages at risk of contact with the juvenile justice system or of coming under the jurisdiction of the juvenile court system). Issues in counting Hispanic youth. It is clear that additional issues arise in the identification of groups. These issues are addressed in depth in chapter 7 of this manual, on DMC of Hispanic youth, so the discussion here is simply an introduction to the issues raised. The rapid growth of Hispanic/Latino communities in the United States, for example, raises a relevant challenge. Hispanics, as a pan-ethnic group, can represent multiple races, DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-16 depending on national origin (e.g., black, indigenous, or of European or Asian descent). As such, the identification of race for recent immigrants is more a foreign term than a term of meaning—ethnicity is more relevant. Similarly, generational status and acculturative stress may reflect more meaningful information for intervention but may represent challenges for data collection. Such challenges and stress may even extend to the selection of language to be used; for example, whether the local community prefers the terms Chicano, Latino, or Hispanic may be a source of tension. Although recommendations for data collection have been offered,3 one important issue is that the terminology be consistent across jurisdictions and across agencies within a jurisdiction. Consistent with Guidelines for Collection and Recording the Race and Ethnicity of Juveniles in Conjunction with Juvenile Delinquency Disposition Reporting to the Juvenile Court Judges’ Commission, discussed in Chapter 7 and included in its entirety as an appendix, OJJDP recommends that jurisdictions ask two questions to more accurately determine the issue of ethnicity and race for youth in the system: (1) a question about ethnic identification (Hispanic, Latino, or the appropriate local terminology) and (2) a question about racial identification. When jurisdictions ask one question instead of two, they lose not only important information but also information that is critical to accuracy. Without a true count of Hispanic youth in the justice system, jurisdictions cannot accurately assess the need for bilingual/bicultural staff and services, written materials in Spanish, certified translators, culturally appropriate programs, and so on, nor can they determine whether dollars allocated to services for Hispanic youth are sufficient and whether monies have been judiciously spent. Moreover, generational status or length of time in the United States may influence linguistic competencies in multiple languages, not just English. Such information is critical to providing needed services for youth whose linguistic choice is non-English. Potential inconsistencies in data definitions. A state or jurisdiction with multiple data systems may encounter problems if these systems use inconsistent methods to collect data about race and ethnicity. This may lead the jurisdiction to identify the same youth in several ways as he or she travels through the justice system, primarily because the data collection systems have different classification schemes and categories into which they subdivide their clients. This is essentially the problem that was previously presented in table 2. Although it may be possible to creatively identify combinations of categories in which the data systems may be treated as consistent, one should exercise great care whenever comparatively analyzing the data from classification systems that differ with respect to race and ethnicity. Common Issues in DMC Data Collection To Obtain RRI Values Missing Data Elements In many juvenile justice information systems, it is not feasible to obtain all of the data elements for the complete RRI model. For example, information on some stage such as referral or diversion may not be recorded or is reported only in a format that does not include race and ethnicity. In such instances, there is little alternative but to exclude these stages from the analysis. As noted earlier, this will have the effect that the rates that are DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-17 available may be based not on the immediately preceding stage of the system but on a stage that is one or more levels ―upstream‖ from the preferred stage. As an example, if the referral information is not available, then the volume of diversion, detention, and petitions filed will all be calculated as rates per 100 arrests rather than as referrals (the preferred base for these rates). And if the arrest information is also unavailable, then the rates will be calculated on the basis of population: ―per 1,000 in the population.‖ In either of these cases, the interpretation of the resulting RRI value is more difficult for two reasons. First, the value is likely to be higher than for other jurisdictions because it contains the impact of DMC in two (or more) major decision stages. Second, because the value contains the impact of multiple decision stages, it is more difficult to identify specific areas that may need attention for the assessment and intervention stages. Unavailable Hispanic Arrest Data National FBI procedures for the Uniform Crime Reports data do not require that jurisdictions report data on non-Hispanic groups or on Hispanic groups. As a result, the analysis of arrest information for Hispanic youth may be problematic. In many jurisdictions, this problem may be avoided by using state or local data systems for law enforcement records that provide information in the required format or that may be programmed to provide such information. In other jurisdictions, however, this may not be available, which creates several difficulties. First, the data on arrests of Hispanic youth are probably included in other groups’ data, the arrest index value will be missing for Hispanic youth, and their impact will be included in other groupings, including white youth. Second, the referral index value will be missing for Hispanic youth because there is no base number for arrests with which to calculate rates of referral for Hispanic youth. One solution for this situation is to ignore the use of arrests as a base (set all arrests for all youth to zero) and then all referral rates will be comparably based on the population. For a more thorough treatment of these issues, see chapter 7 in this volume: DMC Data and Intervention Strategies for Serving Hispanic Youth in the Juvenile Justice System. Data Definitions and OJJDP Recommendations Don’t Match Each state has a separate juvenile code, and in many states there are multiple information systems that capture data according to slightly different operational definitions. The recommendations given earlier (table 1) are only recommendations, not absolute requirements. The goal is to adhere as closely as possible to the meaning and intent of the OJJDP recommendations. This will undoubtedly require some interpretations and some decisions at the state level. One of the critical issues in the application of these interpretations is that the same definition needs to be applied within a state across time, across each racial/ethnic group, and across jurisdictions. As long as the definitions are applied consistently, no biases are introduced into the use of the index. One of the virtues of using an index is that it compares the relative occurrence of events—the emphasis being on the word relative. As a result, the differences in definitions between state juvenile justice codes will have a considerable impact on the volume of activity when one state is compared to another. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-18 However, when the relative treatment of population segments is compared, the differences in definitions should not have as much impact. A Small Volume of Activity Is Hard to Measure This issue of small numbers may refer to the overall number of youth in the system, particularly because those numbers shrink as we examine the ―end‖ of the system, or it may refer to the relative scarcity of some racial/ethnic subgroups. In either instance, small numbers make it more difficult to use statistical procedures to understand what is happening in the justice system. The net effect of such small volumes of activity is to decrease the power of statistical tests. In other words, it becomes more difficult to separate ―real‖ cases of disparities in handling of minority youth from the differences that may occur because of random variation. As a general approach, the reports on the DMC data Web site are set up not to provide results in two situations: the number of events is less than five, or the base for calculating the rate is less than 50 events. Under those circumstances, it is difficult to detect real disproportionate contact for minority youth. Strategies for increasing statistical power include at least two methods: combining multiple years and combining jurisdictions. For example, the state might combine several counties into one region for analytic purposes. In either case, there is an unavoidable tradeoff in which the specificity of being able to locate disparities in a particular year or in a particular place is weakened in order to gain the ability to use a statistical test to see if there are actually disparities at work in the more general setting. Sometimes there is just not enough activity to use statistical procedures; other sources of information may be needed, such as qualitative studies. In some instances, it may be necessary to rely on other processes, including interviews, observation, focus groups, or conversations with the community at large. In earlier years, the OJJDP minimum standard was that the state must examine at least three counties. The selection of these counties reflects the counties with the greatest proportions of minority youth within their juvenile population, as well as reflecting those jurisdictions within the state that contain the greatest numbers of minority youth. The intent of the minimum standard is to enable the state subsequently to make data-driven decisions in selecting appropriate local jurisdictions for targeted DMC reduction efforts. More recently, due to the increasing recognition of the importance of implementing DMC reduction at local levels and the increasing number of states with targeted DMC reduction sites, OJJDP has required that states track DMC data from their DMC reduction sites on a regular basis (preferably annually but every 3 years at a minimum). Therefore, a state should collect data on all counties that are likely to be (or become) specific targeted or pilot sites for DMC activities in the foreseeable future. The state should select which counties to track with some care because the expectation is that, for purposes of monitoring the projects, there will be continuity in the set of counties that are the subject of state reporting on a recurring basis. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-19 A Racial “Minority” May Be the Statistical Majority There are a growing number of jurisdictions in the United States in which one or another group that has been historically denoted as a ―minority group‖ is actually the statistical majority, and a larger number of jurisdictions in which the ―white, non-Hispanic‖ group is not the statistical majority. These changes in demography raise the question of whether the RRI values should be calculated with reference to the white, non-Hispanic rates of contact. From a mathematical perspective, the groups will remain in the same relationship to one another regardless of which group is being used as the reference group, the basis of comparison. However, for purposes of presentation and understanding, it may be useful to change the reference group that is used. The DMC tool allows states to specify the group to be used as the reference point for RRI calculations; this may be separately specified for each county within a state. Homogenous Communities With Few White Youth There are some communities in which the overwhelming majority of youth in the community are members of a racial ―minority group‖ rather than being members of the ―white, non-Hispanic‖ group. Although there may not be issues of disproportionate minority contact within such a community, the experiences of youth in that community may be so different from other places within the United States that DMC is created on a larger scale (statewide or nationally). In other words, even if all youth have the same experience, it is possible that, compared to minority youth in other jurisdictions, they may experience markedly greater (or lesser) contact with the justice system—in ways that may create many of the issues related to DMC—even if there is little apparent disparity as reflected in the RRI values for the jurisdiction. In the absence of a statistically appropriate reference/comparison group within the jurisdiction, state or national rates of juvenile justice activity may be used to determine if the youth in this jurisdiction have different experiences that warrant DMC attention. National DMC information may be obtained from the national DMC databook, http://ojjdp.ncjrs.gov/ojstatbb/dmcdb/index.html. If it is appropriate to obtain other forms of comparative rates, the OJJDP state representatives can assist in designing the request for such comparative data through a technical assistance request. Likewise, there are likely to be some instances in which differences between various minority groups within a jurisdiction need to be studied rather than comparing their differences with white youth. The DMC Web site will permit the analysis of one minority group in comparison to another by changing the reference group from white, nonHispanic to some other group. Systematic Analysis of DMC RRI Values: Five Steps in Interpreting and Analyzing RRI Values To Drive Decisionmaking Experience with the RRI process over several years has created a series of steps in the analysis of index values in order to drive decisionmaking within a community. The objective is to identify a small set (maximum of 3–5) contact point/racial group DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-20 combinations that will be the focus of later assessment, intervention, and evaluation work. In other words, the objective is to select points of focus for the ongoing DMC activity within a community. The five-step process involves: 1. Identifying those RRI values that are statistically significant. 2. From among RRI values that are statistically significant, identifying those with the greatest magnitude, that is, those that reflect the greatest degree of disproportionate contact. 3. From among statistically significant RRI values, identifying those that involve the greatest volume of activity, that is, the largest number of minority youth who potentially may be affected. 4. Comparing the RRI values noted in step 2 or 3 with the range of RRI values across other jurisdictions and noting which jurisdictions may be particularly different from the others. 5. Examining the local context for each of the RRI values identified in steps 1–4 to consider which jurisdictions may be the more feasible target populations for activities designed to reduce disproportionate minority contact. Step 1: Assessing the Statistical Significance of RRI Values In statistics, a result is termed significant if there is statistical evidence that a difference in rates is unlikely to have occurred by chance. A statistically significant difference simply means there is statistical evidence that there is a difference; it does not mean the difference is necessarily large, important, or significant in the usual sense of the word. Statistically significant does not mean that a difference is big or important. A statistically significant difference does mean that there is statistical evidence that a difference in rates is unlikely to have occurred by chance. In other words, we can have confidence that 95 times out of 100 the difference was not random. In the data analysis system provided by OJJDP on the DMC Web site, those RRI values that are statistically significant at a level of 95 percent confidence are designated by red numbers in bold font. Standard statistical textbooks will warn that the ability to reach statistical significance (statistical power) is a combination of the size of the difference between groups and the number of observations. In this case the number of observations will be the volume of case activity occurring on an annual basis. Thus, some RRI values that appear very large, but involve very few cases, may not be statistically significant. The reverse may also be true: RRI values that reflect apparently small differences in the processing of minority youth may be significant if they involve large volumes of activity. Nonetheless, the first filter for analysis and interpretation of the RRI values is for statistical significance because those are the areas in which there is sufficient confidence that, indeed, some level of disproportionate minority contact is occurring. Step 2: Examining the Magnitude of RRI Values Among those RRI values that are statistically significant, some will appear to be more important than others. The objective in this second step is to identify the RRI values in DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-21 which the highest degree of disproportionate contact occurs. The philosophy here is to ask a community to focus on those stages in their justice system where the greatest degree of difference between racial and ethnic groups occurs. It is useful to recall that the RRI is created by dividing the rate of minority contact by the rate of majority contact. If the two rates are equal, then the resulting index value will be 1.00. Values that are both more than and less than 1.00 thus reflect disproportionate contact. In two stages of the system diagram, lower volumes of activity will reflect a disadvantage for minority youth. If the diversion index is less than 1.00, this means that a lower rate of diversion is used for minority youth. In the probation index, a value less than 1.00 typically means that a more restrictive option than probation is being used for minority youth. In these instances, a low value will be problematic for minority youth. However, because the range of values only goes between 1.00 and 0.00, it may not appear that the index values for these stages are as ―serious‖ as those for other stages, in which the more usual values range up from 1.00 without an upper boundary. To deal with this situation, it is useful to consider that an index of 2.00 represents the same degree of disproportionate treatment as an index of 0.50: Both indicate that one group has a rate that is two times greater than the other. Similar equivalences exist between 3.00 and 0.33, 4.00 and 0.25, and 5.00 and 0.20. Thus, an index of 0.33 for the diversion stage would reflect the same degree of disproportionate contact as an index of 3.00 for the referral stage. Table 4 summarizes the areas in which RRI values greater than or less than 1.00 are of greatest concern: Table 4: Relative Rate Index (RRI) Values Area of concern Decision stages or contact points More than 1.00 Arrests Referrals to juvenile court Cases involving secure detention Cases petitioned Cases resulting in delinquency findings Cases resulting in confinement in secure juvenile correctional facilities Cases transferred to adult court Less than 1.00 Cases diverted Cases resulting in probation placement Note: RRI values that cause DMC concern can be greater than 1 or less than 1. Step 3: Examining the Volume of Activity As cases progress through the juvenile justice system, the total volume of activity tends to decrease. The point of that observation is that one criterion for determining where to focus DMC reduction efforts is to examine those locations within the justice system that may have the impact on the greatest number of minority juveniles. From that vantage DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-22 point, it may be useful to identify those contact points (among the ones that are noted as being statistically significant) in which the greatest numbers of minority youth are involved. In applying this logic, it is important to remember that, for the diversion stage, the number of cases that is of concern is really the number of youth who could have been diverted but were not. From that perspective, the relevant volume would be the numbers of youth not diverted. Examining the volume is essentially a judgment call based on the total number of minority youth in the juvenile justice system of the jurisdiction. As a starting point, it may be useful to identify no more than 8–10 combinations of decision points and racial groups in which the volume appears to be the greatest (5–6 would be a better target, if achievable). Step 4: Examining the Comparative Magnitude of RRI Values Relative to Other Jurisdictions In choosing points of contact for further examination and assessment, it may be useful to compare the degree of DMC in a jurisdiction that has the same points of contact as other communities. For example, if a community has a relatively high RRI value compared with other places, it may be the case that standard practices or policies from other jurisdictions may be introduced in ways that facilitate the reduction of DMC. On the other hand, if an RRI value is relatively lower than in other places (even if it is statistically significant), it may be relatively more difficult to achieve additional DMC reduction. The information needed to conduct this comparative analysis may be generated by examination of the collective entries in the OJJDP Web-based DMC data system. At the time of this report, data for 715 jurisdictions have been entered in this system. A tool for comparing a specific jurisdiction to the combined data from these 715 jurisdictions is available in the DMC Tools section of the OJJDP Web site. Specific instructions are available on this Web site and are included in the following section on county comparisons. The information from these more than 700 jurisdictions is used to place each RRI value into the percentile grouping it represents, relative to all jurisdictions, or to all jurisdictions within specific population size groupings. These percentile groupings represent the percentage of jurisdictions that have an RRI value equal to or smaller than the jurisdiction being examined. In most instances, a smaller percentile value means that the RRI value for that decision stage and racial grouping is ―better‖ in the sense that the jurisdiction has an RRI value lower than many of the other jurisdictions and thus has less of an issue with DMC. For diversion and probation RRI values, however, that interpretation is reversed: A smaller value means that the extent of DMC is higher than in other jurisdictions. The 50th percentile ranking would mean that the jurisdiction is at or very near the median RRI value for that decision point and racial grouping. The objective of this criterion is to find those jurisdictions that reflect the highest relative degree of DMC (a higher percentile for most decision points, but a lower percentile for diversion and probation). The basis for comparison. During 2006–2007, data were entered on the OJJDP Web site for 715 jurisdictions. It should be noted that the data reported during this year include statistics from earlier years because of normal lags in data reporting. In addition, because DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-23 those lag periods differ among the states, and because this was the first full year of using the Web-based data collection system, the data reported for these localities were from several different years. Because DMC data usually do not change dramatically across time, such rough comparisons still have value for those who wish to compare localities. To maximize the utility of this DMC Local Data Comparison Tool, OJJDP plans to update it periodically. In order to facilitate comparisons for roughly equivalent-sized jurisdictions, the 715 jurisdictions have been grouped into approximate thirds on the basis of the size of the total juvenile population, as follows. There remains an option for comparison with the entire set of jurisdictions. Additional comparisons may be added over time and will be clearly noted on the comparison tool as they become available. Less than 5,000 total youth. 5,000 through 19,999 youth. 20,000 or more total youth. All jurisdictions. Using the RRI comparison tool. The comparison tool is available from the OJJDP ―DMC Tools‖ Web page (http://ojjdp.ncjrs.gov/dmc/tools.html) or as an Excel spreadsheet (downloaded from http://web.pdx.edu/~feyerhw). The following pages illustrate the output and reports available from the tool. In the example, data from one community are used to illustrate the mechanics of using the tool and an interpretation of the results (see figures 2 and 3, following the DMC tool examples). County Name Select County Size for Comparison Comparison to be used: ANYWHERE County 1 (1 = Large, 2 = Medium, 3 = Small, 4 = Comparison to all counties) Compared with jurisdictions reporting in 2006–7 and having 20,000 youth or more. Asian Native Hawaiian/ Pacific Islanders American Indian or Alaska Native Other/ Mixed All minorities 1.13 0.52 * 1.45 * 2.40 0.83 0.95 1.01 * 1.16 * 0.87 4. Cases diverted 0.90 1.06 0.82 0.90 * 0.93 5. Cases involving secure detention 1.28 0.96 1.05 * 1.86 * 1.22 6. Cases petitioned 1.40 1.54 1.70 * 1.72 * 1.46 Black or AfricanAmerican Hispanic 2. Juvenile arrests 5.67 3. Referrals to juvenile court * DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-24 7. Cases resulting in delinquent findings 0.93 0.84 1.23 * ** * 0.93 8. Cases resulting in probation placement 1.00 1.05 1.02 * ** * 1.01 9. Cases resulting in confinement in secure juvenile correctional facilities 1.12 0.74 ** * ** * 1.07 10. Cases transferred to adult court 1.79 2.88 ** * ** * 2.03 Note: All cells for entry are highlighted in yellow. Instructions: 1. Enter the name of the jurisdiction in cell B1. 2. Select the appropriate comparison group for this jurisdiction, based on the total number of youth of an age at which they might be eligible for juvenile court jurisdiction. 3. Enter the RRI values for each population group and stage of the juvenile justice system. This may be handled by typing the values or by a cut-and-paste operation from the DMC Web site or RRI spreadsheet. 4. Select the type of comparison you want from the tabs at the bottom of the page: (a) The Percentile tab shows the percentile groupings for the RRI values and provides some general guides to interpretation of the findings. Start your analysis using the percentile tab. (The other tabs and materials make more sense, once you understand the ranking of your county on this page of results.) (b) The Combined Percentile tab graphs the percentiles for all groups and decision stages. (c) The Percentile Charts tab shows the percentiles for all decision stages separately by group. (d) The RRI Comparison Charts tab contains separate charts for each population group—showing the range of RRI scores for that group across all stages—and plots the RRI score of your jurisdiction in the range of scores for comparable sized jurisdictions. The range of RRI values is shown by the 20th and 80th percentile as well as the median RRI value for each decision stage. ANYWHERE County Percentile Groups Compared with jurisdictions reporting in 2006–7 and having 20,000 youth or more. Arrest Referral Diversion Detention Petition Delinquency Probation Corrections Transfer Black 95 15 70 30 95 30 70 25 65 Hispanic 55 20 85 15 99 5 80 10 75 Asian 90 95 80 85 95 95 90 0 0 Native 0 0 0 0 0 0 0 0 0 Other 0 0 0 0 0 0 0 0 0 DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring All 70 15 75 40 99 25 75 25 75 1-25 Note: No information is presented for "Hawaiian or other Asian and Pacific Islands" youth because fewer than six jurisdictions reported on that population segment. A value of zero means that the data are not present or are not reported. General interpretation: This jurisdiction has an RRI score for the designated demographic group and decision point in the juvenile justice system that is numerically higher than the indicated percentage of the jurisdictions reporting to the OJJDP Web site. For example, if the value in the cell for "Arrest" and "All" is 70, this would mean that this jurisdiction has an RRI value that is numerically greater than 70% of the jurisdictions reporting to OJJDP. Users of this information should keep in mind that the RRI scores for the diversion and probation stages are typically the reverse of the other stages: A lower numeric value typically means underutilization of diversion and probation (two less restrictive, therefore desirable options). However, a greater numeric value means greater utilization of diversion and probation for minority youth. Figure 2: Percentile grouping: RRI Values for African-American Youth Compared with jurisdictions reporting in 2006–07 and having 20,000 youth or more 100 90 80 70 Percentile 60 50 40 30 20 10 0 Arrest Referral Diversion Detention Petition Delinquent Probation Corrections DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring Transfer 1-26 Figure 3: Percentile grouping: RRI Values for Hispanic Youth Compared with jurisdictions reporting in 2006–7 and having 20,000 youth or more 100 90 80 70 Percentile 60 50 40 30 20 10 0 Arrest Referral Diversion Detention Petition Delinquent Probation Corrections Transfer Step 5: Considering Whether the Agency and Community Context Support DMC Reduction Jurisdictions use the identification stage, in large part, to select fruitful areas (combinations of decision points and racial groupings) to focus their assessment and intervention efforts. The selection of such areas must be informed not only by the statistical properties of the RRI values but also by practical considerations such as the following: Is the agency involved in that decision point amenable to change? Have there been recent events (public relations issues) that make a change in DMC patterns more or less likely? Are funds or resources available that might assist (or hinder, if lacking) the DMC effort at this decision point? Is strong leadership available that is committed to addressing DMC issues? Are best practices models for this decision point available and applicable? Is there support for DMC reduction within the affected minority group and within the political leadership of that group? Are there issues with the affected minority group regarding media attention at this decision point (e.g., potentially high visibility events that could generate support or resistance for DMC)? DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-27 It is important to consider these issues and decide whether the context for a particular decision point and racial group combination represents a viable selection for intervention. If not, the data should still be monitored, but other areas can be selected as initial targets for attention. Combining the Five Steps The RRI Analysis Tracking Sheet (table 5) shown below is designed to organize the results of this systematic analysis. Within each of the cells in table 5, it is possible to insert up to five letters: ―S‖ for those combinations that are statistically significant, ―M‖ for the subsets that have the greatest magnitude, ―V‖ for the subsets that represent the greatest volume of minority youth potentially affected, ―Cm‖ for the subsets that have the highest comparative level of DMC concerns, and ―Ct‖ for the subsets that appear to have a supportive context for DMC reduction. Ideally, there will be a small set of combinations of the contact stage and racial grouping in which four or five of the letters will be present in the applicable cell. These are reasonable targets for the assessment and intervention stages of the DMC process. Table 5: RRI Analysis Tracking Sheet Contact Point Black or AfricanAmerican Hispanic Asian Native Hawaiian or other Pacific Islander American Indian or Alaska Native Other/ Mixed 2. Juvenile arrests 3. Referrals to juvenile court 4. Cases diverted 5. Cases involving secure detention 6. Cases petitioned (charges filed) 7. Cases resulting in delinquency findings 8. Cases resulting in probation placement 9. Cases resulting in confinement in secure juvenile correctional facilities 10. Cases transferred to adult court Directions: Use this tracking sheet to organize your analysis. Fill out the chart with RRI and volume data, using the following keys: S = significant, M = magnitude, V = volume, Cm = compared with medians of 715 counties. Ct = Context is supportive of DMC reduction. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-28 Continued Monitoring of DMC Purpose The purpose of the monitoring activity is at least threefold: The ultimate question that jurisdictions must answer is: Has DMC been reduced? Whether such a change is directly attributable to specific DMC efforts is a secondary issue that requires a specific evaluation, but the first issue for any community is determining whether a high rate of DMC has been reduced and whether the rate of DMC is increasing or decreasing over time. When rates of DMC change, adjustments can be made in the intervention strategies—selecting the next targets and making sure that past gains in DMC reduction are not lost and that the system is managed in a consistent manner. Monitoring and providing feedback of simple data may encourage change. Positive results may provide tremendous encouragement for DMC efforts. The ongoing monitoring of DMC rates keeps the issue alive and fuels the urgency to reverse DMC. Using RRI Values for Monitoring This involves assessing multiple years of information and looking for patterns in the data. The following section includes examples of some of the patterns jurisdictions might expect. The RRI scores and the graphic materials represent the actual results in several counties in a Midwestern state. Constant values. In sample graph 1, relatively flat RRI values indicate system stability and generate greater confidence that the RRI pattern reflects real differences in minority contact rates. In this instance, there is a pattern that African American youth have a greater volume of arrest activity relative to that of white youth and that this pattern is relatively consistent across time. The same consistency applies to the RRI values for all minority youth. In this county’s instance, the arrest stage was not targeted for DMC intervention, and the display simply indicates that not much has changed here. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-29 Sample graph 1 RRI Values for Juvenile Arrests in County A 3 2 2002 I RRI 2.5 1.5 2003 1 2004 0.5 0 African-American All Minorities Increasing RRI values. Sample graph 2 shows a second pattern of increasing RRI values that may appear over time. In this instance, the growth generated a concern that the arrest area for African American youth shows an increasing level of DMC and, therefore, should be examined carefully to become part of ongoing intervention efforts. Sample graph 2 7 6 5 4 3 2 1 0 2002 2003 I RRI RRI Values for Juvenile Arrests in County B 2004 AfricanAmerican American Indian All Minorities Decreasing RRI values. In sample graphs 3 and 4, which show examples from the same county, it appears that DMC issues are headed in an appropriate direction, whether this is due to system change and interventions or to natural changes such as demographic or economic shifts. In this instance, however, because the target for intervention in the DMC arena involved court processing, it appears possible that the intervention has had a desired impact on DMC issues within the court system. A more extensive evaluation would be required to support such a conclusion, but the results are promising. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-30 Sample graph 3 RRI Values for Transfer to Adult Court in County C 2 1.5 2003 1 I RRI 2002 2004 0.5 0 African-American All Minorities Sample graph 4 RRI Values for Secure Detention in County C 2 2002 1 2003 I RRI 1.5 2004 0.5 0 African-American All Minorities Mixed results over time. The system may be unstable relative to DMC issues; that is, the findings for any single year may be a statistical artifact. If the swings are pronounced and the base of volume is large, as is the case with county A (sample graph 5), this may be an area of the system that is undergoing considerable stress and change, an area to watch carefully with respect to DMC and as a possible target for systematic intervention. Discussions with county personnel revealed that the overall use of secure confinement in this county is declining. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-31 Sample graph 5 RRI Values for Secure Confinement in County A 4 3.5 3 2.5 2002 2 2003 1.5 2004 1 0.5 0 African-American Hispanic All Minorities Interpreting the Results Relating any changes in rates or index values to the assessment results and to intervention strategies becomes important. In the brief discussions above, one may see some relationship between changes in DMC levels and possible intervention strategies. In the case of county B, the discussion of these results generated a hypothesis that importation, an influx of African American youth, may be one mechanism by which increasing levels of DMC are created (see chapter 2 for more details on importation). The results of monitoring activity alone are not sufficient to establish that the intervention(s) designed to address DMC are responsible for any changes observed in the monitoring process. For that purpose, an evaluation study is more appropriate; some approaches to that topic are outlined in chapter 5. However, as a first set of information, the monitoring process can provide jurisdictions a sense of whether things are moving in the intended direction. As part of the process of interpreting results, it is also useful to consider other potential explanations for changes. This consideration will give the analysts and policymakers who are working on DMC issues a greater understanding of the context in which they are operating and the way in which they may productively use the monitoring results. Jurisdictions must consider at least four alternatives in addition to simply concluding that the DMC intervention is working as planned (or not working at all). The first of these is the prospect of changes in statutes and/or interpretations of statutes and policy. To the extent that the juvenile justice system has changed (e.g., significant new statutes or changes in decisionmaking authority have occurred), the data collected or the assumptions about the juvenile justice system relationships may no longer be comparable across time. Monitoring results may simply reflect the fact that the data are not comparable rather than indicating a fundamental change in the relative handling of minority youth. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-32 Demographic changes in the state or locality can affect DMC trends. This often implies that changes are taking place over time in the population demographics of the community. These trends bear watching to determine what issues might have an impact on the effectiveness of the juvenile justice system. The overall use of the juvenile justice system, budget constraints within the system, and philosophies about appropriate handling of juveniles may change the rates at which activity occurs that leads to contact for juveniles, including minority youth. As was noted in county A above, there is a dramatic shift in the use of secure confinement, partially based on philosophical reasons but also driven by changing budget models. Those changes in rates may make it difficult to compare the RRI values across time. Finally, changes in data collection systems or standards may occur that alter the way in which youth are categorized. For example, earlier in this chapter, the authors examined the impact of asking Hispanic youth to identify themselves with one question (What is your race or ethnicity?) as opposed to two questions (What is your race? What is your ethnic identification?). Villarruel and his associates argued that changing the way in which such data are collected may dramatically 4 alter the statistical portrait of Hispanic youth. If such a change were implemented in any of the juvenile justice information systems within a jurisdiction, the results of the RRI process would not be comparable across the time boundaries of that change, and a monitoring process would provide misleading results. The result of considering such factors may lead a jurisdiction to conclude it is necessary to go beyond comparing the RRI values to explore data patterns across time. In that context, it may be useful to consider options such as:  Repetition of the assessment analyses that originally supported the selection of intervention strategies.  Additional qualitative and quantitative strategies for understanding what is different, or why nothing seems to be different.  Specific evaluation studies designed to establish the relationship between interventions and changes in the DMC levels within the jurisdiction. Important Caveats The consideration of a monitoring strategy, along with the experience of those jurisdictions that have engaged in DMC work at some point, suggests that the following caveats be taken into account in any monitoring process: Change takes time. Considering that the juvenile justice system is a dynamic set of systems, with many cases under consideration at any one time, it is not surprising that changing the nature of the process takes considerable time and DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-33 effort. Expect that some results will show up slowly over time rather than as discrete and immediate changes in the monitoring graphs. Monitoring is most effective when it is conducted with some frequency. OJJDP requires data updates at least every 3 years, but more accurate and useful monitoring is conducted more frequently—often annually or, in some instances, quarterly or monthly. Monitoring feedback needs to be at the level at which DMC occurs. Feedback also needs to be monitored at the targeted DMC reduction sites as well as at the state level. At local levels, monitoring (or at least the display and interpretation of the results) should be housed within an appropriate agency that can present information with legitimacy and credibility. The use of graphic displays such as those presented above may assist in making the material clear to audiences and to policymakers. Endnotes 1. Although OJJDP requires states to report DMC data at least every 3 years, the Office encourages states to invest in targeted DMC reduction efforts at the local level and report DMC data at these local DMC reduction sites regularly on a more frequent basis. 2. A substantial body of research exists that indicates juveniles of Hispanic/Latino origin may experience contact with the juvenile justice system that is substantively different from that of other groups. Because Hispanic/Latino is not a race, this combined term— race and ethnicity—is used to serve as a recommendation for jurisdictions to systematically and purposefully document how data are collected for Hispanic youth. 3. See F.A. Villarruel, N.E. Walker, P. Minifee, O. Rivera-Vazquez, P. Peterson, and K. Perry, ¿Dónde Está la Justicia? A Call to Action on Behalf of Latino and Latina Youth in the U.S. Justice System,‖ Executive Summary, East Lansing, MI: Institute for Children, Youth and Families. Michigan State University, 2002. 4. See Villarruel et al., note 3. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-34 Appendix A: Data Required To Populate the Cells of the DMC Relative Rate Index Matrix Howard N. Snyder The data required for the DMC Relative Rate Index (RRI) matrix depend, in part, on the structure of each jurisdiction’s juvenile justice system and the data resources that the various subsystems maintain. In general, the RRI matrix requires access to a wide range of information. • Population data can be extracted from data files developed and/or maintained by the U.S. Census Bureau, the Centers for Disease Control and Prevention, or a state-specific resource, if available (when the validity of the federal data are questioned). • Arrest data can be extracted from data files developed and/or maintained by state Uniform Crime Reporting (UCR) Programs or law enforcement agencies. • Court processing data capturing case counts at various stages of court processing can often be obtained from the courts themselves. Most juvenile courts in the nation have automated case management or case tracking information systems. • Detention data, depending on the administrative structure of the local juvenile justice system, can be obtained from the juvenile courts, the executive entity that provides detention services, or the detention centers. • Placement data, depending on the administrative structure of the local juvenile justice system, can be obtained from the juvenile courts, the executive entity that provides placement services, or (when no other source is available) a national data collection effort entitled the Census of Juveniles in Residential Placement (CJRP). • Each of these data requirements presents its own unique challenges to the persons who complete the DMC Relative Rate Index Matrix, but some general knowledge about each may be useful to all who are tasked with this responsibility. Population Data Every decade, the U.S. Census Bureau conducts the decennial census, essentially counting each person living in the United States on April 1st of that year and enumerating the age, sex, race, and ethnicity of each person. Between the decennial censuses, the Census Bureau produces population estimates based on the decennial data and other available information resources. In censuses prior to the 2000 census, persons were asked to report if they were of Hispanic origin (or not) and to select from a list of four categories the one race to which they most closely identified, either white, black or African American, American Indian and Alaska Native, or Asian and Pacific Islander. This process resulted in a racial/ethnic coding structure with eight categories (i.e., the four races each with subcategories of “Hispanic” or “non-Hispanic”). DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-35 Once again, for the 2000 census, persons were ethnically self-classified as being of “Hispanic origin” or “not of Hispanic origin.” However, in 2000, the Bureau changed the race question. First, the “Asian and Pacific Islander” category was divided into Asian and Native Hawaiian and Other Pacific Islander, making five race categories. Then, instead of asking for a single race, persons were presented with the five racial categories and asked to “Check all that apply.” This process enabled individuals to classify themselves into one of 31 possible racial categories—5 single-race categories and 26 mixed-race categories. Together, the Hispanic ethnicity and the race question yielded 62 possible race/Hispanic ethnicity categories. The census did not ask the mixed-race respondents to identify the race to which they most closely identified. Therefore, all population data flowing from the U.S. Census Bureau for the year 2000 and after includes mixed-race categories. For some uses, the existence of a mixed-race code causes problems. This occurs when a companion data system codes the race in single-race categories. For example, the FBI’s current racial coding structure in its Uniform Crime Reporting (UCR) Program codes arrestees into one of four races: white, black or African American, American Indian and Alaska Native, and Asian or Pacific Islander. How should analysts calculate race-specific arrest rates if all that were available to them were Census population data (with its 5 single race codes and its 26 mixed race categories) and the UCR arrest counts (with its 4 single race codes)? To calculate a race-specific arrest rate, divide the number of arrests in a specific racial group by the number of persons in the residential population who are of that racial group. To calculate these rates, the analyst could combine the two population counts for Asian and Native Hawaiian/Other Pacific Islander to form a new group that would be more congruent with the UCR’s Asian/Pacific Islander category; but the problem of the mixed-race population count still exists. Within the mixed-race group, there are probably some persons who, if asked to identify the race to which they most closely relate, would code themselves into each of the four single-race groups that the UCR Program uses. However, from the available data, the analyst could not confidently spread the mixed-race counts into the single-race categories. So, the existence of the mixed-race population group makes the number of persons identified in each single-race group an undercount; and as the proportion of mixed-race persons in the population increases (which is occurring in the juvenile populations), so does the error in the value of single-race population counts. Luckily, for this situation, the Centers for Disease Control and Prevention has done the statistical work to spread the mixed-race population counts and produce population estimates for the years following the 2000 decennial census into the pre-2000 four singlerace categories. Analysts who prepare the DMC Relative Rate Index matrix may easily access the data through the data dissemination package entitled Easy Access to Juvenile Populations (www.ojjdp.ncjrs.gov/ojstatbb/ezapop/), which is available in OJJDP’s Statistical Briefing Book. The opening screen of this package is displayed below. The selection requests the population counts for youth ages 10 through 17 for Los Angeles County, California, for the year 2003. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-36 The table generated from this request is displayed below. The counts for the four racial groups are presented and are subdivided by Hispanic/nonHispanic. The Easy Access to Juvenile Populations tool will generate such county- and state-level tables for the years 1990 and onward. It can also generate tables for males and females and for other age groupings. An analyst with the standard UCR arrest data and the population counts from the Easy Access package could then calculate juvenile racespecific arrest rates for each county in the state and the state as a whole. Arrest Data There are about 18,000 law enforcement agencies in the United States. Most counties have many more than one agency that may arrest juveniles—some have more than 100. This presents a problem for analysts when all of the agencies do not use the same information system. If RRI analysts need the count of white juveniles arrested in a particular year, they might be forced to contact several agencies and hope that the definitions of race and arrest (and possibly offense) are all compatible. Luckily, for DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-37 analysts in most places in the country, a state-level entity already has been assigned to report arrest statistics to the FBI. Since the 1930s, the FBI’s UCR Program has asked local law enforcement agencies to report their arrests. Until the 1990s, each reporting agency sent the FBI aggregate counts by gender of arrests within 29 offense categories, subdivided into several age categories: younger than 10, 10–12, 13–14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25–29, 30–34, 35– 39, 40–44, 45–49, 50–54, 55–59, 60–64, and older than 64. So, for example, from the UCR aggregate data, analysts can know the number of arrests involving persons age 16 for burglary in a particular year for each reporting law enforcement agency. These counts for all law enforcement agencies in a county (or a state) could be summed to yield this statistic for a larger geographical area. DMC work and the RRI matrix require that reporting agencies subdivide arrests by race. Independent of the aggregate reporting of arrests by gender and age, the UCR also collects aggregate arrest data within the 29 offense categories broken into the four race categories of white, black or African American, American Indian and Alaska Native, and Asian or Pacific Islander. The UCR does not collect these aggregate race-specific arrest data separately for males and females, and the only age detail that the UCR captures for race-specific arrest data is “juvenile” and “adult” (i.e., persons younger than age 18 and those age 18 or older). Therefore, it is not possible from the aggregate UCR arrest data to obtain counts of the number of burglary arrests of 16-year-old black youth; the available detail limits counts to burglary arrests of black persons younger than age 18 or black persons older than age 17. The lack of age detail in the UCR’s aggregate race-specific arrest data may cause some RRI analysts problems, specifically in states where the original jurisdiction of the juvenile court ends before the 18th birthday. In most states, the UCR’s definition of a juvenile (i.e., younger than age 18) is consistent with the general definition that the state’s juvenile justice system uses. However, in about a dozen states, persons age 17 are routinely processed within the adult criminal justice system, and, in a few states, this is true for youth age 16. Therefore, in these states, the FBI’s age dichotomy in their racespecific arrests of juveniles and adults is inconsistent with other data available about their juvenile justice systems. The UCR’s aggregate arrest reporting does not collect information on the Hispanic ethnicity of arrestees. However, some states collect these aggregate counts, independent of the UCR Program. If an RRI analyst has access to such data, he or she should take care to understand the reporting rules. It is likely that these data are reported at the same age (i.e., juvenile and adult) and gender (i.e., no gender) detail as is race. If so, it would be impossible, using these aggregate counts, to remove the Hispanic counts from each of the four race counts. As a result, each of the four race counts contains arrests of Hispanics to an unknown degree. In summary, somewhere within most states’ aggregate data exist annual counts of arrests of persons younger than age 18 broken down into four race categories for a large number DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-38 of law enforcement agencies. If your state’s definition of juvenile is consistent with the UCR’s definition of juvenile (i.e., persons younger than age 18), then the UCR racespecific arrest data would be a likely source of the arrest information needed for the RRI. In most states, these data are compiled at a single point; typically, a Uniform Crime Reporting Program is based at the state police agency, within the state’s criminal justice planning agency, or within the state’s statistical analysis center. (A list of state UCR reporters can be found in the back of each annual Crime in the United States report. The list from Crime in the United States 2004 is presented in appendix B of this chapter.) For states in which the definition of a juvenile is inconsistent with the UCR’s youngerthan-age-18 reporting category, or where analysts would like to have their arrest data broken down by both race and sex (and possibly even Hispanic ethnicity), a potential option may exist. In the late 1980s, the FBI expanded the UCR data collection effort from aggregate reporting to incident-based reporting. That is, instead of a department reporting that it made 10 arrests for burglary of persons ages 25 to 29 (an aggregate count), the new incident-based reporting requirements asked the agency to report for each burglary arrestee the person’s age, sex, race, and Hispanic ethnicity. Since the early 1990s, the FBI has collected these incident-based reporting records under the UCR’s National IncidentBased Reporting System (NIBRS). When an agency moves from aggregate to incidentbased reporting, the information potential of the arrest data increases substantially. From incident-based reporting departments, an RRI analyst can obtain detailed counts of arrests at just about any level of age/sex/race/Hispanic ethnicity desired. For example, from the incident-based reporting data, an analyst can derive the number of burglary arrests of white, non-Hispanic males younger than age 16. The number of law enforcement agencies collecting incident-based data and the number reporting to NIBRS is constantly increasing. Analysts should investigate the availability of NIBRS data in their jurisdiction. Court Processing Data The majority of the data needed to complete the RRI matrix could come from juvenile court management information systems. Such systems are common across the nation. The RRI analyst should seek out those who administer their local juvenile court management information systems and request the needed statistics. The information may already be a standard part of the court’s reporting effort or could be easily produced. If not, court programmers could prepare a new extract program to summarize the existing data. Most courts will provide these data if the importance of the request is clear to them and if the request gives unambiguous and detailed definitions of the statistics desired. If the statistics do not already exist, there may be some expense involved; but having those who know the data do the work is always far less expensive and less time consuming than having someone unfamiliar with the data set do it—assuming they can even obtain access to it. Most juvenile courts that collect automated information contribute their data to the National Juvenile Court Data Archive, a resource housed at the National Center for Juvenile Justice, the research division of the National Council of Juvenile and Family Court Judges, and supported by grants from OJJDP. Currently, courts with juvenile DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-39 delinquency jurisdictions that serve about 1,800 of the 3,000 counties in the United States contribute their data to the archive. So, court data exist in most counties. Most of these systems collect information on the demographics of the youth referred (including race and ethnicity); date of referral; offense(s) referred; the processing decisions of diversion, petitioning, transfer/waiver, and adjudication; and the disposition of the case, including probation or out-of-home placement. If all else fails and the data are housed in the National Juvenile Court Data Archive, the RRI analyst could request access to the archived data from the jurisdictions of interest. This process begins with a detailed letter to the archive requesting access to specific data files and detailing the types of analyses that will be performed on these data. The archive will then forward the request to the original data supplier(s) seeking permission to release the file(s).Generally, permission to release the data come with conditions to which the data requestor must agree contractually before the data are released. Also the archive will charge a small fee to oversee this process and prepare the data set(s), along with the SPSS (Statistical Packages for the Social Sciences) program(s) to read them. Detention Data Many juvenile court information systems capture information on the court’s use of detention within each case processed. If so, then the detention information needed for the RRI Matrix could be found in the court data (see above). However, in more and more jurisdictions, detention information is collected in an information system separate from the court system. This occurs often when detention centers are not administratively within the judicial branch of government. When the source of detention data is not the court, analysts should take care to understand the nature of the detention data, especially their unit of count. When detention information is within a court information system, the use of detention is likely to be tied to the court case. When this occurs, the court data can answer such questions as: How many delinquency cases involving white youth also involved the use of detention prior to adjudication? In this situation, the unit of count for detention is the court case. That is, a youth may have been detained more than once in the case, but the unit of count indicates whether detention ever occurred—yes or no. When detention information is extracted from a stand-alone detention information system, the detention information is often not tied to a specific case. In a year, a single youth may have had several detention admissions; if these were tied to one case is unknown. In such situations, the unit of count for detention would be the number of admissions, not the number of court cases with detentions. For the RRI matrix, it does not matter which of the possible units of count is used, just that the unit of count is clear and that the analysts understand how different units of counts may result in different RRI indexes. For example, a youth is arrested, detained, adjudicated, and ordered to weekend detention for a period of 3 months. When the unit of count is “Detention within case—Yes or No,” this scenario would yield “one case detained.” If the unit of count were detention admissions, the scenario would yield more DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-40 than a dozen detention admissions. If some case types were more likely to experience multiple detention admissions, their influence on the RRI would vary with unit of count. Census of Juveniles in Residential Placement. If no local detention information exists, there is a source of state-level detention (and placement) information that could be used to fill the RRI matrix. OJJDP implemented the Census of Juveniles in Residential Placement (CJRP) in 1997. The 1997 CJRP asked juvenile residential custody facilities in the United States to complete an individual record that described each youth assigned a bed in the facility on the last Wednesday in October. CJRP data were collected again in 1999, 2001, and 2003. It is important to understand what CJRP collects and what it does not. The CJRP facility inclusion criteria are as follows: residential facilities in operation on the census reference date, residential facilities that are either publicly or privately (or tribally since 1999) operated, and residential facilities intended for juvenile offenders (although some hold adults as well). Specifically excluded are nonresidential facilities, detention centers operated as part of adult jails, facilities exclusively for drug abusers or dependent/ neglected youth, foster homes, and federal correctional facilities (e.g., Immigration and Naturalization Service, Bureau of Indian Affairs, U.S. Marshals, or Bureau of Prisons). CJRP is not sent to adult facilities or to facilities exclusively for drug or mental health treatment or for abused or neglected children. Inclusion criteria for individual-level data are as follows: youth younger than age 21, assigned a bed in a residential facility at the end of the day on the census reference day, charged with an offense or court-adjudicated for an offense, and in residential placement because of that offense. CJRP collects an individual record on each juvenile held in the residential facility, with information on the juvenile’s gender, date of birth, race, placement authority, most serious offense charged, court adjudication status, date of admission, and security status. Once again, these data are requested for all offenders younger than 21 years of age in the facility. It should be emphasized that CJRP provides 1-day population counts of juveniles in residential placement facilities. One-day counts give a picture of the standing population in facilities. One-day counts are substantially different from annual admission and release counts, which give a measure of facility population flow. One-day count statistics overrepresent youth with longer lengths of stay (more serious offenders, those in longterm placements) and underrepresent youth with short lengths of stay (those in detention). The CJRP data do not capture information on the county of offense or the county for which the youth is being held. CJRP does collect for each youth the state in which the offense occurred and the state in which the facility is located. Therefore, CJRP data can yield only state-level counts. With these data, state-level analyses can display the number of youth that the courts have placed in a single state regardless of whether the youth was placed in a facility in the state or elsewhere in the nation. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-41 The National Center for Juvenile Justice developed and maintains the Census of Juveniles in Residential Placement Databook for OJJDP to make CJRP data available to a wide variety of users. The CJRP Databook is available at www.ojjdp.ncjrs.gov/ojstatbb/cjrp/. It contains a large set of predefined state-level tables detailing the characteristics of juvenile offenders in custody (age, sex, race/ethnicity, offense, type of facility, and placement status). Users can view custody population profiles for a single state but not for a particular county within the state. Downloaded tables can be saved and imported into spreadsheet software for further analysis. This application is periodically modified or expanded. (Although the CJRP data files are not generally publicly available due to confidentiality concerns, they may be made available to analysts on a case-by-case basis. Researchers should contact OJJDP for information regarding access requirements and procedures.) For RRI analysts, the CJRP tables will provide a 1-day count of the number of youth detained in their state in the target year using the population restrictions detailed above. CJRP captures the race/ethnicity of these youth in the following coding structure: white, not of Hispanic origin; black, not of Hispanic origin; Hispanic; American Indian or Alaskan Native, not of Hispanic origin; Asian or Pacific Islander, not of Hispanic origin; and Other. (The “Other” code is rarely used and is likely to indicate a mixed-race youth.). A typical table from the CJRP Databook appears below. This table shows the number of California youth by sex and race/ethnicity in detention status on the census date in 2003. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-42 Placement Data As with detention data, many juvenile court information systems capture information on the court’s use of out-of-home placement within each case processed. If so, then the placement information needed for the RRI matrix could be found in the court data. When placement information is with a court information system, the use of out-of-home placement is likely to be tied to the court case. When this occurs, the court data can answer such questions as: How many delinquency cases involving white youth were placed out-of-the-home at case disposition? In this situation, the unit of count for out-ofhome placement is the court case. That is, the court may have placed the youth more than once in the case or in more than one facility, but the unit of count indicates whether outof-home placement ever occurred in the case. In many jurisdictions, however, out-of-home placement information is collected in an information system separate from the court system. When the source of placement data is not the court, analysts should take care to understand the nature of the placement data, especially their unit of count. When out-of-home placement information is extracted from a stand-alone correctional information system, the placement information often is not tied to a specific case. In a year, a single youth may have several facility admissions; it is often unknown if these were tied to one single case. In such situations, the unit of count for placements would be the number of admissions, not the number of court cases in which the youth was placed out of the home. For placement information in the RRI matrix, it does not matter which of the possible units of count an analyst uses, just that the unit of count selected is clear and that the analyst understands how different units of counts may result in different RRI indices. For example, a youth is arrested, detained, adjudicated, and ordered to the custody of the state department of juvenile corrections. When the unit of count comes from a court data system and is “Out-of-home placement within the case—Yes or No,” this scenario would yield one case placed out of the home. If the unit of count was commitment to the state department of juvenile corrections, the scenario would yield one commitment. However, if the correctional information system could only monitor flow into a facility and a youth passes through several facilities during the commitment experience (e.g., a diagnostic and evaluation center, a state training school, a halfway house, recommitment to the training school following a parole violation, and finally another halfway house), the unit of count would yield five placements. If some case types were more likely to experience multiple placements, then their influence on the RRI would vary with unit of count. As with detention, if no locally available placement information exists, the CJRP data could serve as a source of state-level placement information to fill the RRI matrix. CJRP also has its unique counting rules and characteristics that any analyst using the CJRP data should thoroughly understand. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-43 Conclusion Analysts who complete the DMC Relative Rate Index matrix should realize that much of the needed data already exist, although they are not always easy to find or easy to access. RRI analysts should be students of the sources and types of relevant information available within their states and counties. Analysts should read the statistical reports of law enforcement, juvenile courts, and other entities that handle youth within the juvenile justice system. By doing so, they will develop an understanding of what data are available, what statistics are reported routinely, and who could be their colleagues in the task to generate and interpret the DMC Relative Rate Index matrix. DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-44 Appendix B: State UCR Reporting Agencies, 2004 Listing Alabama Alabama Criminal Justice Information Center Suite 350 770 Washington Avenue Montgomery, Alabama 36104 (334) 242-4900 www.acjic.state.al.us Alaska Alaska Department of Public Safety Criminal Records and Identification Bureau 5700 East Tudor Road Anchorage, Alaska 99507 (907) 269-5765 American Samoa Department of Public Safety Post Office Box 1086 Pago Pago American Samoa 96799 (684) 633-1111 Arizona Access Integrity Unit Uniform Crime Reporting Program Arizona Department of Public Safety Mail Drop 1190 Post Office Box 6638 Phoenix, Arizona 85005-6638 (602) 223-2239 www.dps.state.az.us Arkansas Arkansas Crime Information Center One Capitol Mall, 4D-200 Little Rock, Arkansas 72201 (501) 682-2222 www.acic.org DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-45 California Criminal Justice Statistics Center Department of Justice Post Office Box 903427 Sacramento, California 94203-4270 (916) 227-3515 Colorado Uniform Crime Reporting Colorado Bureau of Investigation Suite 3000 690 Kipling Street Denver, Colorado 80215 (303) 239-4222 www.cbi.state.co.us Connecticut Uniform Crime Reporting Program 1111 Country Club Road Middletown, Connecticut 06457-9294 (860) 685-8030 www.state.ct.us/dps/crime_analysis/crime_analysis.asp Delaware Delaware State Bureau of Identification Post Office Box 430 Dover, Delaware 19903-0430 (302) 739-5901 District of Columbia Research and Resource Development Metropolitan Police Department 300 Indiana Avenue, N.W. Washington, D.C. 20001 (202) 727-4174 www.mpdc.dc.gov Florida Criminal Justice Information Services Uniform Crime Reports Florida Department of Law Enforcement Post Office Box 1489 Tallahassee, Florida 32302-1489 (850) 410-7121 DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-46 Georgia Georgia Crime Information Center Georgia Bureau of Investigation Post Office Box 370748 Decatur, Georgia 30037-0748 (404) 270-8467 www.ganet.org/gbi/ Guam Guam Police Department Planning, Research and Development Building #233 Central Avenue Tiyan, Guam 96913 (671) 475-8422 Hawaii Crime Prevention and Justice Assistance Division Department of the Attorney General Suite 401 235 South Beretania Street Honolulu, Hawaii 96813 (808) 586-1150 www.hawaii.gov/ag/cpja Idaho Bureau of Criminal Identification Idaho State Police Post Office Box 700 Meridian, Idaho 83680-0700 (208) 884-7156 www.isp.state.id.us/identification/ucr/ Illinois Uniform Crime Reporting Program Illinois State Police 2nd Floor 500 Iles Park Place Springfield, Illinois 62703 (217) 782-5794 www.isp.state.il.us DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-47 Iowa Iowa Department of Public Safety Wallace State Office Building East Ninth and Grand Des Moines, Iowa 50319 (515) 281-8494 www.dps.state.ia.us/ Kansas Kansas Bureau of Investigation Information Services Division Incident Based Reporting Section 1620 Southwest Tyler Street Topeka, Kansas 66612 (785) 296-8279 www.accesskansas.org/kbi/ Kentucky Criminal Identification and Records Branch Kentucky State Police 1250 Louisville Road Frankfort, Kentucky 40601 (502) 227-8790 www.kentuckystatepolice.org Louisiana Louisiana Commission on Law Enforcement Uniform Crime Reporting 12th Floor 1885 Wooddale Boulevard Baton Rouge, Louisiana 70806 (225) 925-4440 www.cole.state.la.us/lucr.htm Maine Records Management Services Uniform Crime Reporting Division Maine Department of Public Safety Maine State Police Suite 1 45 Commerce Drive Augusta, Maine 04333-0042 (207) 624-7276 www.maine.gov/dps/ DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-48 Maryland Central Records Division Incident Reporting Section Maryland State Police 1711 Belmont Avenue Baltimore, Maryland 21244 (410) 298-3883 Massachusetts Crime Reporting Unit Uniform Crime Reports Massachusetts State Police 470 Worcester Road Framingham, Massachusetts 01702 (508) 820-2111 Michigan Uniform Crime Reporting Unit Criminal Justice Information Center Michigan State Police 7150 Harris Drive Lansing, Michigan 48913 (517) 322-1424 www.michigan.gov/msp Minnesota Criminal Justice Information Systems Bureau of Criminal Apprehension Minnesota Department of Public Safety 1430 Maryland Avenue East St. Paul, Minnesota 55106 (651) 793-2400 www.bca.state.mn.us/ Missouri Missouri State Highway Patrol Criminal Records & Identification Division CJIS Section—UCR Program Office 1510 East Elm Street Post Office Box 9500 Jefferson City, Missouri 65102-9500 (573) 526-6278 www.mshp.dps.missouri.gov/MSHPWeb/PatrolDivisions/CRID/index.html DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-49 Montana Montana Board of Crime Control Post Office Box 201408 Helena, Montana 59620-1408 (406) 444-4298 www.mbcc.state.mt.us Nebraska Uniform Crime Reporting Section The Nebraska Commission on Law Enforcement and Criminal Justice Post Office Box 94946 Lincoln, Nebraska 68509-4946 (402) 471-3982 www.nol.org/home/crimecom/ Nevada Uniform Crime Reporting Program Records and Identification Bureau 808 West Nye Lane Carson City, Nevada 89703 (775) 687-1600 x235 www.nvrepository.state.nv.us New Hampshire Uniform Crime Reporting Unit New Hampshire State Police New Hampshire Department of Public Safety 33 Hazen Drive Concord, New Hampshire 03305 (603) 271-2509 New Jersey Uniform Crime Reporting Unit New Jersey State Police Post Office Box 7068 West Trenton, New Jersey 08628-0068 (609) 882-2000 x2392 www.njsp.org DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-50 New York Statistical Services New York State Division of Criminal Justice Services 8 th Floor, Mail Room 4 Tower Place Albany, New York 12203 (518) 457-8381 http://criminaljustice.state.ny.us North Carolina Crime Reporting and Criminal Statistics State Bureau of Investigation Post Office Box 29500 Raleigh, North Carolina 27626-0500 (919) 662-4509 http://sbi2.jus.state.nc.us/crp/public/Default.htm North Dakota Information Services Section Bureau of Criminal Investigation Attorney General's Office Post Office Box 1054 Bismarck, North Dakota 58502 (701) 328-5500 www.ag.state.nd.us Ohio Office of Criminal Justice Services 14th Floor 140 East Town Street Columbus, Ohio 43215 (614) 466-7782 Oklahoma Uniform Crime Reporting Section Oklahoma State Bureau of Investigation 6600 North Harvey Oklahoma City, Oklahoma 73116 (405) 879-2533 www.osbi.state.ok.us DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-51 Oregon Law Enforcement Data System Division Oregon State Police Post Office Box 14360 Salem, Oregon 97309 (503) 378-3055 x55002 Pennsylvania Bureau of Research and Development Pennsylvania State Police 1800 Elmerton Avenue Harrisburg, Pennsylvania 17110 (717) 783-5536 http://ucr.psp.state.pa.us Puerto Rico Statistics Division Puerto Rico Police Post Office Box 70166 San Juan, Puerto Rico 00936-8166 (787) 793-1234 x3113 www.policia.gobierno.pr Rhode Island Rhode Island State Police 311 Danielson Pike North Scituate, Rhode Island 02857 (401) 444-1156 www.risp.ri.gov/ South Carolina South Carolina Law Enforcement Division Post Office Box 21398 Columbia, South Carolina 29221-1398 (803) 896-7016 www.sled.state.sc.us South Dakota South Dakota Statistical Analysis Center 3444 East Highway 34 Pierre, South Dakota 57501-5070 (605) 773-6312 www.dci.sd.gov DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-52 Tennessee Tennessee Bureau of Investigation 901 R.S. Gass Boulevard Nashville, Tennessee 37216 (615) 744-4000 www.tbi.state.tn.us Texas Uniform Crime Reporting Crime Information Bureau Texas Department of Public Safety Post Office Box 4143 Austin, Texas 78765-9968 (512) 424-2091 www.txdps.state.tx.us/crimereports/citindex.htm Utah Data Collection and Analysis Uniform Crime Reporting Bureau of Criminal Identification Utah Department of Public Safety Post Office Box 148280 Salt Lake City, Utah 84114-8280 (801) 965-4812 www.bci.utah.gov Vermont Vermont Crime Information Center 103 South Main Street Waterbury, Vermont 05671 (802) 244-8727 www.dps.state.vt.us/cjs/crimestats.htm Virginia Criminal Justice Information Services Division Virginia State Police Post Office Box 27472 Richmond, Virginia 23261-7472 (804) 674-2143 www.vsp.state.va.us/crimestatistics.htm Virgin Islands Virgin Islands Police Department Alexander Farrelly Justice Complex Saint Thomas, Virgin Islands 00802 (340) 774-2211 DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-53 Washington Uniform Crime Reporting Program Washington Association of Sheriffs and Police Chiefs Suite 200 3060 Willamette Drive, Northeast Lacey, Washington 98516 (360) 486-2380 www.waspc.org West Virginia Uniform Crime Reporting Program West Virginia State Police 725 Jefferson Road South Charleston, West Virginia 25309 (304) 746-2237 www.wvstatepolice.com Wisconsin Wisconsin Office of Justice Assistance Suite 610 131 West Wilson Street Madison, Wisconsin 53702-0001 (608) 266-3323 http://oja.state.wi.us/ Wyoming Uniform Crime Reporting Criminal Records Section Division of Criminal Investigation 316 West 22 nd Street Cheyenne, Wyoming 82002 (307) 777-7625 http://attorneygeneral.state.wy.us/dci/ DMC Technical Assistance Manual, 4th Edition • Chapter 1: Identification and Monitoring 1-54 Chapter 2: Assessment Michael Leiber, Dorinda Richetelli, and William Feyerherm* Although the identification stage of the DMC process provides jurisdictions with a description or an account of the extent of minority overrepresentation, the assessment stage is an indepth examination of how DMC occurs. An assessment is a search for the factors that contribute to DMC, with the goal that the results may lead to strategies or interventions to reduce DMC. This chapter discusses mechanisms that may result in DMC and explains how to conduct the assessment. It is important to note that the nature of the assessment process necessarily depends on the preceding identification stage. The logic of the assessment phase builds on the results of the identification process. If a community has sufficient identification information for all or most of the major stages in the juvenile justice system, then it can use the findings to further refine and focus the needed inquiry of the assessment. On the other hand, if a community does not have adequate information to complete the identification process, then the techniques suggested here for conducting the assessment will be less successful. To improve on the value that the assessment study can provide for addressing DMC, it is imperative that the community complete, or come as close as possible to completing, the identification process. An Overview of Assessment The assessment process looks more carefully at the decision points that the identification process has targeted to determine how DMC is created or amplified, specifying the mechanisms at work in a particular jurisdiction. The outcome of the assessment study should result in an understanding of the DMC process that will permit policymakers to make choices about strategies for reducing DMC. To accomplish this purpose, the authors suggest a multistage investigative process: Stage 1: Generate possible explanations. At this stage, the starting point is to choose specific stages, groups, and jurisdictions to explore. This is the likely outcome from the identification stage. Using community leaders, agency personnel, and key informant processes, analysts should generate a set of plausible/possible explanations for the level of DMC observed in the jurisdiction (by stage and racial/ethnic group) for the targeted stages, groups, and jurisdictions. * About the authors: Michael Leiber, Ph.D., is a professor in the Wilder School of Government and Public Affairs at Virginia Commonwealth University in Richmond, Virginia. Dorinda Richetelli is Vice President at Spectrum Associates in Avon, Connecticut. William Feyerherm, Ph.D., is Vice Provost of Research at Portland State University in Portland, Oregon. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-1 Stage 2: Identify the types of data and the pattern of results needed. These should be consistent with the possible explanations and will distinguish between the possible explanations. Stage 3: Obtain the data. Identify sources of the needed data, focusing on those that are most readily available and suitable for comparison over multiple time periods. If data sources are available, make sure that you know how the data are collected and what each data item actually means. If the needed data are not available, then develop plans to collect them. You could use existing files, collect additional data, or develop a hybrid model in which you collect additional/supplemental data on a periodic basis. Stage 4: Analyze the data and identify the most likely mechanism(s) creating DMC in this jurisdiction. Conduct the analyses according to the patterns you expect to emerge (stage 2). Examine the data analysis to see whether the patterns you have observed are consistent with possible explanations. If the data results are consistent with more than one explanation, you may need to plan additional analyses that may help distinguish between options. Develop feedback methods for taking the data results back to the community and key informants to verify the interpretations and begin the process of selecting interventions. Stage 1: Generating Possible Explanations: Mechanisms Leading to DMC To assess and address DMC issues, jurisdictions must explore and identify the mechanisms by which DMC is created. This section will explore the major mechanisms that the research literature has identified, briefly explain the means by which each mechanism operates, and provide some simple examples of the mechanism. The authors will then return to the full list to describe the pattern of Relative Rate Index (RRI) values that you might expect to find if this mechanism is at work and, finally, provide some ideas of the types of assessment work and analysis that might lead to greater confidence that a particular mechanism was actually a significant contributor to DMC in a specific jurisdiction. The material that follows is a partial list of mechanisms found in the research literature. The list is modified from the OJJDP-funded Justice Research and Statistics Association (JRSA) publication Seven Steps To Develop and Evaluate Strategies To Reduce Disproportionate Minority Contact (DMC) by Ashley Nellis, available at www.jrsa.org/jjec/about/dmc_guidebook.html. You may wish to use the list as a stimulus to practitioners, policymakers, and community members as they think about how their juvenile justice system operates and examine possible methods by which DMC is introduced into their system. These activities should generate a listing of mechanisms for this specific community to explore. Differential Behavior The research literature raises the possibility that the rates at which youth from various racial and ethnic subgroups are involved in delinquent activity may differ (e.g., Lauritsen, DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-2 2005). Differing rates of involvement is not a universal phenomenon, nor is it presented here to suggest that disproportionate contact is acceptable. As the Juvenile Justice and Delinquency Prevention Act specifies, one of the means of addressing DMC is through prevention activities, which may not only address DMC but also provide substantial benefits to children and youth generally. Several forms of differential behavior are plausible contributors to DMC, including: Involvement in a different set of offense categories (often including more serious activities such as possession or sale of controlled substances), involvement in gang-related activity, and more frequent involvement in offenses generally and in offenses with higher levels of severity. Involvement in delinquent activities at an earlier age. Involvement with other social services or justice-related systems, such as the child welfare system (dependency or neglect cases). It is noteworthy that many other social services systems are also establishing initiatives or standards related to cultural competency and issues similar to DMC, thus providing the opportunity for cross-system collaboration in addressing issues of racial or cultural disparities. Mobility Effects: Importation/Displacement One of the realities of modern life is easy access to automobiles and other means of mobility, so that youth who reside in one community may, in fact, spend considerable time in other jurisdictions. While present in those other jurisdictions, it is possible that youth may commit delinquent behavior, resulting in their being arrested and, perhaps, processed further in a jurisdiction other than their own home area. When arrest statistics are compared to census statistics on juvenile population, which are based on the area of residence, the result may be that the rate of juvenile arrests in one area may appear either higher or lower than would be expected. Several forms of such mobility-related DMC have been observed. Seasonal Mobility Seasonal mobility occurs when a community has an influx of juveniles during a particular season, frequently either a holiday season (spring break) or a vacation season (summer break). A community may be a destination for many families or youth; depending on the patterns of movement, this may result in higher numbers of youth of color in a community than were recorded in census estimates. For example, many resort communities draw youth from larger urban areas during school holidays. That influx will temporarily change the demographic composition of the juvenile population. As an extreme example, one midwestern county discovered that the arrests of African American youth exceeded the total number of youth estimated in the census as county residents. Further exploration suggests that this county serves as a summer retreat destination for many families, which has the impact of substantially increasing the number of African American youth in the community during the summer. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-3 Attractive Nuisance Attractive nuisance is a term that might be applied to a number of commercial or entertainment areas, particularly in urban settings. For example, a shopping mall or entertainment facility may be located in a suburban community or an urban neighborhood that has lower proportions of minority residents but draws youth from across an urban area. It is likely that the demographic profile of youth in such a location reflects a higher proportion of minority youth than does the census estimate for the area immediately surrounding the facility. Immigration- and Migration-Related Mobility Immigration- and migration-related mobility may have an impact on communities to create higher levels of DMC, particularly where policies of the U.S. Citizenship and Immigration Services (USCIS, formerly Immigration and Naturalization Service) are a major concern. To the extent that jurisdictions detain Hispanic (or other) youth suspected of illegal immigration, DMC numbers are likely to be affected. Moreover, as networks of illegal behavior become more organized and youth from other countries join as participants, DMC numbers may be exacerbated. For example, in one community that monitors RRI numbers, the juvenile agency noticed increasing RRI values for Hispanic youth at the detention stage. The agency generated a list of possible factors, including concerns such as availability of interpretive services, availability of alternative programs, staffing changes, etc. When the agency presented this list to its advisory council, one judicial officer noted that she had seen several cases involving youth from another country who were explicitly brought to the United States as runners for drug trafficking. Upon further exploration, it became clear that there were multiple examples of this phenomenon and that when such youth were excluded from the activity counts, the RRI values were reduced for Hispanic youth. Such a finding provides an opportunity for collaboration between the juvenile justice system and both prevention workers and other agencies concerned with such activity. Institutional Effects Institutional effects may occur when a jurisdiction provides residential or detention capacity for a number of other jurisdictions. For example, if a county operates a regional detention facility, then it might appear that its volume of detention activity is higher than in surrounding counties, and if the county includes these nonresident youth in its RRI calculation, it might create erroneous results. This artificial effect is usually eliminated by calculating detention activity or residential placements based on the county making the commitment placement, not the county physically holding the youth. Indirect Effects ―Indirect effects‖ is a broad term that reflects the fact that in this society, economic status, education, location, and a host of risk factors associated with delinquent behavior, among other factors, are linked with race and ethnicity. These factors, in turn, are related to delinquent activity or to other forms of contact within the justice system. Thus, the DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-4 impact of race or ethnicity is not direct but is ―indirect‖ through these third factors. Those effects in terms of DMC issues are at least threefold: Specific Risk Factors Specific risk factors, which are correlated with race or ethnicity, may lead to differential offending issues. Risk factors such as poor school performance or living in disorganized neighborhoods are more likely to occur to minority youth, putting them at a greater risk of system involvement. As an example, Sampson (1987) discovered that male unemployment is related to family disruption, a risk factor related to delinquency rates, thus creating a set of links with particular impact on African American youth. Programming Access/Eligibility Access to or eligibility for programming (public or private) may be affected as well. For example, access to some forms of behavioral health or substance use treatment is often contingent on medical insurance coverage. That coverage is, in turn, often contingent on economic circumstances, which places many minority families at a disadvantage in obtaining such services. The use of alternative private schools as a preventive measure is also highly related to economic circumstances, again creating a link to race and ethnicity. Juvenile justice decisionmakers report that, in some situations, the only way they can obtain needed treatment services for minority youth is to commit them to state custody, thus adding to the DMC levels for that community. Decisionmaking Factors Decisionmaking factors used within the juvenile justice system may be linked to race and ethnicity. For example, a number of studies have indicated that juvenile justice decisionmakers respond differently to youth from an ―intact‖ two-parent family setting than to youth from a single-parent home. A greater proportion of minority youth in those justice systems lived in single-parent households or other family structures that created a difference in handling within the justice system (Bishop and Frazier, 1996). Thus, what appears to be a decision based on relevant factors made in ―good faith‖ may still contribute to DMC. An alternative may be to expand the search to look for an adult willing to take responsibility for the youth, thus reaching past the two-parent home to examine the capacity of other family structures. Differential Opportunities for Prevention and Treatment The allocation of prevention and treatment resources within communities is seldom uniform or universally accessible across the entire community. In some instances, those allocations create a disadvantage for minority youth. This can occur in at least four ways: Access Access may be limited by geography, hours of operation, or other means. For example, if a program is located in an area of a community that is not accessible through public transportation, the unintended outcome may be that only families who have access to DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-5 private automobiles may participate. If a program is structured so that it is available only during normal working hours, then youth whose parents cannot leave their place of employment during work hours may be unable to participate. If a program is not located in those sections of a community with high concentrations of minority youth, then minority youth are less likely to access it. Eligibility Eligibility criteria may be used in many programs to define a set of youth most likely to benefit from the program or to exclude those youth that program leaders believe will likely disrupt the program or otherwise be less likely to benefit from the program resources. Some of those eligibility criteria may work to the disadvantage of minority youth. For example, drug court or mental health programs may have entry criteria that exclude youth with some types of prior delinquency or other histories. These criteria may be more likely to exclude minority youth. When such criteria are evidence based, they may suggest other intervention strategies to address DMC (e.g., to work on the factors that lead to these eligibility criteria differences). Implementation Implementation characteristics may play a role in encouraging or discouraging minority youth participation. The physical tone of a facility may be inviting or discouraging, may indicate an appreciation of multiple cultures, or may be sterile and institutional. Staff attitudes and demeanor may be welcoming or the opposite. For those youth who do not speak English, the lack of materials and interpretive services in their own language may create barriers to participation. These and other attributes may affect a program’s capacity to retain minority youth participation over time, which is important to achieving the intended prevention or intervention outcomes. As an example, examination of an intervention program to improve the social skills and employment opportunities of troubled African American delinquent males ―one step away from the state training school‖ revealed that these youth were not likely to complete the 4-month program because the lead staff members were neither African American nor male. In this instance, the characteristics of staff seemed to be critical to success with minority clients. This does not, of course, mean that such an impact will occur for all programs or all youth, simply that implementation characteristics need to be considered when differential success is present. Effectiveness Effectiveness is the capability to achieve intended outcomes. Many prevention or treatment programs have been developed initially with a particular group of youth in mind, often white youth. Whether the prevention/treatment model is sufficiently culturally adapted or neutral is a question that is frequently noted in the compilation of evidence-based programs, such as the OJJDP Model Program Guide. The issue for examination in DMC is whether the program outcomes (e.g., prevention) are accomplished at equal rates for youth of differing racial and cultural backgrounds. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-6 Differential Processing or Inappropriate Decisionmaking Criteria Differential processing or inappropriate decisionmaking criteria can be an issue in determining program eligibility, implementing diversion programs, and selecting alternative decision outcomes. The fundamental questions are: What are the bases or criteria on which decisions are made? Are those criteria applied consistently across all groups of youth? Are the criteria structured in a manner that places some groups at a disadvantage? As an example of such issues, consider the use of the term ―gang-related,‖ which is frequently cited as a factor in decisions about how to handle juveniles. To assess its impact, it is important to know how a jurisdiction defines the term, how it is created, and whether the question about being gang related is only asked for youth from certain areas of the community. If so, then use of this criterion likely will place minority youth at some disadvantage relative to white youth, especially white youth from areas of the community not believed to be gang affected. As another example, consider the use of ―family‖ in some detention decisions. It is common to find that one of the criteria for releasing a youth from custody is that a family member must be willing to retrieve the youth. But if the definition of family member extends only to a parent, then the youth from a singleparent home is at a disadvantage. Moreover, the youth who is living with a brother or sister, an aunt or uncle, a grandparent, or other adult is at a disadvantage in such a situation. In many jurisdictions, minority youth are more likely to live in these alternative living arrangements; therefore, the way in which the decision criteria are structured may place such youth at a disadvantage in terms of consideration for being released from detention (or not held in detention at all). A last example centers on the requirement by states that before a youth may participate in diversion at intake, he or she must admit guilt. Although the criterion itself may be racially neutral, studies have raised questions concerning the extent to which minority youth, because of past discriminatory practices and/or distrust of the juvenile justice system, are more likely not to admit guilt and, therefore, are less likely to be involved in diversion than white youth (e.g., Leiber, 1994). Justice by Geography Justice by geography concerns the concept that youth in general, and minority youth in particular, may be processed or handled differently in one jurisdiction than in another within the same state. Differing responses may occur based on whether the youth was processed in an urban versus a rural setting or an urban versus a suburban setting, differences in resources (availability of diversion services), or differences in operating philosophies between jurisdictions (for instance, how a jurisdiction defines ―accountability‖ for youthful misconduct or whether a jurisdiction uses deterrence as a primary rationale for system action as opposed to other philosophies of public safety) (e.g., Bridges and Steen, 1998; Feld, 1991). For example, in Iowa, a study discovered that in one jurisdiction, the juvenile court adhered to an ideology of juvenile accountability and racial stereotyping of African American youth as being more delinquent and in need of intervention. This resulted in blacks being subjected to different case processing and DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-7 case outcomes than similarly situated whites. In another jurisdiction, the juvenile court espoused a strong emphasis on parens patriae at a time when multiple minority groups were moving into the area and local perceptions held that these groups did not adhere to middle-class standards of dress, demeanor, marriage, and respect for authority. As a consequence, the court responded to minority youth differently than white youth (Leiber, 2003). Another example of justice by geography can occur when minority youth in a large jurisdiction (e.g., a state) are concentrated in areas or jurisdictions (communities) where rates of processing differ from those prevalent in other portions of the larger jurisdiction. The end result is that minority youth are more likely to live in jurisdictions where higher rates of contact with the system occur; therefore, in the aggregate state-level calculations, minority youth are more likely to have high rates of system contact compared with white youth who live in other jurisdictions. A similar explanation can lead to lower levels of DMC when minority youth live in jurisdictions in which lower levels of system processing occur. The essential characteristics of justice by geography involving the last example are twofold: Jurisdictions have a wide variation in the rates of juvenile justice system activity. The geographic distribution of minority youth populations correlates strongly with the variation in rates of juvenile justice system activity. A further explanation of these characteristics emerges from the following example. In one midwestern state, researchers were discussing the results of the identification process with probation supervisors. One astute supervisor noted that the RRI values at the state level were higher than the values for any of the counties in the state. The explanation was that the jurisdictions in which minority youth live in that state were also the jurisdictions that had higher rates of juvenile justice activity (e.g., arrest, detention, prosecution, etc.). As a result, the minority youth in that state not only experienced a higher level of contact than their counterparts within their own community, but, compared with white youth in other sections of the state, their rates of juvenile justice system contact/activity were much higher. The identification of justice by geography as a mechanism leading to DMC is particularly difficult in a system of government that embraces local variation and adaptation. The recognition that these variations may have unintended consequences may lead to discussions within and across jurisdictions about the basis for local variations in practice. This does not mean that any particular local practice is ―wrong,‖ simply that policymakers need to be aware of the consequences of the differences in policy and practice across communities. Legislation, Policies, and Legal Factors With Disproportionate Impact Policies enacted through legislation or through administrative action may sometimes contain elements that create a disadvantage for minority youth. These disadvantages may DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-8 occur for a variety of reasons, but the most common are those that target some specific aspect of delinquent behavior, those that target specific locations, and those that use prior delinquent or criminal history as an element of the policy. As examples, consider the following: Policies that target certain types of offenses or offense characteristics may have a disproportionate impact on minority youth. For example, statutes that define drug offenses tend to treat crack cocaine more seriously than powdered cocaine, which, given the usage patterns for the two forms of cocaine, creates a disadvantage for minority youth. Likewise, policies that treat gang activity more seriously than comparable activity by nongang members may place minorities at a disadvantage based on greater likelihood they will be perceived as gang involved. Policies that target location issues (e.g., certain types of offenses near schools or public housing areas) may place minority youth at a disadvantage given the location characteristics. For example, an Illinois automatic transfer law mandates that 15- and 16-year-old youth charged with a drug offense that occurs within 1,000 feet of a school or public housing project are automatically tried in adult court. Although white youth use and sell drugs at similar or higher rates than youth of color, the impact of the law has almost wholly affected African American and Latino youth (www.buildingblocksforyouth.org/illinois/). As another example, Portland, Oregon, has an ordinance permitting police to exclude an individual from specified drug- or prostitution-―free‖ zones. Individuals violating that exclusion order are subject to arrest for trespassing. Since those zones are in portions of the city with the highest proportions of minority residents, it is not surprising that the ordinance creates DMC results. Policies that mandate specific handling (e.g., moving a case to adult court) may have eligibility or threshold criteria based on prior delinquency or offense histories. The use of criteria such as ―three strikes‖ may place a minority youth at a higher risk of application of such sanctions when minority youth as a category have more extensive records of justice system contact (e.g., Feld, 1999; Bishop, 2005). This is not to say that all such policies or practices that result in differences in treatment are necessarily wrong or need to be modified. What is suggested is that if such policies result in accentuating DMC, then policymakers, analysts, and community members should be aware of those consequences and ensure that the policies are well founded and that the jurisdiction considers whether to continue those policies, end them, or seek to modify them in order to address DMC effects. In some communities, for example, an intentional decision to reduce gang activity may result in an increase in DMC measures, which is predicted and understood as a consequence of that public safety objective. The point is not to expect to eliminate all such disparate impacts at once, but rather to examine and monitor these impacts when they occur to ensure that public safety, rehabilitation of gang members, and fair juvenile justice system response all are kept in an intentional balance. For example, although a short-term increase in DMC may be likely to result from a gang-suppression initiative, DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-9 the long-term expected outcome of a comprehensive and balanced approach would be reduced levels of DMC. Accumulated Disadvantage One of the more disturbing aspects of the DMC issue is that the impact on minority youth as a group tends to accumulate, rather than dissipate, through the system,. This phenomenon is displayed in at least two different ways. Simple Accumulation There may be a higher rate of arrest for minority youth, followed by a lower rate of diversion, higher rates of formal processing as delinquent, etc. In most stages of the juvenile justice system, minority youth (particularly black and Hispanic youth) appear to receive handling that is either harsher than their white counterparts or equal to their white counterparts. Thus, although the differential treatment at any particular stage may appear ―small,‖ the cumulative impact across the entire juvenile justice system may be relatively large. The impact here is essentially equivalent to compound interest—a 10-percent difference in volume of activity (RRI value = 1.10) that occurs at each of eight decision stages in the juvenile justice system will accumulate into a rate of DMC that is more than double the level of overall contact for that minority group. Impacts On Later Decisions Another example where race and ethnicity may work indirectly through factors that influence decisionmaking is the impact of earlier stages on later stages of the justice system, such as the impact of pre-adjudicatory detention. Studies have indicated that decisions made at earlier stages, such as detention, affect outcomes at later stages and, in particular, judicial disposition. That is, detention strongly predicts more severe treatment at judicial disposition. Although minority youth and white youth who have been detained may be treated similarly, because the former group is more likely to be detained, they receive more severe dispositions than do their white counterparts. Consequently, race or ethnicity may not directly influence judicial disposition, but its effects may be masked, operating through a racially linked criterion of pre-adjudicatory detention (e.g., Leiber, and Fox, 2005). Stage 2: Identifying Data Types and Expected Results Once the team has identified a short list of potential mechanisms that it will explore, the next task is to identify the types of data and results that might be expected to be consistent with those hypothesized mechanisms. The logic here is not to prove that a particular mechanism is at work, but rather to explore the possibility that it is at work and to rule out those mechanisms that have less support and are not consistent with the data available in the jurisdiction. The table on pages 12–14 summarizes the types of data needed and the types of data patterns you might expect for each mechanism. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-10 Most of the data types and patterns are based on the expectation that the jurisdiction has some form of database available that can provide refined information. In the examples noted above with respect to the importation mechanism, to test this mechanism, one would first examine the RRI values based only on cases involving residents of the jurisdiction. For example, if an attractive nuisance is expected to draw youth from outside the county, then assessing the RRI values when calculated only for youth residing in the county should provide a much lower RRI value (closer to 1.00). You can conduct that analysis only if the county has a database that includes information on whether a youth is a resident of the county. Likewise, examining geographic access requires that the county locate both the residences of the youth and the location of programs. Three general types of data are suggested in the following table. The first is termed RRIlevel data, which means the ability to create volume counts for subsets of cases, as in the examples in the preceding paragraph. The second is termed case-level data, which implies the ability to examine attributes of specific cases and combine them in tabular form. For example, in the assessment of differential opportunities for prevention or intervention programs, issues of program retention and completion are noted. Those issues will require (1) specific data about juvenile cases referred to such programs and (2) an ability to count the cases (youth) who enter, stay, and complete the programs. The third level of data is termed transactional data; this means data systems with the ability to track individual cases through multiple stages in the juvenile justice system and attach many attributes of the youth to the data—for example, any of the items considered under the heading of indirect effects. If a community does not have a data system adequate to provide the needed information, it will have to design methods to create or acquire data with those characteristics for the assessment process. Because it is anticipated that many communities are likely to fit this description, a subsequent section of this chapter discusses the design of methods for gathering such data. It is also likely that some communities will have database systems that can address many, but not all, of the analyses projected here. For such communities, it may make sense to use their database systems to move as far as possible in the assessment process and then supplement those findings with additional data collection and analysis strategies along the lines of those suggested in the sections on stage 3 and stage 4. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-11 Data Types and Expected Patterns Resulting From Various Mechanisms That Create DMC DMC Mechanism Data Type and Analysis Data Pattern Expected Transactional data, multivariate analysis, or multiple controls in cross-tabulations to explore RRI subsets. High correlation of offense type, age at first offense, or other system involvement with race or ethnicity. 1. Differential Behavior Different offense categories. More frequent involvement. When multivariate analysis is conducted and the variables representing offense type, age at first offense, or other system involvement are used as control variables, the correlation of race/ethnicity with system contact stages is significantly reduced, or Involvement at an earlier age. Involvement with other service systems. When subsets with similar characteristics are tested, the RRI value is reduced. 2. Mobility Effects: Importation/Displacement Case-level data with Seasonal mobility. information about residence of Attractive nuisance. youth, nationality, and Immigration and migration. seasonality. When RRI scores are calculated based only on resident youth, the RRI values should be substantially lower than the values calculated for all youth. If seasonal mobility is expected, then the nonresident cases will cluster within expected date/time slots. 3. Indirect Effects Specific risk factors. Access and eligibility for programming. Decisionmaking factors. Transactional data with information on characteristics thought to result in the indirect effects (e.g. economic status, family structure, detention status). High correlation of the variables believed to carry the indirect effects with race/ ethnicity. When multivariate analysis is conducted, the correlation of race/ethnicity with system contact stages is significantly reduced, or When subsets of cases with similar characteristics are tested, the RRI value is reduced. (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-12 Data Types and Expected Patterns Resulting From Various Mechanisms That Create DMC (continued) DMC Mechanism Data Type and Analysis Data Pattern Expected 4. Differential Opportunities for Prevention, Treatment Access. Eligibility criteria. Implementation characteristics. Case-level data with information about program entry, retention and outcomes, residential location. Effectiveness. Program utilization rates for services differ by race/ethnicity; geographical mapping of service locations does not correspond with locations of minority youth. Stated eligibility criteria are correlated to race/ethnicity within the pool of those who might participate in the program. Program retention/ completion data are correlated with race/ethnicity among those who enter the program. Qualitative studies of the program climate, customer satisfaction studies of participants, and those who do not complete the programs show racial/ethnic differences. Outcome measures among program completers are correlated with race/ethnicity. 5. Differential Processing/Decisionmaking Criteria What are the criteria on which decisions are made? Are those criteria applied consistently? Are the criteria stated to create disadvantage? Transactional data with information about the variables or items that may be used as decision criteria. Multiple regression or crosstabulations with controls. The expected criteria are closely related to decision outcomes. The criteria are related to race/ethnicity within the set of cases eligible for each decision. (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-13 Data Types and Expected Patterns Resulting From Various Mechanisms That Create DMC (continued) DMC Mechanism Data Type and Analysis Data Pattern Expected Transactional or case-level data with information on characteristics thought to impact case outcomes. Compare jurisdictional case outcomes and identify various structural characteristics of communities and organizational characteristics of the juvenile court, for example: 6) Justice by Geography Differences in localities may exist in terms of crime, detection and court referrals, case proceedings and outcomes, and development and use of services that may be the result of factors unique to localities. Multivariate analysis or multiple controls in crosstabulations RRI-level data. Jurisdictions have a wide variation in the rates of juvenile justice system activity. The geographic distribution of minority youth populations correlates strongly with the variation in rates of juvenile justice system activity. 7. Legislation, Policies, Legal Factors Case-level data showing who is Policies about offense affected by the policies and types or characteristics. who is not. Policies about location issues. Impact rates (cases that the policies affect) are greatly different by race/ethnicity. Policies that mandate specific handling. 8. Accumulated Disadvantage Simple accumulation. Impacts on later decisions. RRI-level data, transactional data including race/ethnicity and the outcomes of multiple decisions within the juvenile justice system. Relatively low RRI values at most decision points, all in the direction that indicates disadvantage for minority youth. Decisions early in the system, especially detention, will be strongly correlated to later decisions. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-14 Stage 3: Obtaining the Data The extent of DMC and the contributing factors varies by state and within individual jurisdictions. Recognizing this, OJJDP encourages states and localities to develop innovative approaches to conduct the assessment. A DMC assessment, however, must resolve several methodological issues, including which jurisdictions and decision points and what type of research design and data or subjects are most appropriate and feasible. Before addressing these methodological issues, this section discusses the need for state and local DMC committees to plan and collaborate with researchers on the DMC assessment study before, during, and after it is undertaken. Planning The process of planning should be examined from at least two perspectives: that of the state or local agency, and that of the persons or groups conducting the assessment study. State and/or Locality Perspective The DMC lead agency typically coordinates DMC activities statewide and gets local decisionmakers and other community representatives involved at both the state and local levels. This collaboration between state and locality is extremely important in the assessment phase. The information needed to conduct the assessment can be quite extensive and often crosses agency lines, so it is imperative that key agency personnel participate in the process from the start. These individuals will know what the data are, what problems might exist with the data, and what barriers interested parties may face when attempting to access the data. Further, DMC tends to be an emotional issue, and, with its emphasis on causes, some may view the assessment phase as a mechanism for placing blame or attempting to enact quotas. Therefore, participation of key personnel at the onset can help alleviate these concerns and facilitate the collection of the data. Thus, there is a need from the start to discuss and identify who the key actors and agencies are and to map a strategy to involve them before any assessment study is conducted. Once central personnel are on board, a committee should discuss issues that pertain to the cost of the study, what should be studied, what kind of assessment study should be conducted, and what the process will be for recruiting someone either internally (inhouse) or externally (such as an agency or a university) to conduct the study. The committee should examine past assessment research that has been conducted locally as well as national studies to provide direction for the proposed study in terms of what kind of assessment may be needed and the skill needed to conduct the research. In the past, few state planning agencies, state advisory groups, or DMC subcommittees (state or local) possessed, inhouse, the technical expertise to conduct a formal assessment study. A formal study generally involves both quantitative and qualitative techniques that include following the same youth from initial contact with the police or the juvenile court to a final case outcome. The use of multivariate analyses is also incorporated to examine the relationships of many factors (i.e., race, ethnicity, gender, crime type, crime severity, DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-15 etc.) at one time on decisionmaking. Because of the complexities of a formal assessment study, many states and localities choose to contract with other agencies, organizations, or universities that possess the needed qualifications. Ideally, an assessment study should examine as many decisionmaking stages as possible with relevant independent variables and the use of multivariate procedures. This kind of assessment study takes into consideration decisionmaking as a process and attempts to emulate the factors that influence case proceedings. If data and resources are not available to conduct such a detailed assessment study, states and/or localities, with input from the researcher, may decide to conduct a study that focuses only on those decision points with the highest RRI values, to include fewer independent variables, or to exclude multivariate analysis. While reducing the decision points examined, reducing the data elements included, or excluding multivariate analysis will reduce the time and resources needed to conduct the study, the results of the study will likely inspire less confidence than a more robust study. Regardless of what kind of assessment study the committee plans, jurisdictions should consider naming a person who is involved in the research as a member of the DMC committee. This provides the committee access to information on the progress of the assessment study and the chance to benefit from the experiences and knowledge of the person or group conducting the research. In addition to discussions concerning the specifics of the assessment study, the committee should also focus on issues involving the delivery of the final product. Things to consider are a period of time to provide feedback on drafts prior to the completion of the writeup of the findings, at a minimum a final report that includes not only the results but recommendations, and oral presentations to the committee as well as to the state advisory group. OJJDP encourages the state and/or locality to contact the Office to address issues that may arise concerning the planning phase of the assessment research. Some states and localities, for example, have requested technical assistance to help in the planning phase. DMC Assessment Researchers Many of the same things discussed with regard to planning for the state and DMC subcommittees apply to the researcher(s) considering undertaking the assessment study. In addition to those concerns, the DMC researcher should be clear about what the committee wants and expects. In most instances, the state and DMC subcommittees will rely on the researcher for input and direction for what should be done and how the assessment can be conducted. Besides the expertise that the researcher may possess, it is imperative that he or she also examine previous assessment studies to help in the process of planning the research and working with the state and/or localities. As part of the planning, the researcher should, in conjunction with developing what is to be studied (including what type of assessment study should be conducted), work with the state and/or locality to identify the data DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-16 source(s) the researcher might need to conduct the assessment research. It is imperative that the researcher develop a thorough understanding of the workings of the particular juvenile justice system(s) to be studied. It is through this understanding that the researcher may assist the state/locality in developing a study that gets at the important issues for the system. The committee and the researcher should discuss deliverables and timeframes early in the process. That is, what kind of report should the researcher develop and what should the researcher include in the report. Other responsibilities could include being a member of a state or local DMC subcommittee, providing oral presentations, and working with federal technical consultants. The committee and the researcher should also discuss what responsibilities, if any, the researcher may have once the assessment research is completed. This may include such issues as who owns the data once they are collected and what can be done with the data following the completion of the assessment study. That is, can the researcher attempt to publish the assessment study and other work from the data? If so, what responsibilities does the researcher owe to the state and/or locality? Methods Because the extent of DMC and the contributing factors varies among jurisdictions and the data and resources available to conduct an assessment study may vary, OJJDP encourages states and localities to choose the type of assessment study that meets their needs. Irrespective of what type of assessment study a jurisdiction conducts, however, it must address several methodological issues. Site Selection To conduct a DMC assessment, either a formal assessment study or something less, the state, jurisdictions, or localities must decide where to focus their efforts. Although a state would ideally choose to conduct a statewide DMC assessment, financial and time limitations as well as practicality may prevent such an undertaking. Results from the RRIs, census information, and crime reports should provide a guide to areas for study. For example, a jurisdiction should direct any DMC effort where it can influence the lives of as many youth as possible; therefore, those jurisdictions with a high concentration or large number of minority youth are good candidates. The site selection process should also consider other structural factors, such as urban versus rural settings, and the concentration of racial poverty and inequality (Sampson and Laub, 1993). For example, in Washington State, disproportionality was associated with urbanization and levels of violent crime and chronic offending. While most assessment studies have focused site selection efforts exclusively on counties (Leiber, 2002), others have gone a step further and identified areas within the selected counties for assessments. Smaller units of geographical measurement can provide more helpful information for deciding on action or interventions to address DMC than larger DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-17 areas. In Pennsylvania, for example, police precinct data were obtained in the counties that had the greatest problems with minority overrepresentation to determine where the activities selected to address DMC could have the greatest impact. This type of information enabled Pennsylvania to channel resources to the localities where the greatest impact could be expected. Although dependent on the availability of resources and what the RRI reveals, site selection should include at least two to three areas, if not more, to allow for comparisons and consideration of the effects of justice by geography. What may explain DMC in one jurisdiction may not explain DMC in another. Decision Points Selection A number of studies have shown that minority overrepresentation may occur at any point in the system (e.g., Pope and Feyerherm, 1992; Bishop, 2005; Hamparian and Leiber, 1997; Leiber and Mack, 2003; Hsia, Bridges, and McHale, 2004). Understanding the relationships among decision points is also critical. The experiences of OJJDP’s five DMC pilot states indicate that ―understanding overrepresentation is a matter of understanding how a specific juvenile justice system operates, with all its interdependent parts, to result in more minority juveniles entering and penetrating further into the system‖ (Devine et al., 1998: 4). As discussed, the influence of race or ethnicity on any one point may be enhanced or canceled out at a following point. Thus, the assessment research should approach the juvenile justice system in a holistic manner. Because the effect of previous decisionmaking may influence a youth’s further movement into the system and because this effect may be greater for minority youth, the committee must consider multiple decision points to capture the process of decisionmaking. Examining all the significant contact points (police contact, arrest, referral to juvenile court, intake, diversion, petition, adjudication, judicial disposition, detention, transfer to adult court) will provide the greatest confidence in understanding where, how, and why DMC exists (Nellis, 2005). In most states, though, considerations of funding, time, data, and logistics prevent such a comprehensive formal assessment. If a jurisdiction must limit the scope of its research to just a few decisionmaking points, the most important determinant of the points to be studied should come from an examination of the RRIs and previous research, including past assessments. In addition, states may try to use answers from questions concerning why DMC exists to aid in the determination of what stages to examine. Research Design Most states and localities have relied on either quantitative or qualitative research designs to conduct their assessments (Pope and Leiber, 2005). Quantitative data are in the form of numbers, such as the number of referrals to nonsecure facilities or the number of youth arrested. Quantitative studies are used when statistically reliable results are desired. Qualitative data are used to gather indepth information about something but do not provide statistically reliable results. Qualitative data are often in the form of words or text, not numbers, e.g., a description of the decisionmaking process the juvenile probation officer uses or the text of written policies and procedures. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-18 Results from a quantitative study typically allow for more generalizations concerning decisionmaking practices and procedures than those obtained from a qualitative study. On the other hand, a qualitative approach can provide better insights and allow for a meaningful interpretation of quantitative data. To understand why DMC exists, both quantitative and qualitative data must be examined. Quantitative Research Design To undertake a quantitative study, researchers must consider a number of factors. Sampling Depending on a variety of factors, a researcher may not want to examine every case processed in the juvenile court or every police-youth contact but instead may choose a subset or sample of those cases, especially if the jurisdiction in question processes a large number of cases in the specified timeframe. The sampling technique the researcher will use depends, to a large degree, on the decision points to be examined, as well as the number of cases processed or size of the juvenile court. If, for example, the research examines the police decision to arrest, the researcher must ensure that the available data are representative of police contacts with minority youth: there must be an adequate number of cases for each type of police decisionmaking outcome (i.e., release, referral, arrest). If the juvenile court is the target of the assessment, a researcher must first determine the number of cases to be included in the research and calculate the number of youth for each racial group to be studied. In an analysis that will examine decisionmaking across a number of decision points, it is imperative to start with an initial sample of 500 to 1,000 cases per research site. This is because, given that youth drop out from the point of intake to judicial disposition, most often only 10 percent of the cases at intake reach judicial disposition. As a result, with an initial sample of 500 to 1,000, it is likely that only 50 to 100 cases will be available at the judicial disposition stage for purposes of analysis. Decision Points To Be Examined Because most state assessment efforts have focused on the decisions encompassed by the juvenile court and, to a lesser degree, arrests, the following discussion centers on police and the juvenile court. Police. Variable selection for this decision point (arrest or not arrest) requires that the researcher ask the question: What factors play a role in a police officer’s decision to arrest? Unfortunately, many of the existing client-tracking data systems in juvenile justice do not include arrest information; the first point of contact reported in the system is usually referral to juvenile court. The type of information that law enforcement agencies routinely collect also tends to be of limited utility in conducting quantitative analyses of minority overrepresentation because they do not record many of their contacts with youth. For example, data may be available DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-19 on youth who were arrested but not on youth who came into contact with the police and were just warned and released. If researchers examine police case files or reports, information is needed on all police youth contacts, not merely those resulting in arrest. There is little research on the relationship between law enforcement decisionmaking and minority overrepresentation (Conley, 1994) that provides guidance on the selection of variables. The research that does exist suggests that variables must reflect certain characteristics of the youth, the situation that led to police involvement, the officers themselves, victims, the community, and how the police agency is organized (e.g., Black and Reiss, 1970; Carter, 1986; Harstone and Richetelli, 2001; Jackson, 1992; Pope and Snyder, 2003; Sampson, 1986; Smith and Visher, 1981; Smith, 1986). Examples of such variables are presented in exhibit 1. Exhibit 1: Examples of Variables That Might Influence the Decision To Arrest Characteristics of Youth Officer Characteristics Community Characteristics Race Race Economic situation Age Age Racial/ethnic composition Gender Gender Extent of racial segregation Demeanor Education Status of race relations Family situation Length of service Knowledge of the suspect Characteristics of Contact Victim Characteristics Police Agency Characteristics Type of crime Age Police expenditures Reason youth was contacted, taken into custody, or arrested Race Deployment patterns Gender Organization ideology Involvement of a weapon Place of contact Presence of bystanders Victim’s wish to press charges Relationship between youth and victim DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-20 Juvenile Court. Although key decision points of the juvenile justice process may vary across the country, similarities include referral to juvenile court, diversion, detention, petition/charges filed, delinquency findings, probation, confinement in a secure correctional facility, and transfer to adult court. When the juvenile court is the subject of research, and if resources and the data allow, decisionmaking at these stages should be examined. If not, at least two or more stages should be examined. Data/Analysis Considerations Variables. The previous section focused on the decisionmaking points that could or should be examined. In addition to using decisionmaking data, it is important that researchers include independent variables in the assessment study. The independent variables should include the seriousness of the case (e.g., type of charge, severity of the charge, use of a weapon, victim injury) and the juvenile’s prior involvement with the juvenile justice system (e.g., prior referral, adjudication, placement), as well as ―extralegal‖ factors such as race, ethnicity, gender, family status, etc. (e.g., Bishop and Frazier, 1988; Bridges et al., 1995; Leiber and Fox, 2005; Leiber, 2003; Sampson and Laub, 1993). In multivariate analysis, the more information on these independent factors that are included, the greater the confidence a researcher can have in the results. Examples of variables are presented in exhibit 2. Exhibit 2: Examples of Variables That Might Influence Juvenile Court Decisionmaking Characteristics of Youth Characteristics of Current Offense Community Characteristics Race Race Economic situation Gender Age Racial/ethnic composition Age Gender Extent of racial segregation Demeanor Education Status of race relations Family situation Length of service School situation Knowledge of the suspect Prior Court Involvement Prior delinquency Severity of past disposition Youth under authority of court at time of current offense Other Characteristics Cooperativeness of youth and family Youth’s mental health history Type of legal representation Race of victim Relationship between victim and youth DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-21 Missing Decision Points. If data that are crucial for a particular decision point are not readily available via computer records, researchers should consider manual abstraction of the data. For example, if in a particular jurisdiction, system practitioners theorize that the disparity in the use of pre-adjudicatory detention for minority and white youth is attributable to the fact that parents of minority youth are more difficult to contact, they should undertake an effort to gather that data. The police or detention staff may record whether a parent was contacted on a form even though they may not enter the information into a computer. If researchers think this issue may have a major impact on what happens to the youth, it is important that they include this information in the multivariate analysis. The analysis may reveal that it is not a significant factor in the greater use of detention for minority youth, in which case system practitioners would need to be educated. However, if it is revealed that not being able to reach a parent is a significant factor in the use of detention, then strategies could be developed to address the issue. Defining Race and Ethnicity. Of all the pieces of information to be collected, it is perhaps most critical that ―race/ethnicity‖ and ―referral to juvenile court‖ be correctly and consistently defined. Many states have treated race as a dichotomy: white versus minority (Hamparian and Leiber, 1997; Pope et al., 2002). This classification of race fails to capture differences in case processing and outcomes that may exist among different minority groups, defined both in terms of racial grouping and ethnicity. Disproportionate minority arrest, secure detention, and commitment to secure corrections are not equivalent issues for all minority groups: in most states and localities, African American juveniles are arrested and confined at a greater rate than youth of other minority groups. Defining Decision Outcomes. Similarly, the disposition at intake has been inappropriately defined, most often as release/diversion versus a recommendation for further court proceedings or petition. Putting released youth and diverted youth into one category may mask differences in the use of release and participation in diversion for whites compared with minorities. Previous research and results from state RRIs have shown that white youth are more likely to be diverted from formal court proceedings than are minority youth (Leiber and Stairs, 1999), and the failure to differentiate among these outcomes precludes an examination of this important decision. Appendix A of this chapter presents an example of a survey instrument that includes definitions of decision outcomes as well as variables that may influence juvenile court decisionmaking. Qualitative Research Design Although quantitative research will help a jurisdiction determine the precise decision points at which DMC may occur and the factors that may significantly impact decisionmaking, researchers can use qualitative research to develop a deeper understanding of some of the issues around DMC and decisionmaking. Types of Qualitative Research Typically, qualitative research includes focus groups and/or indepth interviews. Focus groups of 8 to 10 participants, lead by a moderator using a semistructured discussion guide, are brought together to discuss a particular issue. The moderator ensures that all DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-22 participants are given the opportunity to contribute to the discussion, and the group explores the various important aspects of the discussion topics. Indepth interviews are typically semistructured, one-on-one discussions between an interviewer and interviewee. Appendix B of this chapter presents an example of a semistructured interview instrument. Selecting a Qualitative Research Methodology There are a number of factors that the researcher must consider when determining whether to use focus groups or indepth interviews: geography, candor/confidentiality, and cost. Geography. Traditionally, a focus group includes participants from a small geographical area because they would all need to travel to one central location for the meeting. A study might include a series of groups to cover the various geographic areas within a jurisdiction. Some jurisdictions may have access to videoconferencing technology that makes it possible to include participants who are geographically dispersed in the same focus group. Indepth interviews are useful when the participants are geographically dispersed, making it difficult to get many of them to a central location for a focus group. The researcher conducts the interviews one participant at a time, typically at a location that is convenient for the interviewee (e.g., his or her office). When necessary, the researcher can conduct indepth interviews via telephone; however, it is preferable to conduct the interviews in person, as a rapport develops more easily between the interviewer and interviewee when they meet face-to-face. Candor/Confidentiality. When conducting focus groups with system practitioners, it is most effective to hold separate groups with administration and line staff from the agencies. This will increase the comfort level of line staff so they can speak honestly and candidly about their experiences within their agency without risking on-the-job repercussions. In addition, it may be necessary to hold separate focus groups with personnel from different justice system agencies. In some jurisdictions, police officers, probation officers, and corrections officers may hold a constructive group conversation, but in other jurisdictions, the group may become a forum for the ―blame game,‖ with each agency blaming the other for the justice system’s problems. As indepth interviews are conducted one-on-one, confidentiality is less of an issue, assuming the interviewer gains the confidence of the interviewee. Cost. No hard-and-fast rules exist on how many focus groups or indepth interviews researchers should conduct for a particular study. It is always necessary to balance the issue of cost against the number of participants included in the qualitative research. Obviously, the more focus groups/interviews conducted, the greater the costs. However, it is important that the study include enough focus groups/interviews to ensure that the findings are not based only on certain geographical areas, certain types of system practitioners, or a few strongly opinionated practitioners. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-23 There are many variables that can greatly impact the cost of conducting focus groups or one-on-one interviews. Focus groups provide the advantage of gathering information from multiple practitioners (e.g., 8 to 10) at the same time. However, if a series of groups must be conducted to cover various geographical areas, and separate groups need to be held with administrators and staff as well as different justice agencies, the number of groups can become large, greatly increasing the cost of the qualitative research. Where the focus groups are held will have a big impact on the cost of the groups. Professional facilities equipped to handle the logistics of conducting focus groups provide a neutral environment in which to hold the discussions. Although there are obvious advantages to conducting focus groups in such facilities rather than in a conference room of a local state office building, it is more costly to do so. One-on-one interviews involve time and travel expenses for each interview conducted. If a large geographical area must be covered, the travel expenses can add up. Although it is possible to conduct the interviews via telephone rather than in person, thereby reducing the travel costs per interview, it is more difficult to get the interviewee to talk candidly about difficult issues. When To Use Qualitative Research It is important to remember the strengths and weaknesses of qualitative research. Its biggest strength is that it provides indepth information about issues; its biggest weakness is that there are no means of statistically assessing the confidence to place in the information. To examine DMC issues, qualitative research should typically be used with quantitative research, not instead of quantitative research. Prior to Quantitative Research. Qualitative research can provide researchers with a thorough understanding of the juvenile justice system being examined. Although researchers can learn how a particular jurisdiction works based on a review of legislation and policy and procedure manuals, they can use qualitative research to learn how practitioners actually implement the system. A cautionary note about using qualitative research prior to quantitative research: It is important that the qualitative research not be used to limit the scope or focus of the quantitative research. By its very nature, qualitative research includes a small number of participants. Therefore, having a small number of system practitioners determine which decision points or which areas of the jurisdiction should or should not be examined using quantitative methodologies could lead to the exclusion of important information from the study, resulting in a flawed assessment of the justice system. After Quantitative Research. Qualitative research can be useful after quantitative research has identified specific problem areas within the juvenile justice system. Via focus groups or indepth interviews, system practitioners can provide their perspectives on what may contribute to DMC at particular decisionmaking points. In addition, researchers can ask practitioners for their suggestions on how to address the problems found. This provides an opportunity for possible solutions to come from different perspectives within the DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-24 various justice agencies (and from different levels within the agencies). Also, given that system practitioners must ultimately implement any actions developed to address DMC issues, providing them with an opportunity to make suggestions at the beginning will increase their motivation to implement the changes. Stage 4: Analyzing the Data and Identifying the Most Likely Mechanism(s) Creating DMC Once data are obtained, the next step is to analyze the data to determine if they fit the patterns expected in terms of the DMC mechanisms identified. The table in the discussion of stage 2 outlines many of the patterns you might expect. However, a jurisdiction and its research team must always keep an eye out for unexpected results. Just because the key system players did not identify a particular mechanism during the planning process does not mean that it is not an issue within the jurisdiction. As noted previously, the kind of statistical analysis procedures that researchers might use depends on the level of confidence desired for the assessment findings and on the data that may be accessible for the study. Cross-tabulations and other statistical procedures that compare only two variables at a time are relatively simple to do and can provide very useful information. Because these types of statistical procedures allow for the examination of associations rather than prediction, the level of confidence in the results would be lower than would be the case with research that used multivariate procedures. For example, analysis of race by detention status may inform researchers that a relationship exists between the two. However, the effects of the severity of the charge, prior record, age, and other factors on the decision to detain are unknown. Thus, it is not possible to determine whether it is race/ethnicity and/or other legal and extralegal variables that explain the decision to detain. To increase the level of confidence, researchers may want to test for relationships beyond two-way comparisons and attempt to include as many variables as possible. Multivariate techniques, on the other hand, make it possible to estimate the influence of a variable on a dependent variable or a decisionmaking stage while simultaneously controlling for the effects of a large number of other variables. Ordinary Least Squares Regression (OLS) and Logistic Regression are two examples of multivariate techniques. The latter procedure is the statistical technique researchers are most likely to use, because many of the decisionmaking stages can and should be expressed as a dichotomy (e.g., adjudicated delinquent versus not, detained versus not). Further, specific techniques allow researchers to explore the possibility that factors other than race and ethnicity may condition decisionmaking or work in combination with other variables. Being African American and female may have an association with case outcomes, whereas just being African American may not have such an interaction effect. The use of multivariate analyses in the form of regression also allows researchers to test for indirect racial/ethnic effects. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-25 It is not the purpose of this section to provide a primer on data analysis, because the analyst or contractor conducting the DMC assessment will have a basic understanding of the subject. However, some nuances of this type of analysis should be described— perhaps most importantly those concerning the methods of multivariate analysis or, more specifically, conducting either regression-based analyses (i.e. logit analysis or hierarchical multiple regression) or cross-tabulations with multiple controls. In either instance the logic is similar, consisting of several simple questions: Does the variable (mechanism) have the hypothesized relationship to race/ ethnicity? Does the variable (mechanism) have the expected relationship to the juvenile justice decision that is being explained? If the impact of that variable is held constant, does the relationship of race/ ethnicity to juvenile justice system decisions become markedly reduced? If the answer to all three items is yes, then support exists for the interpretation that this mechanism helps to explain DMC. An example may help: See exhibit 3 (the numbers presented are hypothetical; the examples are based on real situations). Exhibit 3: Example: Detention, Access to Alternatives, and Geography Assume that a community has a high RRI value for African American and Hispanic youth for the detention stage. After discussing the issue with law enforcement officers, detention workers, judicial staff, community agency directors, and others, the research team believes that access to detention alternatives may explain part of that high level of DMC. Researchers collect information about the location of alternative programs such as afterschool centers and other options and classify the neighborhoods in the community into those that have available options for youth and those that do not. Of the 3,000 youth referred to the court each year for possible detention intake, researchers can identify the neighborhoods in which most of them live. In the identification stage, the researchers found results as shown in table 1, as follows: Table 1 White African-American Hispanic Arrest Volume 1550 900 550 Detention 218 252 143.5 Detention Rate 14.06 28.00 26.09 RRI 1.99 1.86 In examining the first question above, researchers find that the available detention alternatives definitely do not serve minority youth. As seen in table 2 below, while overall 48 percent of youth live in neighborhoods with alternative programs, only 11 percent of African American youth and 27 percent of Hispanic youth live in such neighborhoods. (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-26 Exhibit 3: Example: Detention, Access to Alternatives, and Geography (continued) Table 2 White African American Hispanic Total Live in neighborhoods with detention alternatives? Yes No 1200 350 100 800 150 400 1450 1550 Total 1550 900 550 3000 Percent yes 77 11 27 48 In addressing the second question (see table 3 below), researchers find that the availability of detention alternatives is clearly related to the use of detention. Youth from the neighborhoods with available services are detained at a 10-percent rate when arrested, while those from neighborhoods not served show a 30-percent rate of detention. Table 3 Live in neighborhoods with detention alternatives Yes No Total Detained? Yes No 152 1298 462 1088 614 2386 Total 1450 1550 3000 Percent Detained 10 30 20 Finally, to answer the third question, researchers can look at table 4, which contains all three variables: race and ethnicity, neighborhood, and detention. Table 4 Live in Neighborhoods With Alternative Services Arrest Volume Detention Detention Rate White 1200 120 10.00 African American 100 12 12.00 Hispanic 150 20 13.33 Live in Neighborhoods Without Alternative Services White 350 98 28.00 African American 800 240 30.00 Hispanic 400 124 31.00 RRI 1.20 1.33 1.07 1.11 In this table, compared with the table that came from the identification process, the size of the RRI values is substantially smaller, indicating that a substantial part of the impact of race on detention is carried through the neighborhood in which the youth lives, and especially whether that neighborhood has available alternatives as a substitute for detention. In this instance, researchers would conclude that geographic access is a mechanism worth addressing in terms of DMC at the detention stage. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-27 In reality, Multnomah County, Oregon, reached results like those in exhibit 3 when it assessed the mechanism contributing to DMC at the detention stage. In addition, researchers hypothesized other mechanisms that, in fact, supported an analysis similar to the suggestions here for the assessment phase. Not only was there a gap in the geographic availability of alternatives to detention services, but the researchers’ analysis identified and confirmed two other issues. The first was related to the implementation of services in the juvenile department, specifically to the cultural competence of staff. The county used extensive training programs and intentional recruitment and hiring to increase the number of staff from racial and ethnic minority populations. Beyond that, the researchers’ examination of decisionmaking criteria revealed that the criteria were not being applied with the desired level of consistency and held substantial disadvantages for minority youth. The Annie E. Casey Foundation and others have documented the resulting interventions elsewhere. The point for this manual is that jurisdictions should not assume that only one mechanism is at work to create DMC. It may be very likely that the analysis will support a finding that several mechanisms are in place and that the successful intervention strategy may be one that addresses multiple mechanisms. Once the data analysis has taken place, it is useful and necessary to describe the results to the groups who helped to identify the possible lists. This step is necessary for several reasons. First, it provides feedback for those groups with respect to the ideas that they generated, showing which ones appear to be supported and which ones are not supported. It permits clarification and testing of the explanations that are used. The objective of the assessment step is not just to collect data; it is to generate explanations for DMC that jurisdictions may use to address the issue. From that vantage point, those explanations are essentially stories about how the juvenile justice system operates. It is important to check the plausibility of those stories and explanations. By focusing on the plausible explanations, it may be possible to start these same groups thinking about solutions. Finally, by identifying the mechanisms that are at work to create DMC, one also sensitizes those working in the system to those mechanisms, serving not only to help them avoid using those mechanisms but also to make them aware of changes and help them watch for changes in the future. This may assist in the ongoing monitoring of DMC in the jurisdiction. Conclusion This chapter lays the groundwork for the assessment process, which means asking questions about how DMC is created within a jurisdiction and then obtaining data to validate the answers received to those questions. The objective is not to arrive at a complete sociological and psychological description of the juvenile justice system but DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-28 instead to reach a plausible understanding of the way the juvenile justice system operates and creates DMC. Ultimately, the objective of the assessment phase is to provide enough information so that jurisdictions may choose to implement DMC reduction strategies and interventions based on evidence from their own community. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-29 References Bishop, D. 2005. ―The role of race and ethnicity in juvenile justice processing,‖ in Our Children, Their Children: Confronting Racial and Ethnic Differences in American Juvenile Justice. D. Hawkins and K. Kempf-Leonard (eds.). The John T. and Catherine MacArthur Foundation Research Network on Adolescent Development and Juvenile Justice. Chicago, IL: University of Chicago Press, pp 23–82. Bishop, D., and Frazier, C. 1988. ―The influence of race in juvenile justice processing.‖ Journal of Research in Crime & Delinquency 25: 242–263. Bishop, D., and Frazier, C. 1996. ―Race effects in juvenile justice decision-making: Findings of a statewide analysis.‖ Journal of Criminal Law & Criminology 86:392–414. Black, D., and Reiss, A. 1970. ―Police control of juveniles.‖ American Sociological Review 35: 63–77. Bridges, G., Conley, D., Engen, R., and Price-Spratlen, T. 1995. ―Racial disparities in the confinement of juveniles: Effects of crime and community social structure on punishment,‖ in Minorities in Juvenile Justice, K. Kempf-Leonard, C. Pope, and W. Feyerherm (eds.). Thousand Oaks, CA: Sage Publications, pp. 128–152. Bridges, G., and Steen, S. 1998. ―Racial disparities in official assessments of juvenile offenders: Attributional stereotypes as mediating mechanisms.‖ American Sociological Review 63:554–570. Carter, D. 1986. ―Hispanic police officers’ perceptions of discrimination.‖ Police Studies 9(4): 204–210. Conley, D. 1994. ―Adding color to a black and white picture: Using qualitative data to explain racial disproportionality in the juvenile justice system.‖ Journal of Research in Crime & Delinquency 31:135–148. Devine, P., Coolbaugh, K., and Jenkins, S. 1998. Disproportionate Minority Confinement: Lessons Learned From Five States. Bulletin. Washington DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Feld, B. 1991. ―Justice by geography: Urban, suburban, and rural variations in juvenile justice administration.‖ Journal of Criminal Law & Criminology 82:156–210. Feld, B. 1999. Bad Kids: Race and the Transformation of the Juvenile Court. New York, NY: Oxford University Press. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-30 Hamparian, D., and Leiber, M.J. 1997. ―Disproportionate confinement of minority juveniles in secure facilities: 1996 national report.‖ Prepared for the Office of Juvenile Justice and Delinquency Prevention. Champaign, IL: Community Research Associates. Harstone, E., and Richetelli, D. 2001. A Reassessment of Minority Overrepresentation in Connecticut’s Juvenile Justice System. Spectrum Associates Market Research. Hsia, H., Bridges, G., and McHale, R. 2004. Disproportionate Confinement—2002 Update. Summary. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Jackson, P. 1992. ―Minority group threat, social context, and policing,‖ in Social Threat and Social Control, A. Liska (ed.). Albany, NY: State University of New York Press, pp. 89–102. Lauritsen, J. 2005. ―Racial and ethnic differences in juvenile offending,‖ in Our Children, Their Children: Confronting Racial and Ethnic Differences in American Juvenile Justice. D. Hawkins and K. Kempf-Leonard (eds.). The John T. and Catherine MacArthur Foundation Research Network on Adolescent Development and Juvenile Justice. Chicago, IL: University of Chicago Press, pp 83–104. Leiber, M. 1994. ―A comparison of juvenile court outcomes for Native Americans, African Americans, and whites.‖ Justice Quarterly 11:257–279. Leiber, M. 2002. ―Disproportionate minority confinement (DMC) of youth: An analysis of efforts to address the issue.‖ Crime & Delinquency 48(1):3–45. Leiber, M. 2003. The Contexts of Juvenile Justice Decision Making: When Race Matters. Albany, NY: State University of New York Press. Leiber, M., and Fox, K. 2005. ―Race and the impact of detention on juvenile justice decision making.‖ Crime & Delinquency 51(4):470–497. Leiber, M., and Mack, K. 2003. ―The individual and joint effects of race, gender, and family status on juvenile justice decision-making.‖ Journal of Research in Crime & Delinquency 40(1):34–70. Leiber, M., and Stairs, J. 1999. ―Race, contexts, and the use of intake diversion.‖ Journal of Research in Crime and Delinquency 36:56–86. Nellis, A. 2005. Seven Steps To Develop and Evaluate Strategies To Reduce Disproportionate Minority Contact (DMC).Washington, DC: Juvenile Justice Evaluation Center. Pope, C.E., and Feyerherm, W. 1992. Minorities and the Juvenile Justice System: Full Report. Rockville, MD: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-31 Pope, C.E., and Leiber, M. 2005. ―Disproportionate minority contact (DMC): The federal initiative,‖ in Our Children, Their Children: Confronting Racial and Ethnic Differences in American Juvenile Justice. D. Hawkins and K. Kempf-Leonard (eds.). The John T. and Catherine MacArthur Foundation Research Network on Adolescent Development and Juvenile Justice. Chicago, IL: University of Chicago Press, pp. 351–389. Pope, C.E., Lovell, R., and Hsia, H. 2002. Synthesis of Disproportionate Minority Confinement (DMC) Literature (1989–1999). Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Pope, C.E., and Snyder, H. 2003. Race as a Factor in Juvenile Arrests. Bulletin. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Sampson, R. 1986. ―Effects of socioeconomic context on official reaction to juvenile delinquency.‖ American Sociological Review 51: 876–885. Sampson, R. 1987. ―Urban black violence: The effect of male joblessness and family disruption.‖ American Journal of Sociology 93:348–382. Sampson, R., and Laub, J. 1993. ―Structural variations in juvenile court processings: inequality, the underclass, and social control.‖ Law & Society Review 27:285–311. Smith, D. 1986. ―The neighborhood context of police behavior,‖ in Communities and Crime, A. Reiss and M. Tonry (ed.). Crime and Justice: A Review of Research, vol 8. Chicago: University of Chicago Press, pp. 313–342 Smith, D., and Visher, C. 1981. ―Street-level justice: Situational determinants of police arrest decisions.‖ Social Problems 29:167–177. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-32 Appendix A: Example—Youth Interview (Spectrum Associates) Date: Time Started: Time Ended: Youth ID: DOB: Gender: Male = 1 Female = 2 RACE: Black = 1 Hispanic = 2 White = 3 Town of Residence: Hi, my name is As you were told by the Long Lane staff when you were asked to take part in this study, I work for Spectrum Associates, which is a private research organization. I do not work for DCF, Long Lane, the police, the court, or the State of Connecticut. We are conducting a research project to learn more about the state’s juvenile justice system and how the police, juvenile courts and Long Lane treat juvenile offenders. We hope this study will help make the system better meet the needs of youth who come into contact with police, the courts, and Long Lane. As part of our study, we are talking with 30 kids at Long Lane. You have my word that everything you tell me will be kept confidential. That is, our report on the study might say something like one-half of the kids we spoke with felt they were helped by Long Lane and one-half of the kids said they were not. But, it will not name anybody. So, please answer the questions honestly. Nobody will see your answers. Also, please don’t discuss this interview (either the questions or your answers) with any other kids as we don’t want their answers to be influenced. I. GENERAL QUESTIONS (LONG LANE, POLICE, AND JUVENILE COURT) Long Lane 1. How long have you been at Long Lane? ____________ (RECORD MONTHS) (IF ASKED: this placement only) DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-33 2. At the time you were placed at Long Lane School (for this placement): Did you want to come to Long Lane, or .................. 1 (GO TO Q.5) Did you want to go to some other facility................ 2 (GO TO Q.3) NO PREFERENCE ............................................... 8 (GO TO Q.8) 3. Why didn’t you want to come to Long Lane? Where did you want to go instead? 4. Did you tell your lawyer that you didn’t want to go to Long Lane? Yes ...........................................................................1 No.............................................................................2 (GO TO Q. 8) 5. Why did you want to come to Long Lane? 6. Did you tell your lawyer that you wanted to come to Long Lane? Yes ...........................................................................1 No.............................................................................2 7. Which of the following was most important in your wanting to come to Long Lane? Your friends were here.............................................1 You thought the staff here could help you, or .........2 You thought you would need to serve less time here than if you were placed in another facility ...........3 NONE, WOULDN’T ANSWER ...........................9 8. How helpful has your stay at Long Lane School been so far? Would you say: Very helpful .............................................................5 Somewhat helpful ....................................................4 Neither helpful nor harmful .....................................3 Somewhat harmful, or ..............................................2 Very harmful ............................................................1 How has Long Lane (helped) (harmed) you? DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-34 9. Would you say that Long Lane staff: Treats all the kids the same, or.................................1 (GO TO Q.11) Treats some kids better than others ..........................2 (GO TO Q.10) DON’T KNOW.......................................................9 (GO TO Q.11) 10. Please tell me more about that. (PROBE: What types of kids are treated better and what types of kids worse? Do some types of staff treat kids better than others while others don’t? How so?) 11. Would you say that most of the staff at Long Lane usually: Have been fair in how they treat you, or ................ 1 (GO TO Q.13) Have not been fair in how they treat you ............... 2 (GO TO Q.12) DON’T KNOW..................................................... 9 (GO TO Q.13) 12. What have they done that was unfair to you? Which type of staff? Police I am now going to ask you a few questions about the police. 13. Thinking back to your experiences with the police, would you say that police officers: Treat all the kids they stop the same, or................. 1 (GO TO Q.15) Treat some kids better than others ......................... 2 (GO TO Q.14) DON’T KNOW..................................................... 9 (GO TO Q.15) 14. Please tell me more about that. What types of kids are treated better and what types of kids worse? 15. Would you say that the police officers you have dealt with usually: Were fair in how they treated you, or .................... 1 (GO TO Q.17) Were not fair in how they treated you ................... 2 (GO TO Q.16) DON’T KNOW..................................................... 9 (GO TO Q.17) DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-35 16. What have they done that was unfair to you? Juvenile Court 17. Let’s switch over to juvenile court. Would you say that people at the court like the juvenile probation officer, your attorney, the prosecutor and the judge: Treat all the kids the same, or ..................................1 (GO TO Q.18) Treat some kids better than others ...........................2 DON’T KNOW.......................................................9 (GO TO Q.18) Please tell me more about that. What types of kids are treated better and what types of kids worse? Are some types of court staff less fair than others? How so? 18. Would you say that people at the court were: Fair in how they treated you, or ...............................1 (GO TO Q.19) Not fair in how they treated you ..............................2 DON’T KNOW ........................................................9 (GO TO Q.19) Who was unfair to you? How? II. RACE-SPECIFIC QUESTIONS The next group of questions ask you about whether you think the juvenile justice system handles Black, Hispanic and White youth the same or differently. I will ask you questions about Long Lane first, then the police, and then juvenile court. Long Lane 19. Overall, would you say that Long Lane staff treat Black, Hispanic and White juveniles the same or differently? The same ................................................................ 1 (GO TO Q.25) Differently .............................................................. 2 (GO TO Q.20) DON’T KNOW..................................................... 3 (GO TO Q.25) DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-36 20. Who gets treated the best at Long Lane? Would you say residents that are: Black ........................................................................1 Hispanic, or ..............................................................2 White ........................................................................3 DON’T KNOW.......................................................9 (GO TO Q.22) 21. Why do you feel that way? (PROBE: Specifically, how do they get treated better? What type of staff treats them better?) 22. Who gets treated the worst by the Long Lane staff? Would you say residents that are: Black ........................................................................1 Hispanic, or ..............................................................2 White ........................................................................3 DON’T KNOW ......................................................9 (GO TO Q.24) 23. Why do you feel that way? (PROBE: Specifically, how do they get treated worse? What type of staff treats them worse?) 24a. (HISPANIC YOUTH) Do you think that the Hispanic staff treat you better, the same or worse than the other staff? Same.........................................................................1 (GO TO Q. 25) Better, or ..................................................................2 Worse .......................................................................3 How so? 24b. (BLACK YOUTH) Do you think that the Black staff treat you better, the same or worse than the other staff? Same.........................................................................1 (GO TO Q. 25) Better, or ..................................................................2 Worse .......................................................................3 How so? DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-37 24c. (WHITE YOUTH) Do you think that the White staff treat you better, the same or worse than the other staff? Same.........................................................................1 (GO TO Q. 25) Better, or ..................................................................2 Worse .......................................................................3 How so? Police Let’s switch to the police. 25. Overall, would you say that the police usually treat Black, Hispanic and White juveniles the same or differently? The same ................................................................ 1 (GO TO Q.30) Differently .............................................................. 2 (GO TO Q.26) DON’T KNOW..................................................... 9 (GO TO Q.30) 26. Who gets treated the best by the police? Would you say: Black ........................................................................1 Hispanic, or ..............................................................2 White juveniles ........................................................3 DON’T KNOW.......................................................9 (GO TO Q.28) 27. Why do you feel that way? (PROBE: Specifically, how do they get treated better?) 28. Who gets treated the worst by the police? Would you say: Black ........................................................................1 Hispanic, or ..............................................................2 White juveniles ........................................................3 DON’T KNOW.......................................................9 (GO TO Q.30) 29. Why do you feel that way? (PROBE: Specifically, how do they get treated worse?) DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-38 30. Overall, do you think that police officers are most likely to arrest: Black kids.................................................................1 Hispanic kids ............................................................2 White kids, or ...........................................................3 the police do not consider the kid’s race or ethnicity in their arrest decisions .............................................4 DON’T KNOW.......................................................9 Juvenile Court Let’s go back to people at the juvenile court, like the juvenile probation officer, your attorney, the prosecutor and the judge. 31. Overall, would you say that the juvenile court usually treats Black, Hispanic and White juveniles the same or differently? The same ................................................................ 1 (GO TO Q.36) Differently .............................................................. 2 (GO TO Q.32) DON’T KNOW..................................................... 3 (GO TO Q.36) 32. Who gets treated the best by the juvenile court? Would you say: Black ........................................................................1 Hispanic, or ..............................................................2 White juveniles ........................................................3 DON’T KNOW.......................................................9 (GO TO Q.34) 33. Specifically, how do they get treated better? What type of court people treat them better? 34. Who gets treated the worst by the juvenile court? Would you say: Black ........................................................................1 Hispanic, or ..............................................................2 White juveniles ........................................................3 DON’T KNOW.......................................................9 (GO TO Q.36) 35. Specifically, how do they get treated worse? What type of court people treat them worse? DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-39 36. Overall, do you think that the juvenile courts give more severe placements and punishments to: Black ........................................................................1 Hispanic, or ..............................................................2 White juveniles ........................................................3 DON’T KNOW.......................................................9 (GO TO Q.38) 37. In what ways do they get harder punishments? 38. Do you think that the juvenile court is most likely to place: Black kids at Long Lane ..........................................1 Hispanic kids at Long Lane .....................................2 White kids at Long Lane, or ....................................3 the juvenile court makes no distinction in placing Black, Hispanic and White kids at Long Lane .....4 DON’T KNOW.......................................................9 LAST QUESTION 39. One last question. What do you want to do when you leave Long Lane? I want to thank you very much for taking the time to talk with me. It was very helpful. As you know, to show our appreciation to the kids who are helping us out on this study, each of you is receiving $10. We have given $10 to the staff here at Long Lane to deposit in your account You should receive a deposit receipt within a few days. As I mentioned earlier, please do not talk to the other kids here about the questions on this interview or your answers as we don’t want their answers to be influenced. I would really appreciate it. DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-40 Appendix B: Example—Juvenile Court Coding Instrument for Case Records, Iowa, 2005 (1) Case Number (identification no.) (2) Study ID (sample no. coded case) (3) Date of proceedings (4) Age (5) Race 1) White 2) Black 3) Other 4) American Indian 5) Hispanic 6) Asian 8) No information (6) Gender 1) Male 2) Female 8) No information (7) Education (highest grade completed) School Performance (8) School Status 1) Attending 2) Attending but problems 3) Not attending 4) Other (GED, alternative high school) 8) No information 9) Not applicable Family (9) Number of siblings (brothers/sisters) 88 No information DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-41 Parents (10) Mother’s education (highest grade completed) 88 No information 99 Not applicable (11) Employment 1) Yes 2) No 8) No information 9) Not applicable (12) If employed, type of job 1) Clerical 2) Craftsman 3) Farmer 4) Laborer 5) Manager 6) Operative 7) Professional 8) Service 88) No information 99) Not applicable (13) Father’s education (highest grade completed) 88) No information 99) Not applicable (14) Employment 1) Yes 2) No 8) No information 9) Not applicable (15) If employed, type of job 1) Clerical 2) Craftsman 3) Farmer 4) Laborer 5) Manager 6) Operative 7) Professional 8) Service 88) No information 99) Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-42 (16) Receiving welfare 1) Yes 2) No 8) No information (17) Family status 1) Married 2) Living together 3) One family member present 8) No information 9) Not applicable (18) Who is taking care of the child 1) Mother 2) Father 3) Grandmother 4) Other (anything else) 5) Parents 6) Relatives (two present) 7) Foster parents 8) No information 9) Independent living (19) Family cooperative 1) Yes 2) No 8) No information 9) Not applicable Legal (20) Number of prior criminal offenses 88) No information (21) Previous criminal offense type (most serious) 1) Theft/unauthorized use 2) Burglary/breaking and entering 3) Disorderly conduct/jaywalking/obstruction/criminal mischief 4) Aggravated assault 5) Criminal trespassing 6) Receiving stolen property 7) Resisting arrest/escape 10) Robbery 11) Carrying a concealed weapon 12) Drug offense 15) Simple assault DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-43 16) 17) 18) 19) 21) 22) 23) 24) 25) 26) 27) 28) 29) 30) 31) 32) 33) 34) 35) 36) 37) 38) 77) 88) 99) (22) Rape/sexual assault Prostitution/soliciting Loitering Arson Forgery/fuffi Vandalism Murder Vehicular homicide Indecent exposure Tampering with a car Traffic offense Extortion Terrorism Cruelty to animals Fraudulent misrepresentation Failure to give assistance Kidnapping Explosives Alcohol Interference Delinquency by profanity DWLS Other No information Not applicable Previous criminal offense 1) Simple misdemeanor 2) Serious misdemeanor 3) Aggravated misdemeanor 4) Class A felony 5) Class B felony 6) Class C felony 7) Class D felony 8) No information 9) Not applicable Prior court supervision, prior supervision (type) Informal adjustment (type) (23) Just stay out of trouble 1) Yes 2) No 8) No information 9) Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-44 (24) Refer to other agency 1) Yes 2) No 8) No information 9) Not applicable (25) Community service 1) Yes 2) No 8) No information 9) Not applicable (26) Community service (informal or formal probation) 1) Park service/ maintenance service 2) People service 3) Both 4) Other 8) No information 9) Not applicable (27) If community service, how many hours 88 No information 99 Not applicable (28) If community service, did successfully complete 1) Yes 2) No 8) No information 9) Not applicable (29) Restitution 1) Yes 2) No 8) No information 9) Not applicable (30) Restitution ($ amount), if over $1,000 put 998 8) No information 9) Not applicable (31) If restitution , did successfully complete 1) Yes 2) No 8) No information 9) Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-45 (32) Volunteer supervision 1) Yes 2) No 8) No information 9) Not applicable (33) Length of volunteer supervision (days) 8888) No information 9999) Not applicable (34) Official supervision 1) Yes 2) No 8) No information 9) Not applicable (35) If official supervision, length in days 8888) No information 9999) Not applicable (36) Intensive supervision 1) Yes 2) No 8) No information 9) Not applicable (37) If intensive supervision, number of contacts per week 8) No information 9) Not applicable (38) If intensive supervision, method 1) Phone 2) In person 3) Both 8) No information 9) Not applicable (39) If intensive supervision, length in days 8888) No information 9999) Not applicable (40) If intensive supervision, did successfully complete 1) Yes 2) No DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-46 (41) Shoplifting program 1) Yes 2) No 8) No information 9) Not applicable (42) If shoplifting program, did successfully complete 1) Yes 2) No 8) No information 9) Not applicable (43) Education (tutoring) 1) Yes 2) No 8) No information 9) Not applicable (44) If tutoring program, did successfully complete 1) Yes 2) No 8) No information 9) Not applicable (45) Home of relatives (other than guardians) 1) Yes 2) No 8) No information 9) Not applicable (46) Foster home/group home 1) Yes 2) No 8) No information 9) Not applicable (47) Residential setting 1) Yes 2) No 8) No information 9) Not applicable (48) Training school 1) Yes 2) No 8) No information DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-47 9) Not applicable (49) Was child found to be delinquent (adjudicated) 1) Yes 2) No 8) No information 9) Not applicable (50) Waived to adult court 1) Yes 2) No 3) Waiver stipulation 8) No information 9) Not applicable (51) Length of time from current situation to last involvement in juvenile court (days) 8888) No information 9999) Not applicable (52) Number of current criminal charges 8) No information 9) Not applicable (53) List most serious current criminal offense 1) Theft/unauthorized use 2) Burglary/breaking and entering 3) Disorderly conduct/jaywalking/obstruction/criminal mischief 4) Aggravated assault 5) Criminal trespassing 6) Receiving stolen property 7) Resisting arrest/escape 10) Robbery 11) Carrying a concealed weapon 12) Drug offense 15) Simple assault 16) Rape/sexual assault 17) Prostitution/soliciting 18) Loitering 19) Arson 21) Forgery/fuffi 22) Vandalism 23) Murder 24) Vehicular homicide 25) Indecent exposure 26) Tampering with a car 27) Traffic offense DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-48 28) 29) 30) 31) 32) 33) 34) 35) 36) 37) 38) 77) 88) 99) Extortion Terrorism Cruelty to animals Fraudulent misrepresentation Failure to give assistance Kidnapping Explosives Alcohol Interference Delinquency by profanity DWLS Other No information Not applicable (54) Seriousness of criminal offense 1) Simple misdemeanor 2) Serious misdemeanor 3) Aggravated misdemeanor 4) Class A felony 5) Class B felony 6) Class C felony 7) Class D felony 8) No information 9) Not applicable (55) How many person were present when criminal offense took place (besides person that committed offense) 0) None 1) One 2) Two 3) Three 4) Four of more 8) No information 9) Not applicable (56) If committed new crime, was person still under court authority 1) Yes 2) No 7) No information 8) Not applicable (57) Violated condition of probation 1) Yes 2) No 8) No information DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-49 9) Not applicable Stages (58) Intake 1) Dismiss 2) Informal adjustment 3) Further court processing/petition 4) Dismiss and can’t locate 8) No information 9) Not applicable (59) At intake stage was child cooperative (always answer) 1) Yes 2) No 8) No information 9) Not applicable If Informal Adjustment (60) Held open, no other intervention/just stay out of trouble 1) Yes 2) No 8) No information 9) Not applicable (61) Refer to another agency 1) Yes 2) No 8) No information 9) Not applicable (62) Community service 1) Yes 2) No 8) No information 9) Not applicable (63) Community service (informal or formal probation) 1) Park service/maintenance service 2) People service 3) Both 4) Other 8) No information 9) Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-50 (64) If community service (how many hours) 88 No information 99 Not applicable (65) Restitution 1) Yes 2) No 8) No information 9) Not applicable (66) Restitution ($ amount), if over $1,000 put 998 8) No information 9) Not applicable (67) Volunteer supervision 1) Yes 2) No 8) No information 9) Not applicable (68) Length of volunteer supervision (days) 8889) No information 10000) Not applicable (69) Official supervision 1) Yes 2) No 8) No information 9) Not applicable (70) If official supervision length in days 8888) No information 9999) Not applicable (71) Intensive supervision 1) Yes 2) No 8) No information 9) Not applicable (72) If intensive supervision, number of contacts per week 8) No information 9) Not applicable (73) If intensive supervision method 1) Phone DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-51 2) 3) 8) 9) In person Both No information Not applicable (74) If intensive supervision, length in days 8888) No information 9999) Not applicable (75) Shoplifting program 1) Yes 2) No 8) No information 9) Not applicable (76) Education (tutoring) 1) Yes 2) No 8) No information 9) Not applicable (77) Petition (if youth gets this far, have to answer yes or no) 1) Yes 2) No 8) No information 9) Not applicable (78) Waived to adult court (if person reached this stage need to answer 1, 2, or 3) 1) Yes (if yes, stop now) 2) No 3) Waiver stipulation 8) No information 9) Not applicable (79) Initial appearance (if yes to petition) 1) Contested (fights case) 2) Uncontested (will not fight case) 8) No information 9) Not applicable (80) If yes to petition consent decree (proceedings suspended/open) 1) Yes 2) No (if no, go to variable 98) 8) No information 9) Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-52 If yes to consent decree, informal adjustment type (81) Just stay out of trouble 1) Yes 2) No 8) No information 9) Not applicable (82) Refer to another agency 1) Yes 2) No 8) No information 9) Not applicable (83) Community service 1) Yes 2) No 8) No information 9) Not applicable (84) Community service (informal or formal probation) 1) Park service/maintenance service 2) People service 3) Both 4) Other 8) No information 9) Not applicable (85) If community service, how many hours 888 No information 999 Not applicable (86) Restitution 1) Yes 2) No 8) No information 9) Not applicable (87) Restitution ($ amount), if over $1,000 put 998 8) No information 9) Not applicable (88) Volunteer supervision 1) Yes 2) No 8) No information DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-53 9) Not applicable (89) Length of volunteer supervision (days) 8888) No information 9999) Not applicable (90) Official supervision 1) Yes 2) No 8) No information 9) Not applicable (91) If official supervision, length in days 8888) No information 9999) Not applicable (92) Intensive supervision 1) Yes 2) No 8) No information 9) Not applicable (93) If intensive supervision, number of contacts per week 8) No information 9) Not applicable (94) If intensive supervision, method 1) Phone 2) In person 3) Both 8) No information 9) Not applicable (95) If intensive supervision, length in days 8888) No information 9999) Not applicable (96) Shoplifting program 1) Yes 2) No 8) No information 9) Not applicable (97) Education (tutoring) 1) Yes 2) No DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-54 8) No information 9) Not applicable (98) Adjudication (here if yes to petition and no to consent decree) 1) Yes 2) No 8) No information 9) Not applicable Disposition (99) Straight probation 1) Yes 2) No 8) No information 9) Not applicable (100) Refer to another agency 1) Yes 2) No 8) No information 9) Not applicable (101) Community service 1) Yes 2) No 8) No information 9) Not applicable (102) Community service (informal or formal probation) 1) Park service/maintenance service 2) People service 3) Both 4) Other 8) No information 9) Not applicable (103) If community service, how many hours 888 No information 999 Not applicable (104) Restitution 1) Yes 2) No 8) No information 9) Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-55 (105) Restitution ($ amount), if over $1,000 put 998 8) No information 9) Not applicable (106) Volunteer supervision 1) Yes 2) No 8) No information 9) Not applicable (107) Length of volunteer supervision (days) 8888) No information 9999) Not applicable (108) Official supervision 1) Yes 2) No 8) No information 9) Not applicable (109) If official supervision, length in days 8888) No information 9999) Not applicable (110) Intensive supervision 1) Yes 2) No 8) No information 9) Not applicable (111) If intensive supervision, number of contacts per week 8) No information 9) Not applicable (112) If intensive supervision, method 1) Phone 2) In person 3) Both 8) No information 9) Not applicable (113) If intensive supervision, length in days 8888) No information 9999) Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-56 (114) Shoplifting program 1) Yes 2) No 8) No information 9) Not applicable (115) Education (tutoring) 1) Yes 2) No 8) No information 9) Not applicable Placement (116) Home of relative (other than guardian) 1) Yes 2) No 8) No information 9) Not applicable (117) Group home/foster home 1) Yes 2) No 8) No information 9) Not applicable (118) Residential setting 1) Yes 2) No 8) No information 9) Not applicable (119) Training school/mental health institute 1) Yes 2) No 8) No information 9) Not applicable Detention (120) Stage intake 1) Yes 2) No 8) No information 9) Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-57 (121) Where 1) Home detention 2) Youth shelter 3) Detention facility 4) Combination of 1,2,3 5) Jail 8) No information 9) Not applicable (122) Initial appearance 1) Yes 2) No 8) No information 9) Not applicable (123) Where 1) Home detention 2) Youth shelter 3) Detention facility 4) Combination of 1,2,3 5) Jail 8) No information 9) Not applicable (124) Adjudication 1) Yes 2) No 8) No information 9) Not applicable (125) Where 1) Home detention 2) Youth shelter 3) Detention facility 4) Combination of 1,2,3 5) Jail 8) No information 9) Not applicable (126) Gender of probation officer 1) Male 2) Female (127) Length of detention for variable 120 (time is in hours) 888 No information 999 Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-58 (128) Length of detention for variable 122 (time is in hours) 888 No information 999 Not applicable (129) Length of detention for variable 124 (time is in hours) 888 No information 999 Not applicable (130) If legal counsel (type at any time) 1) Court appointed 2) Obtained 8) No information 9) Not applicable (131) For variable 77, if no petition why not 1) Out of court settlement 2) Not enough evidence 3) Moved away/ran away/joined services 8) No information 9) Not applicable (132) Site of coding 1) Black Hawk County 2) Polk County 3) Woodbury County 4) Scott County (133) If no adjudication, why not 1) Out of court settlement 2) Not enough evidence 8) No information 9) Not applicable (134) Race of probation officer 1) White 2) Black 3) Spanish 4) Other 8) No information (135) Prior disposition for previous most serious offense 1) Dismissed 2) Informal adjustment 3) Adjudication 4) Waived DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-59 5) Adjudication and placement (136) If answered 5 to variable 135, then answer 1) Home of relative (other than guardian) 2) Group home/foster home 3) Residential setting (youth shelter) 4) Training school 9) Not applicable (137) List second serious criminal offense type 1) Theft/unauthorized use 2) Burglary/breaking and entering 3) Disorderly conduct/jaywalking/obstruction/criminal mischief 4) Aggravated assault 5) Criminal trespassing 6) Receiving stolen property 7) Resisting arrest/escape 10) Robbery 11) Carrying a concealed weapon 12) Drug offense 15) Simple assault 16) Rape/sexual assault 17) Prostitution/soliciting 18) Loitering 19) Arson 21) Forgery/fuffi 22) Vandalism 23) Murder 24) Vehicular homicide 25) Indecent exposure 26) Tampering with a car 27) Traffic offense 28) Extortion 29) Terrorism 30) Cruelty to animals 31) Fraudulent misrepresentation 32) Failure to give assistance 33) Kidnapping 34) Explosives 35) Alcohol 36) Interference 37) Delinquency by profanity 38) DWLS 77) Other 88) No information 99) Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-60 (138) Seriousness of criminal offense 1) Simple misdemeanor 2) Serious misdemeanor 3) Aggravated misdemeanor 4) Class A felony 5) Class B felony 6) Class C felony 7) Class D felony 8) No information 9) Not applicable (139) With the current referral, was there any mention of drugs involved 1) Yes 2) No 3) Alcohol 4) Both drugs and alcohol 8) No information 9) Not applicable (140) With the current referral, was there any mention of a weapon involved (gun, knife, stick, club, pipe) 1) Yes 2) No 8) No information 9) Not applicable (141) Was an examination ordered 1) Mental health 2) Substance abuse (CADS) 3) Combination of 1 & 2 6) No 8) No information (142) If adjudicated, disposition sentenced to detention facility 1) Yes 2) No 9) Not applicable DMC Technical Assistance Manual, 4th Edition • Chapter 2: Assessment 2-61 Chapter 3: Preparation at the Local Level Mark Soler and Lisa Garry* There are several benefits of preparation prior to launching a local DMC reduction effort. These include establishing relationships with and among key local stakeholders, explaining the key goals of DMC reduction efforts, and identifying available data and research on DMC. After a brief overview of potential sources of financial support for local DMC reduction efforts, this chapter outlines a six-step preparation process: (1) establishing a steering committee, (2) identifying leadership, (3) reaching consensus, (4) conveying a sense of urgency, (5) setting priorities, and (6) organizing the work. The next section outlines basic tasks for the steering committee, based on lessons learned from the Baltimore City DMC Reduction Initiative. The chapter then briefly describes the W. Haywood Burns Institute approach to building community momentum for DMC reduction efforts and the Annie E. Casey Foundation’s Juvenile Detention Alternatives Initiative (JDAI) Framework. The chapter’s appendix presents the JDAI Core Strategies Matrix developed by the Burns Institute and the Casey Foundation. An Initial Issue: Financial Support An initial issue is how to support local efforts financially. The Office of Juvenile Justice and Delinquency Prevention (www.ojjdp.ncjrs.gov) awards Title II Formula Grant funds to the states for system improvement and programmatic efforts as well as technical assistance and other support for DMC reduction and other reforms. OJJDP’s Juvenile Accountability Block Grant and Title V Incentive Grants for Delinquency Prevention Grant funds provide additional financial support to states for juvenile justice reform, intervention, and prevention services in the area of juvenile delinquency. The work of the Burns Institute, described below, is supported in many communities with the Title II Formula Grant funds passed through states to communities. State and county agencies may also support DMC reduction efforts with state and/or local dollars. Local foundations, particularly community foundations, may be a source of funds. The Council on Foundations (www.cof.org) and other organizations provide information on locating and contacting community foundations. Several national foundations, such as the Annie E. Casey Foundation (www.aecf.org) and the John D. and Catherine T. MacArthur Foundation (www.macfound.org), support juvenile justice reform initiatives—e.g., the Casey Foundation’s Juvenile Detention Alternatives Initiative and the MacArthur Foundation’s Models for Change—that have DMC reduction components. * About the authors: Mark Soler is Executive Director at the Center for Children’s Law and Policy in Washington, DC. Lisa M. Garry is the Baltimore City DMC Coordinator in Baltimore, Maryland. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-1 Steps in Local Preparation Establishing a Steering Committee The local preparation effort should begin with the establishment of a steering committee. The committee should include key stakeholders in the juvenile justice system, such as the chief judge in the juvenile court, chief juvenile probation officer, senior prosecutor in the juvenile court, senior public defender, and police captain or lieutenant in charge of juvenile cases. It is important to have chiefs or senior officials on the committee to ensure that committee decisions are implemented. The committee should also include nontraditional stakeholders (i.e., persons with an interest in DMC from the perspective of program services rather than system policies and practices). These representatives are identified from community-level leadership, such as directors of community groups, civil rights organizations, child advocates, parent advocates, and others in the community who are concerned with DMC issues. The committee should also include young people or representatives of young people who are or have been in the system, to anchor the work to the population most affected. For example, the composition of DMC workgroup members in Cook County, Illinois (South Suburbs), is exemplary in its engagement of nontraditional stakeholders, such as community-based service providers, grassroots leaders, and community residents. A community-inclusive steering composition is advantageous to the development and expansion of community-based services and programs as detention alternatives and supports for youth and families within the least restrictive settings. Identifying Leadership DMC is a difficult issue to address, so it is critical to identify strong leadership for the steering committee. This usually means the chief judge in the juvenile court or chief juvenile probation officer, since they will most likely control policy changes that may be necessary to implement DMC reduction. Leadership by high-level administrators of the judicial and/or probation system also conveys and lends validity to the message that DMC reduction is an important issue within system agencies. Reaching Consensus The first task of the steering committee is to reach consensus on the goals of the DMC effort and the responsibilities of participants in the effort. If the local effort will conduct its activities in conjunction with a state DMC effort, state leaders should use the preparatory phase to introduce DMC as a shared vision and responsibility of both state and local entities. This approach will incorporate the concerns, insights, and innovations of local communities in the broader context of statewide DMC activities. At the early stage, this foundational step in preparation at the local level will be based more on dialog than on data research. Dialog among all participants on the committee will help establish consensus, prioritize problematic decision points, and create a sense of urgency among local leaders and stakeholder groups. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-2 It should not be surprising if members of the steering committee come to the effort with different expectations and understandings of the goals. The preparatory dialog will reveal the shared agendas and differences of position among stakeholders concerning variables contributing most to DMC. Leadership of the steering committee should anticipate and plan for differences among stakeholders in goals, priorities, or strategies. Creative tension helps the collaborative build consensus and trust. Conveying a Sense of Urgency At the same time, however, “urgency” speaks to the level of importance that key stakeholder agencies and opinion leaders assign to DMC. DMC reduction will not occur as an afterthought or a sidelight to other initiatives: the leadership of the steering committee must convey a sense of urgency about the issue. Do the stakeholders consider DMC to be an important issue? Do they believe that fair and equitable treatment of minority youth will improve their outcomes and, therefore, reduce concerns about public safety? What changes in policy are they prepared to consider to reduce DMC? What resources can they bring to the effort? A sense of urgency is a reflection of a search for active solutions to ensure a fair and equitable system. Setting Priorities The steering committee should lead the effort to use data that have been gathered to prioritize system decision points and develop targeted interventions. DMC may occur at any key decision point in the system—arrest, referral to juvenile court, diversion, secure detention, petition (charges filed), delinquent findings, probation placement, secure confinement, and transfer to adult court. There is value and wisdom to addressing one decision point at a time. The DMC-related processes of assessment, intervention, and evaluation are often time-consuming and grueling exercises of determination and will on the part of key stakeholders. Agreement about which decision points are the largest contributors to DMC will vary by jurisdiction. Although setting priorities should be based on data, levels of collaboration, cooperation, community readiness for change, and availability of resources in some parts of the system rather than others may also drive priorities. Consensus, urgency, and priority are keystones of DMC preparation and planning. The process may take months, or even longer, to accomplish. Local communities should not move forward until these prerequisites are met. By the same token, when these prerequisites are met, the community should acknowledge and celebrate them as real successes in the DMC planning process. Organizing the Work The next step in local-level DMC preparation is defining success. How do local stakeholders define success based on their own perspective of need and their collective experience in local juvenile justice work? Although the goal is to reduce overrepresentation at particular points in the system, there are many ways to move toward that goal, such as adoption of an objective risk assessment instrument to control detention DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-3 admissions, development of new community-based programs and services as alternatives to secure detention, modification of police procedures to better track contacts with minority youth, adoption of policies to reduce transfer to adult criminal court, and reduction in post-dispositional placements in secure confinement. As with the consensusbuilding process, reaching agreement among stakeholders on the definition(s) of success may be a struggle, but it is an important one. Basic Tasks for the Steering Committee: Examples From the Baltimore City DMC Reduction Initiative The lessons learned from the Baltimore City DMC Reduction Initiative during its early planning and preparation activities led to the identification of several basic tasks for the steering committee. Articulating Local DMC Goals and Objectives The core goal and objectives of the DMC Initiative in Baltimore City are to reduce overrepresentation of youth of color at the secure detention decision point by altering conditions, policies, and practices that contribute to their overrepresentation in the system and by influencing the culture and values of the system and community toward least restrictive and community-based sanctions. The objectives are clearly articulated within a set of deliverables that detail specific tasks, task assignments, and proposed timelines of the governing body. The deliverables then serve as a tangible workplan to which the governing body, its staff, and consultants hold themselves accountable and by which they measure progress toward policy and practice reforms and, ultimately, DMC reductions. Additionally, routine dissemination of the workplan to partners and community interests helps to communicate the governing body’s work, findings, and progress. Mediating Discussions To Acknowledge and Respect Differences of Opinion Without Compromising Progress Initial discussions about the key system decision points at which disproportionality was greatest unveiled differences of opinion and perspectives among key stakeholders within the governing body. Although some were of the opinion that disparate law enforcement practices resulted in the high numbers of juveniles being presented at intake, others were convinced that disparate system policies and practices following juvenile arrest were the issue and priority. Opinions differed even more when the jurisdiction began experiencing an increase in the number of difficult-to-place, post-adjudicated juveniles who were in confinement, and thereby increased disproportionality even further. Through its mediation of the DMC discussions and debates, the governing body facilitated a shift from disagreements on the extent of disproportionality at various decision points to a mutually agreeable focus on decision points where DMC reduction victories could be gained more readily and rapidly. City stakeholders agreed that the governing body’s influence over policies and practices at the doors of detention was greater than its influence over the public attitudes and perceptions that were driving law enforcement DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-4 practices. Ultimately, DMC activities in Baltimore City were better served through this shift. Keeping the DMC Agenda at the Forefront of All Juvenile Justice Activities in the Jurisdiction The task of keeping the issue of DMC front and center in all discussions and activities in juvenile justice is strongly tied to the principle of creating and maintaining a sense of local urgency on the issue. Early in the planning process, Baltimore City’s DMC steering group identified all other local planning and strategy initiatives that had responsibility for shaping and driving outcomes for youth of color who were either involved, or at-risk for involvement, in the juvenile justice system. The intention was to establish racial disparity as an agenda item for the other initiatives. This was accomplished through the active involvement of DMC group members in the other initiatives to apply a “racial lens” to their planning and strategies. For example, working partnerships were formed between the DMC governing body and the other committees within Baltimore City’s detention reform initiative to assess and redevelop detention alternatives, expedite case processing, and study the nexus between juvenile justice and child welfare. Through the staff resource of a full-time DMC coordinator in Baltimore City, the DMC governing body has taken the lead on detention utilization studies and community capacity building to increase community-based resources that serve juveniles in less restrictive settings. Moving Ahead With Intervention Strategies Jurisdictions must be careful not to allow extended dialog and analyses of assessments to immobilize them so that they fail to move forward with active reduction strategies and interventions. This “analysis paralysis” is a common experience in jurisdictions that lack measurable goals and objectives and a detailed work plan to accomplish those goals and objectives. While the State of Maryland had been examining the issue of DMC through various data analyses, reports, and conferences for several years, local communities lacked viable strategies toward attainable goals. The development of its DMC workplan and the decision to hire a full-time DMC coordinator enabled the Baltimore City DMC collaborative to move beyond abstract discussion and to finally focus on the development and implementation of intervention strategies that address day-to-day agency practices and increase community-based resources and detention alternatives. The Burns Institute Process: An Example of Building Community Momentum The W. Haywood Burns Institute is a national organization working with local jurisdictions to reduce the overrepresentation of minority youth in their juvenile justice systems using a data-driven, consensus-based process model that engages both traditional and nontraditional stakeholders. The Institute has a range of services to assist jurisdictions in reducing racial disparities in the juvenile justice system, from consulting to intensive engagement. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-5 The Institute model requires the active commitment and participation of the key traditional and nontraditional stakeholders in the juvenile justice system in each site. This includes judges, prosecutors, public defenders, police, probation, school officials, political leaders, service providers, and community groups. The Institute leads these stakeholders through a process that focuses specifically and intentionally on reducing disproportionate minority contact. Without a committed and intentional approach to reducing DMC, jurisdictions often lose momentum because of changes in stakeholders, inconsistent approaches, and short attention spans. To ensure that sites stay focused, the Institute has developed a manual and workbook to guide them through the process. The Institute’s model of Intensive Site Engagement (ISE) calls for stakeholders to develop a workplan and to meet monthly to move it forward. The model requires each site to hire a full-time local site coordinator to lead the process. In addition, an Institute staff member is assigned to each site, attends all local meetings, and is available for the local site coordinator and stakeholders to contact for technical assistance and guidance. Site coordinators from each of the Institute sites meet twice yearly to compare best practices and are available year-round for consultation with one another. The Institute also provides a Readiness Assessment Consultation (RAC) to enable a site to assess its readiness to address DMC. In 2006, the Institute is working in Baltimore, Maryland; Louisville, Kentucky; Pima County, Arizona; Cook County (Lawndale), Illinois; Cook County (South Suburbs), Illinois; St. Clair County, Illinois; Peoria County, Illinois; San Francisco, California; San Jose, California; and Seattle, Washington. The Juvenile Detention Alternatives Initiative Framework: An Example of Changing the System To demonstrate that jurisdictions can establish more effective and efficient systems to accomplish the purposes of juvenile detention, the Annie E. Casey Foundation established the Juvenile Detention Alternatives Initiative (JDAI) in 1992. The objectives of JDAI are to reduce the number of children unnecessarily or inappropriately detained, to minimize the number of youth who fail to appear in court or reoffend pending adjudication, to redirect public funds toward successful reform strategies, and to improve conditions of confinement. As minority youth are consistently overrepresented in juvenile detention facilities, the elimination of disparate treatment and decisionmaking for these youth is a core JDAI strategy. The other core strategies are collaboration, reliance on data, objective admissions screening, alternatives to secure detention, expedited case processing, strategies for special detention cases, and rigorous facility inspections. In May 2005, the Foundation developed system assessment frameworks for sites participating in JDAI. The frameworks provide a structured, content-specific way for local jurisdictions to examine their detention policies, practices, and programs and to gauge progress in the JDAI’s eight core strategies for detention reform. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-6 As part of this process, the Foundation and the Burns Institute developed a framework for viewing the JDAI core strategies “through a racial lens.” This JDAI Core Strategies Matrix, which can be found in this chapter’s appendix, can help jurisdictions to examine detention policies, practices, and programs and can also serve as a workplan to help jurisdictions accomplish the following: Structuring a diverse DMC collaborative that has authority, common agendas, shared responsibilities, and community inclusiveness. Relying on both quantitative and qualitative data to assess and reform detention utilization and to determine the most effective allocation and placement of community resources. Eliminating intentional and unintentional biases in detention admission screening. Developing culturally and racially competent detention alternatives. Equalizing case processing at all system decision points to minimize delay and ensure equity in the judicial process. Addressing disparate handling of special detention cases (e.g., writs, warrants, violation of probation). Ensuring that confinement conditions are acceptable and competent. The use of the JDAI Core Strategies Matrix to meticulously examine system practices holds a high value for jurisdictions working within the JDAI core strategy frameworks. Equally important, non-JDAI jurisdictions that are planning or expanding a DMC initiative may also find the DMC framework useful for creating a workplan around the system components most vital to ensuring a racially equitable and competent juvenile justice system. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-7 Appendix: JDAI Core Strategies Matrix Collaboration Issue What We’re Looking For, Why This Is Important Authority Is there an official imprimatur that reducing racial disparities is an explicit responsibility of the JDAI collaborative? Composition Does the collaborative reflect the diversity of the kids and families involved in your juvenile justice system? Review, Observe, and Interview Major Findings Best Practices, Recommendations Do we have the decisionmakers sitting at the table with the appropriate community representatives? Does the collaborative effort include representatives of the impacted neighborhoods of color? Are civil rights advocates at the table? Are community-based service providers at the table? Organizing the work The intentionality and infusion of the racial lens needs to be driven in unison with decisionmakers and communities of color. Is the current configuration, e.g., work group, ad hoc committee, working? Is each subcommittee held accountable for contributions to reducing racial disparities? Common challenges are ―work groups‖ working in a silo, which are expected to ―fix‖ the problem. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-8 Collaboration (continued) Issue Creating a safe place What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Are discussions regarding disproportionality undertaken with respect and tolerance? Are the discussions mainly finger-pointing sessions? Are deliberations based on facts and supported by data or impressions? Have efforts been made to ensure equal and full participation in the discussions and deliberations? Forging a common agenda Do members of the collaborative, including work group members if relevant, have a common understanding of, and embrace, the same agenda: detention as the entry point to the reduction of racial disparities? Members of the collaborative understand that the work entails changing policies and practices under the control of their juvenile justice system. Members of the collaborative reach a consensus on the use of detention in their jurisdiction. A shared value that pretrial detention should not be used as either punishment or treatment. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-9 Reliance on Data What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-10 Issue Disaggregating data by race and ethnicity Baseline data of youth ages 10– 17, disaggregated by race, ethnicity, gender, and geography, should be collected by the foundation to identify the disproportionality and to commence the discussion. Has the collaborative compared the percentage of youth of color in the juvenile justice system with the percentage of minorities in the general youth population? All ensuing data collection— e.g., admissions by reason, risk assessment instrument (RAI) screening, RAI overrides, length of stay (LOS), average daily population, use of alternatives to detention (ADP)—should be disaggregated by race/ethnicity/gender/ geography. Routine management reports present basic utilization statistics by race/ethnicity/ gender to enable stakeholders to identify disparities and to assess trends and change policies and practices. Detention utilization study One of the first steps in planning for reform is to document how detention is currently used through careful data collection and analysis. A thorough description of recent trends and current practices in detention utilization provides the foundation for the problem identification and analysis, as well as the subsequent development of change strategies. The detention utilization study should provide the collaborative with a quantitative picture of how detention use varies for different categories of youth. Reliance on Data (continued) What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-11 Issue Geocoding and community mapping Identify the target area(s), that is the geographic area(s) contributing the highest number of kids in detention. Map the community assets, including community-based organizations currently providing services to youth and their families in the target neighborhoods. Identifying the target neighborhoods and mapping community-based services will assist in informing strategies for effective and efficient alternatives to detention. Routine management reports Using data to monitor progress toward reducing racial disparities and disproportionate minority confinement. The JDAI quarterly reports are an example of fundamental management reports. As the data from the reports raise questions, further data queries should be developed to dig deeper and acquire clarity. Qualitative analysis Digging deeper generally leads to going ―behind the data‖ to look at individual policies and practices to clarify reasons behind the statistics. What are the practices or policies contributing to the statistical disproportionality? Comprehensive annual analysis of racial disparities Is the community informed of the state of racial disparities/DMC on an annual basis in your jurisdiction? Annual reports developed by the system partners help keep eyes on the prize and promote accountability and transparency. Eliminating Bias in Detention Admission Screening Issue Objective criteria and instruments What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Collaborative development of a race- and gender-neutral objective detention admission screening instrument based on risk. The admission screening instrument should be scrutinized to ensure it is eliminating opportunities for disparate decisions. We’re looking to control the front gates in an objective and equitable manner. Bias in statutory criteria Examine your jurisdiction’s statutory detention criteria for any bias and determine whether the criteria are mandatory or discretionary. This examination should include which factors must be taken into consideration to detain and consider collaborative efforts for developing local detention criteria to reduce the number of kids of color brought to the front gate. Testing for unintended bias from screening tools Assess the admission screening instruments’ impact on kids of color. The screening scores should be consistently monitored for disparate application and nuances that can reveal unintended biases. The risk-based detention screening instrument should not add unfair risk points for kids of color. For example: points for being a “gang associate” tend to penalize kids for living in the disinvested neighborhoods where youth of color and their families have long been segregated; limiting release to parent(s) only and not considering extended family members or a responsible adult. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-12 Eliminating Bias in Detention Admission Screening (continued) Issue Multilingual, multicultural intake staff What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Eliminating barriers to returning a youth home. Intake staff that speak and understand the language spoken by the youth and families to facilitate the release of youth in a more timely fashion. Implementing intake procedures 24/7. Intake staff who value, recognize, and appreciate an individual’s race/culture and its significance and role in the lives of youth and families. Quality controls The development of protocols for the implementation of the admission screening instrument. Leadership providing swift and consistent oversight for compliance with the protocols and with the application and scoring of the admissions screening instrument, as well as monitoring overrides. Monitoring for consistency and equity in the application of the admission screening instrument by intake staff. Use of overrides Collecting data to determine if kids of color are being overridden in a disparate manner. What are the override criteria? What are the reasons for the overrides? Do patterns emerge in the criteria invoked for the override relative to youth of color? For instance, criteria that allow for an override if ―parent, guardian or responsible relative refuses to take custody.‖ Collecting this information will assist in informing strategies for changes in policies and practices relative to the particular override criteria. Monitoring for consistency and equity in the application of the DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-13 Eliminating Bias in Detention Admission Screening (continued) Issue What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations admission screening instrument by intake staff. If one worker, for example, is overriding the RAI at a significantly higher rate than other workers or at a significantly higher rate for kids of color, the pattern should be identified and addressed immediately. Automatic detention cases Collecting and analyzing the data to determine whether youth of color fall disproportionately into this category. Conducting a qualitative analysis to determine if changes in policies are necessary; e.g., warrants, and policies that will promote detention alternatives. Monitoring the data to ensure that the automatic detention category is not being disparately applied to youth of color. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-14 Culturally and Racially Competent Alternatives to Detention Issue Target populations What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations The ATD should serve kids who otherwise would be detained. Is the target population based on risk level, e.g., RAI score, or status, e.g., violations of probation (VOPs)? Collect and monitor data informing which kids are being referred to ATD. Are youth of color treated disparately in referrals to ATD? Conduct a qualitative analysis of the target population to determine the needed intervention necessary to inform responsive ATD. Program design Programs that respond to the needs and circumstances of youth of color. Good ATD programs are relationship based, not technology based. Successful ATD programs include partnerships with community-based organizations to provide the appropriate culturally and racially relevant and responsive interventions. Pre-adjudication ATD programs are intended to ensure court appearance and minimize rearrest risk. Post-adjudication programs will typically feature more treatment interventions (e.g., counseling) and sanctions. The ATD is limited in duration of purpose—don’t create a purgatory that will set kids up for failure. Does supervision include face-toface contact? Is the level of supervision based on risk? ATD that offer more than one level of alternative? Collect data on entry to and exits from the programs. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-15 Culturally and Racially Competent Alternatives to Detention (continued) Issue What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Collect data on the rate of referrals by RAI scores to Electronic Monitoring Programs (EMP). Is there an overreliance on the use of EMP with kids of color? Collect data to monitor terminations/failures. Is there a high failure rate of kids of color by a particular program? Conduct a qualitative analysis to determine reasons for failure to inform needed program changes or enhancement and development of ATD. Does the program have a ―no reject‖ policy? Service providers Community-based organizations that provide culturally or racially relevant and appropriate services. Do current service providers have the capacity and are they appropriate, to work with kids of color? Location and access Are programs located in the neighborhoods where relevant youth and families reside? Programs that are accessible to the youth, e.g., getting to the program, isn’t going to pose a hazard to the youth’s safety. Accessing and partnering with community-based organizations that are in the neighborhoods already working with, and touching on, the lives of youth of color and their families. Language and culture Program staff that have the skills set and values to meet the youth’s language and cultural needs. Eliminate barriers, posed by staff’s language limitations that hamper the youth’s success on the ATD. Principles that acknowledge that ―culturally responsive‖ also includes understanding and tolerance of ―youth culture.‖ DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-16 Culturally and Racially Competent Alternatives to Detention (continued) Issue Staffing and services What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Staff who relate, and are responsive to, the needs and circumstances of youth of color and their families. Staff who appreciate the culture of youth and who want to work with youth and help them succeed. Staff who have an awareness and understanding of the dynamics of the neighborhoods where youth and their families reside. Staff who look like, and live in or around the same neighborhoods as, the youth of color and their families. Activities and services that value and honor the race/ethnicity/culture of the youth and their families. Are activities and services designed as a ―one size fits all,‖ or designed to respond to individual needs? Are services designed to build on the strengths of the youth and their families? Are there cultural and relevant racial competency trainings for staff? Is the program’s physical environment reflective of the clientele’s race/ethnicity/culture? Resultsbased accountability Assess current ATD for effectiveness, efficiency, and responsiveness. Does the ATD affect bed displacement of kids of color? Whether the ATD is provided for solely by system folks or in partnership with communitybased organizations, results/outcomes must be established and monitored. Measurable results for preadjudication ATD include minimizing re-arrest and failure to appear (FTA). Contractual agreements between DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-17 Culturally and Racially Competent Alternatives to Detention (continued) Issue What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations system agencies and communitybased organizations that specify expected results and define success. Agreed-upon data collection and methodology, e.g., FTA, rearrest, successful completion, LOS. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-18 Equalizing Case Processing Issue Analysis of decision points What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Each of the juvenile justice system partners map the decisionmaking points relevant to their discipline, that touch upon the children’s lives as they ―process‖ through the system. Collect data relative to each of the decision points and analyze for racial disparities. (Some examples of specific decision points: the District Attorney measuring all filing decisions and processes by race/ethnicity/gender; the Public Defender measuring requests for continuances, e.g., reasons, frequency, by race/ethnicity/gender; the probation department’s recommending or opposing ATD.) Monitor decision point data for trends. Monitor for disparities in arresting charge vs. actual charge filed vs. resulting adjudication. Examining ―race effects‖ throughout case processing Develop an initial mapping of the jurisdiction’s case processing, including time frames for each of the case processing ―steps.‖ Collect the data to determine any disparate outcomes based on race/ethnicity/gender. Use the data to inform changes in policies and practices. Minimizing unnecessary delay Critical examination of case processing with an eye to reveal unnecessary delay for kids of color which contribute to longer lengths of stay in detention. Efficient court and placement system with short lengths of stay in detention. Measure length of stay by race/ethnicity/gender to inform changes in policies and practices. Dedicated staff/expeditor assigned to monitor the status of detained youth and identify any disparities. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-19 Equalizing Case Processing (continued) Issue What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Examine for and reduce delays that can result in pushing kids into detention, e.g., delays leading to FTA, resulting in the issuance of a warrant in turn resulting in detention. Ensuring equal access and due process The administration of justice that is responsive to the circumstances of youth of color and their families. Public transportation conveniently located in the impacted communities and in proximity to court services. Multilingual court personnel, including courtroom interpreters, to minimize barriers for youth of color and their families. Defense counsel knowledgeable of, and experienced in, juvenile law. Defense counsel who understand the circumstances of youth of color. Sufficient number of public defenders to support the caseload. A fair and honest rate of pay for appointed counsel. Ensure that youth are represented by counsel at every stage of the proceedings. Monitor for waivers of counsel by youth, and eliminate such policies and practices. Monitor for disparities in adjudicatory outcomes for kids of color. Consistency and equity Ensuring that kids who are similarly situated are treated in an equitable manner from courtroom to courtroom. A determined and intentional commitment to equitable and consistent treatment of kids of color that reflect the principles of JDAI. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-20 Race and “Special” Detention Cases Issue Data analysis What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Are there disparities in case status by race/ethnicity? Often, kids of color are more likely to have warrants, be charged with VOPs, etc. If disparities are found statistically, are there policy or practice reasons for them? Do youth of color have longer lengths of stay? This is especially likely in the pending placement group; what causes these differences? What are the reasons for warrants, VOPs, and delayed placements? For example, are most warrants for FTA? Most VOPs for positive drug tests? Warrant reduction strategies Are FTA rates high, at first appearance? High FTA rates often include many unintentional absences. Is there a court notification system? FTA can be reduced simply by reinforcing notification of court dates. (Similar gains can be made regarding VOPs by decreasing likelihood that youth miss visits with probation.) Are warrant cases screened with RAI? Many warrant cases pose low public safety risks (after all, the kid was not detained in the first instance), but ―automatic‖ detention policies often mean that risk is never assessed. Is there a differential warrant policy? Do judges indicate whether individual warrants must be detained, or is there simply a blanket policy. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-21 Race and “Special” Detention Cases (continued) Issue Violations of probation What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations How are conditions of probation established; are they too numerous? If there are lots of unnecessary conditions, it is easy to violate youth. Are detained VOP cases equally distributed across staff? Differences between probation officers in use of detention for VOPs indicates that the underlying policies do not structure decisions or control for individual idiosyncrasies. Are graduated sanctions available as alternatives? Systems ought to have options short of detention that are based upon seriousness of the violation, etc. Is there court policy requiring court intervention for technical violations? Can the department handle routine violations administratively? What do we know about the quality of probation supervision generally? In some systems, for example, high caseloads typically mean ineffective case management which, in turn, leads to youth ―failures,‖ negative results that might be avoided through improved supervision. Pending placement cases Do placement options reflect diversity of client population? Are they culturally competent? If placements are not available for non-English speaking youth, for example, they will languish as staff look for a program that can communicate with the clients. Similarly, culturally incompetent programs will surely have higher failure rates as youth abscond or get frustrated and alienated. What are program policies regarding rejection of referrals or termination of clients? Contract conditions can reduce pending placement cases simply by DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-22 Race and “Special” Detention Cases (continued) Issue What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations ensuring that referred clients are accepted or by limiting the numbers of youth getting recycled because of unnecessary ejection from programs. Is there effective dispositional planning? Many places have long pending placement lists because they are uncreative or rigid in their approach to crafting individualized dispositions. Does the placement process delay release? If placement paperwork is not prepared in a timely way, or only sent to one program at a time, days will be wasted. Are there intensive home-based services available? Overreliance on out-of-home placements is often the result of limited nonresidential program options. Effectiveness of counsel Does counsel take steps to reduce likelihood of warrants, or to clear old warrants? Defense lawyers can reduce clients’ jeopardy of detention for FTA simply by taking steps to ensure their clients appear in court as scheduled. Does counsel have capacity to provide effective dispositional advocacy? In many places, the defense fails to offer the court nonresidential alternatives that could minimize pending placement backlogs. Similarly, failure to advocate for appropriate conditions of probation increases odds that violations will occur. Does counsel challenge VOPs? Detention use in VOP cases can be avoided if counsel presents a case against the allegations or the detention. Does counsel review ―special‖ detention cases internally or participate in system case reviews? Placement cases languish absent prodding to expedite arrangements. Warrants may be cleared and set the stage for renewed applications for DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-23 Race and “Special” Detention Cases (continued) Issue What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations release. These developments are more likely if there is a structured review process, either in counsel’s office or by the system generally. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-24 Conditions of Confinement Issue Staff competencies What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Do staff reflect the racial/ethnic composition of detained youth? Detainees are more likely to be able to communicate, feel safe, etc. if the staff reflect them. Similarly, staff biases are less likely to manifest themselves when staff are more diverse. Of particular importance, do nonEnglish speaking youth have staff with whom they can communicate? Do staff routinely receive diversity training? If we want staff to do their jobs in culturally competent ways, they may need training and consistent reinforcement. Are staff efforts to perform work in culturally sensitive and competent ways routinely reinforced? If we want staff to act in certain ways, or reflect certain values, management should create incentives for such behavior (or disincentives for its opposite). Facility programming Does facility offer culturally appropriate programs? Failure to celebrate relevant holidays, or to give equal attention to various racial or ethnic groups, will create an us/them environment. Are there faith-related resources that reflect diversity of religion? Whether for formal services or individual counseling, the diversity of faiths ought to be accommodated by the detention programs. Health and hygiene supplies Are products familiar to different racial/ethnic groups available? Differences across groups need to be accommodated lest minority groups be forced to use ―foreign‖ supplies. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-25 Conditions of Confinement (continued) Issue Access and visitation What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Are youth able to see their lawyers? Detained youth should be able to contact their lawyers by phone and there must be private space for consultations. Can youth call home? Facilities need to provide opportunities for youth to call home (collect) in order to maintain contact. Are visitation policies sufficient to maximize likelihood of contact between youth and family members? If visitation days and times are restrictive, kids are less likely to maintain effective contact with family and will be more likely to be depressed, etc. Food Does food service reflect diversity of detainees? Discipline, restrictions, and restraints Is the use of various disciplinary actions, including loss of privileges, room restrictions and placement in restraints equal across racial and ethnic groups? Is there a sufficiently detailed and observed set of due process protections in place? Is there an accessible, genuine grievance process available to detained youth? Do youth of color experience more incident write-ups or infractions? Overall climate Do youth of color feel safe in facility? Do youth of color feel respected in facility? Does housing tend to segregate youth by race/ethnicity? Are there tensions and hostilities across racial and ethnic groups? DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-26 Specific Strategies To Reduce Racial Disparities Issue Formulate a vision and goals What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Determined leadership! No specific strategy seems more important than the tangible commitment of system leaders to racial justice. System leaders make reduction of racial disparities in detention their priority and use both their formal and informal authority to focus agency strategies to reduce DMC. System leaders engaging staff in the development of a vision establishing the reduction of racial disparities fundamental work. Establishing measurable objectives that are within the control of each partner’s respective system/discipline. Establish formal structures to keep eyes on the prize Intentionality! Intentionality! Intentionality! Keeping all eyes on the prize requires intentionality. Build ties to communities of color Successful efforts to reduce racial disparities and DMC include communities of color at the table. This isn’t an issue that white people are going to solve on their own without the unique perspectives of people of color who are impacted by the polices and practices. Ensure that technical changes are transformed to ―adaptive changes.‖ Establish the organizational infrastructure to sustain system changes. For example: developing and implementing an RAI is a technical change. However, if the infrastructure (e.g., training, protocols, monitoring the data, quality control, etc.) is not developed, addressed, and adhered to, then the change has not been ―adapted‖; the change will slip into the status quo. Relinquishing power to meaningfully engage and promote the unique perspectives and lens brought by people of color. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-27 Specific Strategies To Reduce Racial Disparities (continued) Issue What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations Promoting system accountability and transparency. Building allies with communities of color to effectively reduce racial disparities and DMC. Diversify system workforce Establish measurable goals to establish a workforce reflecting the demographics of the jurisdiction’s children and families. A multicultural workforce of men and women whose values reflect the principles of detention reform and the reduction of racial disparities and DMC. Key positions have bi/multilingual staff. Conduct cultural and relevant racial competency training Ongoing system training to develop staff cultural and relevant racial competencies. Create new or utilize current capacities in key neighborhoods Engaging nontraditional partners/community-based organizations (CBOs) who are already working with youth of color and families in their neighborhoods. Implementation of cultural and racial competence standards by all of the juvenile justice departments. Commitment to, and assisting in, developing the capacities of CBOs to partner in efforts to reduce unnecessary and inappropriate detention, including disproportionality. Informed by the quantitative and qualitative data developed relative to assessing ATD, create ATD in key neighborhoods where kids of color and their families reside. Develop objective tools for key decision points Key decisions, not just the decision to detain, are supported by objective tools. These decision points should be DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-28 Specific Strategies To Reduce Racial Disparities (continued) Issue What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations identified from the mapping of the decision points of all system partners; ―peeling the onion‖ at each point to determine how the decision impacts kids of color. Tools defined by solutions to the disparities uncovered at any decision point. Examples of objective tools include: detention criteria developed in partnership with law enforcement; customer surveys that identify service barriers; criteria without racial bias for assignment to intensive caseloads; criteria for removal from intensive caseloads; partnering with culturally and racially relevant CBOs to improve success rates of kids in pre-and post-adjudication services; multilingual/cultural/racial intake officers to facilitate the youth’s release from detention. Improve defender services Recognition by defenders of their role in policy reform, exposing abusive practices in detention, the overuse of detention, overcrowding, DMC, and disparities in case processing and outcomes for kids of color. Ongoing training in defense advocacy of juveniles. Stop ―dumping‖ of youth from other systems School administrators/ decisionmakers and key mental health personnel must be at the table and actively participate in reaching a consensus as to the use of detention and the implementation of JDAI strategies. Reach a common understanding that it is harmful to children, and inappropriate, to detain kids in order to provide for their health and mental health needs. Develop a ―system of care‖ to leverage resources and provide comprehensive services to children outside of detention. Minimize school as the entry DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-29 Specific Strategies To Reduce Racial Disparities (continued) Issue What We’re Looking For, Why This Is Important Review, Observe, and Interview Major Findings Best Practices, Recommendations point into detention by stopping the criminalization of schoolbased behaviors. Eliminate responsibilities that have been transferred from schools to the juvenile justice system. STOP opening the front door to detention so readily. Include communities of color in decision making It’s not enough to build ties with communities of color; they must be included in, and have an equal voice in the decisions necessary to foster change. Communities of color are at the table providing their unique perspectives in the decisionmaking process. DMC Technical Assistance Manual, 4th Edition • Chapter 3: Preparation at the Local Level 3-30 Chapter 4: Intervention Stephen Gies, Marcia Cohen, and Francisco Villarruel* Once a jurisdiction completes its initial assessment activities and conducts readiness events to prepare local stakeholders, it can select and implement intervention strategies to reduce DMC. A jurisdiction initiates this intervention phase by developing an intervention plan that serves as a road map for how the jurisdiction will proceed to reduce minority overrepresentation in its juvenile justice system. Although the state and/or OJJDP often support this phase through financial and technical assistance, the jurisdiction carries out the majority of the activities in targeted local communities because the factors or combinations of factors that emerge as important are most likely to be jurisdiction specific or community specific. This chapter will guide jurisdictions in the development of initiatives to reduce DMC. It first establishes guidelines for choosing strategies and developing an intervention plan. Secondly, it presents a conceptual framework, similar to the risk and protective factor framework used in prevention science, for use in plotting a course through the mixture of programs, services, and activities to determine the most appropriate strategy for targeting the specific DMC contributing factor(s) identified. The chapter organizes the universe of DMC strategies into three categories corresponding to the different audiences for which they are designed: (1) direct services, which target at-risk or system-involved youth, their families, and communities; (2) training and technical assistance, primarily for juvenile justice personnel; and (3) system change, which strives to modify aspects of the juvenile justice system that may contribute to DMC. For each category, the authors identify the applicable DMC factors, discuss strategies (i.e., programs, services, and/or activities) for addressing DMC, and highlight characteristics, principles, and procedures associated with successful programs. The chapter also includes case studies of seven jurisdictional DMC initiatives that have demonstrated a basic level of effectiveness. Each case study summarizes information about the jurisdiction’s DMC problem, the strategies it used to address the problem, and evidence regarding the strategies’ effects on DMC trends. Overview of the Intervention Phase The goals of the assessment phase were to determine whether DMC exists in the juvenile justice system—and at what decision points—and to identify the factors that contribute to DMC at those decision points. This process requires the jurisdiction to investigate each decision point and determine whether DMC is an issue at that point. If DMC exists, the jurisdiction must find the cause(s). Current research provides seven explanations as to why DMC may occur: (1) differential offending; (2) differential opportunities for * About the authors: Stephen Gies, Ph.D., is Senior Research Associate at Development Services Group, Inc., in Bethesda, Maryland. Marcia Cohen is Vice President for Research and Evaluation at Development Services Group, Inc. Francisco Villarruel, Ph.D., is a professor and fellow in the Department of Family and Child Ecology at Michigan State University in East Lansing, Michigan. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-1 prevention and intervention; (3) differential handling; (4) legislative, policy, and legal factors; (5) justice by geography; (6) indirect effects; and (7) accumulated disadvantage. Chapter 2 discusses these seven factors and an additional factor (mobility effects) in detail. A jurisdiction’s goal at the intervention phase is to address the DMC factors that it identified during the assessment phase by developing a comprehensive set of interrelated intervention strategies to reduce minority overrepresentation in its juvenile justice system. With this goal in mind, the next section offers general guidelines that jurisdictions should consider before selecting the specific strategy they will implement. Intervention Guidelines Jurisdictions should consider five guidelines when developing an intervention plan: (1) design a comprehensive, multimodal approach, when feasible; (2) prioritize strategies to focus on critical decision points; (3) choose interventions that the community is ready to implement; (4) use evidence-based strategies and draw on the successful experiences of current DMC initiatives, as applicable; (5) evaluate the effectiveness of the strategy. Designing a Comprehensive, Multimodal Approach Most often, DMC results from many factors and is present at more than one decision point. For this reason, jurisdictions typically must develop multiple strategies to address DMC and deploy them concurrently at several decision points. This comprehensive, multimodal approach is known as a jurisdictional DMC initiative. In a best-case scenario, a jurisdictional initiative involves the simultaneous implementation of multiple strategies targeting all critical decision points where DMC exists, and is directed at all identified DMC factors. According to Hsia, Bridges, and McHale (2004), the majority of jurisdictions recognize that multiple factors at different decision points contribute to DMC, but they invest primarily in direct services strategies (i.e., delinquency prevention and early intervention programs for minority youth, their families, and communities). Attempts to reduce crime and delinquency among minority youth will lead a jurisdiction to address the demand side of DMC and pay comparatively little attention to implementing systems change that would address the supply side of DMC. This finding is particularly troubling because, as previously stated, multiple factors generally contribute to minority overrepresentation in juvenile justice systems, and, therefore, the solution to the problem requires multiple strategies (Devine, Coolbaugh, and Jenkins, 1998). Nevertheless, whenever indicated (and feasible), OJJDP encourages jurisdictions to choose a variety of strategies drawn from the entire spectrum of options rather than focus on just one segment of the spectrum (e.g., direct services prevention and intervention programs). States and jurisdictions would do well to emulate Washington State’s example of implementing a comprehensive, research-based approach to DMC that incorporates cultural competency training and systems change as well as programmatic interventions. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-2 Prioritizing Strategies To Focus on Critical Decision Points Some jurisdictions may be unable to carry out a comprehensive DMC initiative because of limited resources, time, and/or data. In such cases, they should strongly consider those strategies that will reduce DMC the most and focus on the decision points where DMC is most pronounced or where the most minority youth are affected. For instance, strategies that reduce commitments of minority youth to state juvenile corrections may decrease the number of such youth in confinement. However, strategies designed to reduce overrepresentation at arrest may have a much more lasting and extensive impact on DMC because arrest is highly correlated with accumulated disadvantage for minority youth. Choosing Interventions That the Community Is Ready To Implement Researchers and practitioners alike have found that jurisdictions vary greatly in their interest in, and ability and willingness to try, intervention strategies and thus are at different stages of readiness (Weisheit, 1984; Aniskiewicz and Wysong, 1990; Bukaski and Amsel, 1994). Related factors that may hinder community readiness include a lack of consensus within the jurisdiction concerning the nature of the problem or solution, a dearth of resources, a lack of knowledge about the problem, and an unstable political climate. A community’s level of readiness to implement a particular strategy often plays a major role in determining whether the strategy will be successful. Therefore, OJJDP encourages jurisdictions to assess community readiness and select strategies with the greatest likelihood of successful implementation. For instance, although two jurisdictions may identify arrest as the decision point contributing most to overrepresentation of minority youth, each jurisdiction may address DMC differently, depending on community readiness. One jurisdiction may direct DMC activities at law enforcement, whereas another may address the same decision point through prevention programming that targets youth. The success or failure of either strategy is strongly influenced by the community’s readiness to accept it. Using Evidence-Based Strategies and Drawing on the Successful Experiences of Current DMC Initiatives Evidence-based programs and best practices are programs or sets of differentiated program elements that have been well researched and are associated with positive outcomes (Lipsey, 1995). Unfortunately, the literature on what works to reduce DMC is not as extensive as the literature on what works in delinquency prevention and other areas of juvenile justice. Moreover, jurisdictional DMC initiatives generally consist of many diverse strategies because, as previously mentioned, many factors typically contribute to DMC. On the one hand, jurisdictions are encouraged to design a comprehensive initiative consisting of multiple strategies whenever indicated and feasible. On the other hand, aggregating these strategies into a single initiative makes disentangling the effects of one strategy from DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-3 another extremely arduous and virtually prohibitive, given the current state of the science in DMC research. Thus, while there are no “blueprints”1 for reducing DMC, a number of jurisdictional initiatives composed of diverse strategies that have demonstrated encouraging results in reducing minority overrepresentation at various decision points do exist. Because the research base for these initiatives is not sufficiently rigorous, they cannot be ranked using the same rating system applied to programs in OJJDP’s Model Programs Guide and Database.2 Instead, they are best described as developing or encouraging initiatives that have a strong theoretical base and for which some empirical and anecdotal evidence of effectiveness exists. It is important to note, however, the evidence is not sufficient to establish a reasonable causal inference between the intervention strategy and the outcomes. These initiatives are, nevertheless, excellent foundations from which other jurisdictions can learn. Planning To Evaluate Effectiveness Regardless of the strategies a jurisdiction chooses, the ultimate goal of all interventions is to decrease the overrepresentation of minority youth. It is imperative that a jurisdiction design any DMC initiative to prove that it reduced DMC (i.e., caused a decrease in the RRI for minority youth at the juvenile justice decision points targeted by the strategy). Therefore, the jurisdiction should develop its intervention plan in tandem with a plan to evaluate the initiative’s effectiveness (see chapter 5 for guidelines on evaluating DMC strategies). Conceptual Framework for Selecting DMC Strategies Appropriate strategies for reducing DMC are those that address the specific factors contributing to DMC and target the juvenile justice decision points where the DMC problem exists. For example, if the DMC assessment shows that differential offending is the single most important factor contributing to overrepresentation of minority youth at the arrest decision point, strategies should address the disproportionate involvement of minority youth in crime. It would make little sense to implement a strategy to reduce differential handling of youth in detention, where DMC may not exist. Identifying high-quality programs that can address specific DMC factors in a given community has been one of the most difficult obstacles to developing effective DMC initiatives. The DMC-factor-based conceptual framework presented in this chapter links particular DMC contributing factors to particular strategies for reducing DMC. Researchers adapted the risk and protective factor conceptual model used in prevention research to create the framework. The adaptation appears to be a natural extension of the risk and protective factor model because certain factors can increase or decrease the overrepresentation of minority youth in the juvenile justice system, just as risk and protective factors increase or decrease the likelihood of problem behavior.3 DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-4 According to the risk and protective factor model, a mix of risk and protective factors that influence a youth over the course of adolescent development in either positive or negative ways determines the youth’s potential to engage in problem behaviors such as delinquency, substance abuse, dropping out of school, and HIV/AIDS risk behaviors (Hawkins, Catalano, and Miller, 1992). Risk factors increase, and protective factors decrease, the likelihood of problem behaviors. While no single risk factor is more potent than any other, in general the more risk factors and the fewer protective factors present in life, the greater the probability of problem behaviors (Bry, McKeon, and Pandina, 1982; Newcomb, 1995). Although there are no corresponding protective factors in DMC research as yet, certain factors in the community resemble the risk factors in prevention research in a number of ways. Like risk factors, the presence of these contributing factors—alone or in combination—increases the likelihood of negative effects (i.e., minority overrepresentation in the juvenile justice system). As is true for risk factors, no single DMC factor is more potent than any other; however, the greater the number of DMC factors, the greater the probability of minority overrepresentation. Like risk factors, DMC factors can have a cumulative effect on a youth’s life course that leads to increased involvement with the juvenile justice system. For instance, juvenile justice decisionmakers tend to use prior record and seriousness of offense as the basis for making any determination (e.g., diversion, detention, formal sanctioning). Therefore, any DMC factor that artificially inflates a young offender’s delinquent history will have profound consequences later in life. This DMC-factor-based conceptual framework can help communities identify specific strategies for the DMC factors they identified during the assessment phase. Once a jurisdiction has diagnosed the factors contributing to DMC at particular decision points, it can refer to table 1 to find the most appropriate strategies for addressing those factors. Table 1. DMC Initiatives Categorized by DMC Contributing Factors DMC Contributing Factors Differential Handling Legislative, Policy, and Legal Factors Bernalillo County √ √ Cook County √ √ County DMC Initiatives Differential Offending Hillsborough County King County Mesa County Differential Opportunities √ √ Santa Cruz County Indirect Effects Accumulated Disadvantage √ √ √ √ √ Multnomah County Justice by Geography √ √ √ √ √ DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-5 Suppose, for example, that a universe of 50 strategies is available to deal with DMC issues. Suppose that a jurisdiction has limited resources and, therefore, can address only one DMC factor. The jurisdiction’s assessment identifies differential offending as the single most important factor contributing to DMC. Which of the 50 strategies should the jurisdiction choose? A DMC-factor-based conceptual framework, as illustrated in table 1, enables the jurisdiction to narrow the search to strategies that address differential offending. Instead of considering all 50 strategies, the jurisdiction can choose from the much smaller number of strategies appropriate for a specific DMC factor—in this case, differential offending. Aided by the DMC-factor-based conceptual framework, jurisdictions can move quickly from the assessment phase directly to the intervention phase, without needing to conceptualize an original method for selecting strategies or investigate the entire universe of possible strategies.4 Types of Strategies for Reducing DMC As previously noted, a DMC initiative may include a number of different programs, services, or activities directed at several juvenile justice decision points and DMC factors. The large and varied universe of possible DMC strategies can be organized into three categories based on the target audience: (1) direct services, which address the requirements of youth; (2) training and technical assistance, which focus primarily on the needs of law enforcement and juvenile justice personnel; and (3) system change, which involves altering aspects of the juvenile justice system that may contribute to DMC. Direct services decrease overall crime and delinquency, particularly among minority youth, to reduce the potential for DMC and are sometimes thought of as demand-side strategies. Training and technical assistance strive to equip juvenile justice service providers with the necessary skills and resources to reduce DMC. System change strategies create wide-ranging and long-lasting modifications in policies, procedures, and/or laws that will decrease and prevent DMC. Training and technical assistance and system change are viewed as supply-side responses to DMC. The following sections describe these three types of strategies in detail and give examples of programs and services in each category. Direct Services (DMC factors: differential offending, differential opportunities for prevention and intervention, indirect effects, and accumulated disadvantage) When used as a DMC strategy, direct services focus on giving at-risk and delinquent minority youth greater access to appropriate prevention and early intervention programs so that they receive services needed to build skills, improve social functioning, and form healthy relationships with family members, other adults, and peers (OJJDP, 2001:35). Other types of direct services include diversion programs, alternatives to secure confinement, and advocacy. The majority of states recognize the need for diverse strategies to address the factors at different decision points that typically contribute to DMC; however, most states and communities have invested primarily in direct services strategies consisting of delinquency prevention and early intervention programs that DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-6 target minority youth, their families, and communities (Hsia, Bridges, and McHale, 2004:17). Prevention and Early Intervention Prevention programs generally are proactive and are aimed at youth who exhibit risk factors for delinquency but have not been adjudicated delinquent. They address the recognized factors that can lead to or serve as a catalyst for delinquency or other problem behaviors in minority youth (e.g., lack of education or employment opportunities, attitudes in the community or family that condone criminal activity, lack of parental supervision). Types of prevention programs include family therapy, parent training, cognitive behavioral treatment, mentoring, academic skills enhancement, afterschool recreation, vocational/job training, and wraparound services.5 In jurisdictions where the identified causes of overrepresentation can best be addressed with a programmatic response, it may be necessary to develop new programs and services for minority youth. However, it also may be feasible to modify, enhance, or expand current initiatives. Building on existing resources can be beneficial. For example, youthserving programs already in place may have a foundation in terms of funding, commitment from key stakeholders (e.g., elected officials, chamber of commerce representatives, neighborhood advisory council members, and religious leaders), and evidence of effectiveness that may improve with appropriate modifications and additional support. Although theoretically similar to prevention strategies in that they generally attempt to impede a youth’s trajectory toward delinquent and problem behaviors, intervention programs are operationally different in that they are reactive and assume that youth have been arrested for a delinquent act and either diverted from the juvenile court or adjudicated guilty. Moreover, interventions typically are graduated to reflect that the penalties for delinquent activity should move from limited interventions to more restrictive measures according to the severity and nature of the crime. In other words, youth who commit serious and violent offenses should receive more severe sentences than youth who commit less serious offenses. On the following pages, exhibit 1 summarizes research findings on characteristics of successful prevention and intervention programs, and exhibit 2 identifies four principles shared by programs that have been effective in reducing recidivism. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-7 Exhibit 1: Characteristics of Successful Programs Communities should select programs that are evidence based. They should base their selection of program models on what is known about effective programs. As discussed in Krisberg et al. (1995), the National Council on Crime and Delinquency (NCCD) collected and analyzed information on 209 prevention and intervention programs and identified eight characteristics of effective programs. NCCD found that these eight elements contribute to success in improving outcomes for youth regardless of program type (e.g., prevention, diversion or residential treatment, aftercare) or youth population served. According to this research, successful programs: • Are holistic, simultaneously addressing many aspects of a youth’s life. • Develop individualized treatment plans for youth, providing the services, level of support, and degree of supervision that each juvenile’s needs, skills, and circumstances dictate. • Are intensive, often involving multiple contacts weekly or even daily with at-risk youth, and continue over a relatively long period of time. • Give youth frequent positive and negative feedback (as appropriate) regarding their progress. • Typically operate outside the formal juvenile justice system. • Build on a youth’s strengths rather than focusing on deficiencies. • Use a socially grounded framework for understanding and dealing with a youth’s situation, rather than relying mainly on individual or medical-therapeutic approaches. • Offer services provided by energetic and committed staff. Exhibit 2: Principles for Reducing Recidivism Researchers have scientifically demonstrated a variety of theoretically sound and well-designed intervention strategies that are effective for juvenile offenders (see OJJDP’s Model Programs Guide). Latessa, among others, has identified four principles that effective programs share to appreciably reduce recidivism and that should guide program development and implementation (Latessa, 2004). These principles are: Risk (Who requires the most intensive programs?). Programs should reserve the most intensive correctional treatment and interventions for higher risk offenders (i.e., those who are more likely to recidivate) and should not waste money on services for offenders who do not need them. More importantly, research has clearly demonstrated that when lower risk offenders are placed in more structured programs, their failure rates often increase, which reduces the program’s overall effectiveness. Need (What factors should programs target?). Programs should target risk/need factors that are both dynamic (i.e., amenable to change) and criminogenic (i.e., associated with criminal conduct). Although many risk factors are criminogenic, some, such as age, gender, and early criminal behavior, are static—that is, they cannot be changed in treatment. To be effective, rehabilitation services must focus on both dynamic and criminogenic factors, such as the following: antisocial attitudes, values, and beliefs; antisocial peer associations; substance abuse; lack of problem-solving and self-control skills; and other factors highly correlated with criminal conduct. (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-8 Exhibit 2: Principles for Reducing Recidivism (continued) Programs that target noncriminogenic factors (e.g., self-esteem, physical conditioning, creative abilities, understanding one’s culture or history) will have little effect on recidivism rates. Militarystyle boot camps for offenders are one example of programs that target noncriminogenic factors. Most studies show that boot camps have little impact on future criminal behavior because they tend to focus on factors that are not crime producing, such as drill and ceremony, physical conditioning, discipline, self-esteem, and bonding together offenders. Treatment (How should programs target risk/need factors?). The most effective programs and approaches are behavioral in nature. Behavioral programs have several attributes. First, they focus on the present circumstances and risks that are responsible for the offender’s behavior. Second, they are action oriented rather than talk oriented. (Offenders do something about their difficulties rather than just talk about them.) Third, they teach offenders new, prosocial skills through modeling, practice, and reinforcement to replace antisocial ones such as stealing, cheating, and lying. Examples of behavioral programs include structured social learning programs where new skills are taught and positive behaviors and attitudes are consistently reinforced; cognitive behavioral programs that target attitudes, values, peers, substance abuse, anger, etc.; and family-based interventions that train families in appropriate behaviors. Fidelity (How well does the implemented program reflect the program design?). Fidelity is the degree of fit between the components defined in the program design and the components the community actually implements. Fidelity is essential to determine whether a program caused the measured outcome effects. Concern over fidelity sprang from attempts to explain why so many evaluations of previously validated programs yielded null or inconsistent results (Mayor and Davidson, 2000). Many high-quality programs fail to take adequate steps to monitor and verify program fidelity (CSAP, 2002). A program lacks fidelity when it is implemented inconsistently (e.g., when practitioners differ in the number of sessions they deliver, the length of time they provide for each session, or the number of objectives they address). Research shows that program participants benefit more when there is a high degree of fidelity (i.e., when practitioners are faithful to the details of the program design) (CSAP, 2002). Diversion Diversion programs are a specialized subset of interventions that typically serve youth who are at-risk and/or have been arrested or referred to the juvenile court for status offenses or nonserious delinquent offenses. In the latter instance, jurisdictions use diversion in lieu of formal court processing. The concept of diversion is based on the theory that processing certain youth through the juvenile justice system may do more harm than good (Lundman, 1993). Courts may inadvertently stigmatize some youth for having committed relatively petty acts that might best be handled outside the formal system. In addition, involvement with the juvenile justice system may increase some risk factors for delinquent behavior, such as delinquent peer associations. Diversion also is designed to reduce demands placed on overburdened juvenile courts and detention facilities, so they can concentrate on more serious offenders (Sheldon, 1999). To reduce minority overrepresentation, diversion programs divert youth from formal court processing while still providing a mechanism that holds them accountable for their actions. Examples of diversion programs include community service, informal hearings, family group conferences, victim impact panels, victim-offender mediation, mentoring, teen courts, restitution, and other restorative justice strategies. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-9 Many diversion programs incorporate the restorative justice principles of repairing the harm done to victims and communities. The essence of restorative justice lies in the perspective that crime harms people, communities, and relationships. Consequently, if crime is about harm, then the justice process should emphasize repairing the harm (Bazemore, Seymour, and Rubin, 2000). Restorative justice differs from traditional juvenile justice in the way it views and addresses crime. Whereas the traditional justice system focuses on the offender’s culpability and need for sanctioning and/or treatment, a restorative system takes a broader approach. Howard Zehr (1990) maintains that a restorative system focuses on three primary questions: What is the nature of the harm resulting from the crime? What needs to be done to repair the harm? Who is responsible for the repair? Thus, a restorative system places more emphasis on the victim and views the offender as more than an object of punishment—the offender is held accountable for repairing the harm that the crime caused. A restorative court supports, facilitates, and enforces reparative agreements (Bazemore, Seymour, and Rubin, 2000). Alternatives to Secure Confinement (Detention and Corrections) Like diversion, the concept of secure confinement alternatives is based on the premise that time spent in secure detention or corrections may do more harm than good for some youth. In many instances, juvenile offenders who might qualify for less restrictive treatment or sanctions are securely confined because no appropriate options exist. Detention and corrections alternatives give such youth the benefit of remaining in their communities with greater access to needed resources without endangering the community and at much less expense then secure confinement (OJJDP, 2001:37). This approach keeps less serious or nonviolent offenders at home or in their home communities, thus increasing the availability of secure beds for the most serious and violent offenders (OJJDP, 2001:37). There are many different types of secure confinement alternatives, but all are used as either a pre- or post-adjudication secure confinement alternative to keep youth from being placed behind bars for any length of time (see exhibit 3 for examples). A pre-adjudication program is an alternative to detention and helps ensure that youth appear in court. A postadjudication program is an alternative to correctional placement that is more severe than regular supervision but less restrictive than confinement. In addition, correctional alternatives generally entail a longer length of stay and are more likely to provide an array of treatment interventions designed to effect behavioral change. Examples of secure confinement alternatives include home confinement (or house arrest), day (or evening) treatment facilities, shelter care, specialized foster care, attendant or holdover care, and intensive supervision programs (see exhibit 3). DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-10 Exhibit 3: Alternatives to Secure Confinement Home confinement or house arrest is a community-based program designed to restrict the activities of offenders in the community. Offenders live at home, go to work, run errands, attend school, and fulfill other responsibilities. However, they are closely monitored (electronically and/or through frequent contact with staff) to ensure that they comply with the conditions that the court has set. They are restricted to their residence for varying lengths of time and are required to maintain a strict daily schedule. Home confinement programs can be used as a pre- or postadjudication program. Day (or evening) treatment is a highly structured, nonresidential, pre- or post-adjudication, community-based alternative that provides intensive supervision to ensure the community’s safety and a range of services for offenders to prevent future delinquent behavior. Offenders must report to the treatment facility on a daily basis at specified times (either during the day or in the evening) for a specified period (generally at least 5 days per week) but are allowed to return home at night. The program may provide special weekend activities. Services may include individual and group counseling, recreation, education, vocational training, employment counseling, life skills and cognitive skills training, substance abuse treatment, and referrals to community resources. Shelter care is a pre- or post-adjudicatory alternative that offers residential care for youth who need short-term placement (i.e., for 1 to 30 days) outside the home. Shelter care is used for juveniles who require more intensive supervision than that provided by nonresidential options and for youth who must be detained because no parent or family member is available. Facilities are staff secure or nonsecure. Staff monitor youth 24 hours per day, 7 days per week, and provide a full schedule of structured educational and recreational activities. Specialized foster care is an adult-mediated treatment model that recruits and trains families to offer placement and treatment for youth with a history of chronic and severe delinquency. Usually, youth are closely supervised at home, in the community, and at school. Foster care parents provide one-on-one mentoring and consistent discipline for rule violations. Foster care parents typically receive special training on the needs of youth involved in the juvenile justice system and have access to additional resources to address special situations. Attendant or holdover care is a pre-adjudicatory alternative that provides a safe, temporary setting for youth who do not meet detention criteria but need a place to stay until other suitable arrangements can be made. This alternative is used for youth who require relatively intensive but very short-term supervision following an arrest and prior to a court hearing. It is especially appropriate for runaways from rural areas and minors who violate liquor laws. The programs are staffed by community volunteers or paid personnel and administered by law enforcement, juvenile court, probation, or a nonprofit organization. In more remote areas, staff may be on call. Intensive supervision programs (ISPs) are a community-based, post-adjudication, nonresidential alternative that provides a high degree of control over offenders to ensure public safety, without the additional costs associated with confinement. ISPs have small caseloads, strict conditions of compliance, and high levels of contact and intervention by the probation officer or caseworker. ISPs typically use a variety of risk control strategies (e.g., multiple weekly face-to-face contacts, evening visits, urine testing, electronic monitoring) and deliver a wide range of services to address offenders’ needs. ISPs generally fall into two categories: those that serve probationers who have been assessed as high risk and those developed specifically as alternatives to institutionalization. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-11 Advocacy A direct service strategy aimed at reducing minority overrepresentation need not, and should not, be limited to the provision of supervision and substantive interventions to decrease delinquency and other problem behaviors. Indeed, minority youth and their families may benefit from assistance in successfully navigating the complex procedures, overwhelming multitude of social services agencies, and circuitous organizational tiers that often comprise a juvenile justice system response to a young offender (OJJDP, 2001:40). In general, advocacy connects youth and families with a variety of social networks and service providers to integrate services that multiple agencies provide, ensure continuity of care, and facilitate the development of a youth’s social skills (Sheldon, 1997). Advocacy is a “client-level strategy for promoting the coordination of human services, opportunities, or benefits” (Macallair, 1994:84) that helps youth and families move through this maze. Advocacy services can take many forms. Advocates can assist youth and families in understanding the juvenile court process from arrest to disposition and the roles of the various system representatives. They also can help parents and other family members overcome distrust of and hostility toward the system. Advocates can assist youth and families in gaining improved access to counsel and in understanding their legal rights. They also can facilitate coordination within the justice system and/or between the justice system and other services. Detention advocacy programs offer comprehensive services to youth who are likely to be detained pending adjudication. Once a potential client is identified, a case manager develops and presents to the judge a release plan that includes a list of appropriate community services (e.g., tutoring, drug counseling, family counseling) that will be accessed on the youth’s behalf and specific objectives (e.g., improved grades, victim restitution, drug-free status) that will be used to evaluate the youth’s progress. If the judge deems the release plan acceptable, the youth is released under supervision (Sheldon, 1997). Detention advocacy programs ensure that youth receive frequent, consistent supervision and that families obtain needed support (e.g., assistance in securing employment, daycare, drug treatment, and income supplementation such as food stamps). Training and Technical Assistance (DMC factors: differential opportunities for prevention and intervention; differential handling; legislative, policy, and legal factors; justice by geography; and accumulated disadvantage) All children and youth, regardless of race or ethnicity, have more in common than they have differences, but culture does influence how they behave and how the juvenile justice system perceives them. Although racial discrimination may emerge from these differences some of the time at some stages of the juvenile justice process, little evidence exists that racial disparities result from systematic, overt bias. Instead, such disparities in the juvenile justice system appear to be the indirect result of unintentional racial bias and DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-12 the amplification over time of initial disadvantages (Sampson and Lauritsen, 1997:311). Indirect racial discrimination can occur even when no intention to discriminate exists. For example, a rule stating that all students must not wear anything on their heads could result in discrimination against students whose religion requires the wearing of headwear. In law enforcement and juvenile justice, communities can enable decisionmakers and all personnel who routinely interact with system-involved youth to gain the knowledge and skills they need to work effectively with culturally diverse minority youth and to address indirect (particularly unintentional) racial bias. Cultural competency training and technical assistance is the most common method for disseminating these practical tools. Cultural competency can be defined as a set of congruent behaviors, attitudes, and policies that interface with each other in a system, an agency, or a network of professionals to work effectively in cross-cultural situations (Cross et al., 1989; Isaacs and Benjamin, 1991). The term culture refers to integrated patterns of human behavior— including thoughts, communications, actions, customs, beliefs, values, and institutions— associated with particular racial, ethnic, religious, or social groups. The term culturally competent describes the capacity to function effectively within the context of the behavior patterns characteristic of such groups. Operationally, cultural competency entails transforming knowledge about particular individuals and groups into culturally appropriate (as those individuals or groups define them) standards, policies, practices, and attitudes that increase the quality of services, thereby producing better outcomes (Davis, 1997). Cultural competency is not an either-or issue. Rather, it is a developmental process that occurs along a continuum. Researchers have separated the continuum into six clusters reflecting degrees of competency: (1) cultural destructiveness, (2) cultural incapacity, (3) cultural blindness, (4) cultural precompetence, (5) cultural competence, and (6) cultural proficiency. It has been said that, at best, most social services agencies responsible for meeting the needs of children and families fall somewhere between cultural incapacity and cultural blindness (Cross et al., 1989). This dismal assessment of the cultural competency of social services agencies suggests that minority youth may be treated differently from other youth. For example, Latino youth face many barriers in the justice system, including a lack of bilingual and culturally competent staff, which can result in harsher treatment and profound confusion and frustration for youth and families who speak little or no English (Villarruel, 2002). Cultural Competency Training and Program Development Benefits of Cultural Competency Training Providing cultural competency training is one way in which agencies can increase the effectiveness of staff at all levels. Cultural competency training can engender a deeper awareness of cultural factors (e.g., differences in communication styles, body language and demeanor, language use, beliefs about the family, attitudes toward authority figures) that typically influence decisionmaking about youth (OJJDP, 2001:42). Such training DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-13 seeks to increase knowledge about different cultures, address cultural biases and stereotypes, and produce changes in the practices of individuals and the organizations to which they belong. Cultural competency training can alter the beliefs and behaviors of juvenile justice personnel, system administrators, elected officials, and the general public in far-reaching ways regarding minority overrepresentation. Through such training, for instance, juvenile justice professionals could gain essential contextual information regarding Native American and Alaska Native youth, including the history of conquests by white traders and settlers, genocide, decimation from disease, forced loss of cultural heritage and sacred land, and the evolution of alcohol use, violence, and chronic disease (DeBruyn et. al., 2001). A working knowledge of the history, traumatic losses, cultural frameworks, and cultural change of different racial and ethnic groups could help these professionals identify the combinations of risk and protective factors that violence prevention strategies should address to be most useful for youth in each group. If this type of cultural competency training was mandatory and provided systematically throughout the juvenile justice and related youth-serving systems, cross-cultural understanding within organizations would improve and interventions would become culturally relevant and more likely to be successful (Ellis, Klepper, and Sowers, 2001; DeBruyn et. al., 2001; Federle and Chesney-Lind, 1992). Resources for Cultural Competency Training The American Correctional Association and the Police Executive Forum have developed an excellent example of cultural competency training entitled Training in Cultural Differences for Law Enforcement/Juvenile Justice Practitioners. This OJJDP-sponsored training curriculum is designed to increase the effectiveness of police and other juvenile justice personnel who work with minority victims, suspects, and offenders and to improve policies governing the way police, juvenile court, and corrections personnel treat minority youth. The participant’s manual contains four modules, a cultural diversity action plan, and extensive appendixes. The first module identifies the benefits to professionals of cultural diversity training (e.g., increased safety, compliance with police agency directives, increased job satisfaction, fewer citizen complaints against the police, minimal risk of successful lawsuits, decreased paperwork and hearings associated with complaints and lawsuits, increased community involvement and cooperation).The second module explains the concepts of culture and cultural diversity and defines terms such as culture, subculture, values, prejudice, stereotyping, ethnocentrism, discrimination, racism, acculturation, and assimilation. The third module covers cross-cultural communication, with an emphasis on components of effective communication, factors that impede the communication process, automatic actions people take when they witness an event, and communication styles. The fourth module presents questions that professionals should ask themselves when they interact with people from different cultural groups and gives information on how to respond effectively to people from different cultures. The cultural diversity action plan contains exercises on self-awareness, values, prejudice, communication, goals, and task management. The appendixes focus on juvenile courts, juvenile community programs, juvenile institutional workers, and law enforcement. Training in Cultural Differences for Law Enforcement/Juvenile Justice DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-14 Practitioners (NCJ Number 163318) is available through the National Criminal Justice Reference Services at www.ncjrs.gov. Although not a juvenile justice organization, the National Multicultural Institute (NMCI) is a valuable resource for training, consulting services, and publications on cultural competency and related issues. Founded in 1983, NMCI helps individuals, organizations, and communities create a society strengthened and empowered by diversity. NMCI has extensive experience working with corporations, educational institutions, government agencies, and nonprofit organizations in the areas of workforce diversity, human resources management, multicultural education, and cross-cultural conflict resolution. NMCI organizes conferences in the spring and fall, provides customized training and consulting interventions for organizations, and offers many training manuals and other publications. In the past few years, more than 35,000 participants have attended NMCI’s national conferences and onsite workshops. Conference participants have come from all 50 states and other countries. NMCI training manuals include the following: Developing Diversity Training for the Workplace: A Guide for Trainers, Teaching Skills and Cultural Competency: A Guide for Trainers, Cultural Competency in Healthcare: A Guide for Trainers, Multicultural Case Studies: Tools for Training, and Customizing Diversity Training with Case Vignettes. NMCI is located in Washington, DC. Web site: www.nmci.org/index.htm. Contact information: Maria Morukian, program manager (202–483–0700). Providing Culturally Appropriate Services Providing culturally appropriate services is another way that communities can improve the cultural competence of their social services agencies. This strategy, however, is often implemented in a well-meaning yet inappropriate manner. For example, some programs focusing on drug treatment, delinquency prevention, violence prevention, or mental health problems may focus on services for particular ethnic groups or subcultures but use conventional means to provide these services. A more appropriate method to provide culturally competent services to a specific ethnic group is to use approaches grounded in the culture and traditions of the population being served. This approach is based on the assumption that cultural values and traditions, when incorporated into a program, improve the integrity of the treatment. Thus, for services to be effective, providers must consider the specific characteristics of the group being targeted, and the services must be grounded in the value system, traditions, and language of the targeted group. Examples of Culturally Competent Programming The Strengthening Families Program (SFP) is a classic example of culturally competent programming. SFP is a family skills training program that increases resilience and reduces risk factors for substance abuse, depression, violence and aggression, delinquency, and school failure in high-risk children (ages 6 to 12) and their parents. The SFP curriculum includes three courses (parent skills training, children’s skills training, and family life skills training) taught in fourteen 2-hour sessions. In the first hour, parents and children participate in separate skills-building classes that are each led by two co­ leaders. During the second hour, families practice structured family activities, therapeutic child play, family meetings, communication skills, effective discipline, reinforcing DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-15 positive behaviors in each other, and jointly planning family activities. SFP was developed in 1982, and positive outcomes were first demonstrated in the original research study conducted from 1983 to 1987. Since then, more than 15 independent research replications have yielded significant positive results. SFP has been modified for African American, Asian/Pacific Islander, Hispanic, and Native American families; rural families; and families with young teens. Canadian and Australian versions also have been tested. Although originally developed for children of substance abusers, SFP is effective and widely used with non-substance-abuse parents in many settings. Web site: www.strengtheningfamiliesprogram.org/index.html. Contact information: Strengthening Families, University of Utah (801–581–7718). Another example is the Syracuse Family Development Research Program (FDRP), a comprehensive early childhood program targeting African American, single-parent, economically disadvantaged families. The intervention bolsters child and family functioning (i.e., improves children’s cognitive and emotional functioning, helps children develop a positive outlook, and decreases juvenile delinquency in later years) through home visitation, parent training, and individualized daycare. The original FDRP provided a full range of education, nutrition, health and safety, and human services resources to participating families that began prenatally and continued until the children reached elementary school age. Child development trainers work with participating families during weekly home visitations to increase positive family interaction, cohesiveness, and nurturing. In the FDRP’s Children’s Center, infants are assigned to a caregiver for attention, cognitive and social games, sensorimotor activities, and language simulation. The preschool program supports child-chosen opportunities for learning and peer interaction through specific types of activities that each have dedicated space at the Children’s Center. Contact information: Dr. Alice Sterling Honig, Professor Emerita of Child Development, Syracuse Family Development Research Program (315–443–4296, e-mail ahonig@mailbox.syr.edu). Cultural Competency at the Oregon Youth Authority The Oregon Youth Authority (OYA) is an excellent example of a community agency committed to cultural competency training and program development. In May 2000, OYA’s Juvenile Policy Committee developed and formally adopted a set of organizational protocols that provide a foundation and direction to assist OYA in becoming a culturally competent agency. Within the OYA, the Office of Minority Services (OMS) “. . . provide[s] leadership, advocacy, and guiding principles to assist [OYA] in its ongoing efforts to become a culturally competent organization and address the complex issues of a culturally and ethnically diverse agency. Minority Services supports OYA in embracing the values and strengths of all cultures and the implementation of culturally relevant, gender-specific, and language-appropriate treatment services that empower youth to make positive changes.” In addition, the Juvenile Crime Prevention Advisory Committee (JCPAC) adopted a policy requiring counties to ensure that services are culturally appropriate and gender specific; JCPAC commissioned OMS to develop an interagency committee charged with designing training and technical assistance on cultural competency/gender-specific services to assist DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-16 counties in incorporating these services in their juvenile crime prevention plans. (Exhibit 4 summarizes guidelines for developing culturally competent programs, used in Marion County, Oregon, as presented in Oregon’s Cultural Competency and Gender-Specific Resource Guide.) Exhibit 4: Guidelines for Developing Culturally Competent Programs The Marion County Commission on Children and Families (cited in Oregon’s Cultural Competency and Gender-Specific Resource Guide), developed a list of program guidelines to assist in the process, program design, and evaluation of culturally competent programs. All the guidelines listed are important and, if incorporated, maximize program effectiveness. These guidelines are applicable to all services along the continuum from community-based prevention programs to intensive residential programs, detention, and state institutions. Even if programs serve small diverse populations, they can meet the guidelines and create an effective continuum of care. A. Environment (the community the program operates in) • The program has identified resource people from various cultural populations of the community. • The program has developed and maintains ongoing direct person-to-person contact with these resource persons. • Resource persons are asked to participate in program evaluation(s). • The program has accurate demographic data about various cultural communities it proposes to serve. • The program uses demographic data to ensure that all cultural populations receive equal access and/or equal services as required to meet their needs. B. Management control (the governing and administration system for the program) • People from various ethnic and cultural backgrounds will comprise the board. • The program has written policies or plans, which call for the program to become culturally competent. • The board of directors must approve written policies or plans. C. Resources (the money, technology, human resources, etc., coming into the program) • The program employs ethnic, bilingual, and culturally diverse staff and/or volunteers in positions that have direct contact with clients. • The program has the capacity to serve clients from all ethnic and cultural demographics in the community. D. Change technology (activities/interactions that create positive change or add value) • The program has incorporated the concepts of culturally competent, gender-specific, and language-appropriate services into its ongoing training programs. • All staff members and volunteers receive at least 4 hours of cultural competency training annually. (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-17 Exhibit 4: Guidelines for Developing Culturally Competent Programs (continued) • The program has a resource library containing up-to-date articles, books, tapes, etc., related to cultural issues. • The program provides opportunities, which help clients understand and appreciate cultural differences. E. Facility and grounds (actual site(s) where program activities and services occur) • Architectural and structural barriers are removed to provide reasonable accommodation and access to services. • Posted information is written in a variety of appropriate languages. • The atmosphere in the program acknowledges and welcomes people from diverse cultural backgrounds through artwork, posters, books, etc. F. Services delivered (actual services/products delivered, hopefully with positive change or value added. Include any resources leaving the program, such as staff turnover) • Clients from all cultural and language backgrounds successfully complete the program in similar ratios. • All clients receive equitable services, appropriate and relevant to the clients’ cultural and language backgrounds. • Personnel records reflect nondiscriminatory practices in the hiring, promotion, and retention of staff. G. Feedback/evaluation (information relating to program goals and objectives) • The program assesses its "cultural competency" policy and plans with input from staff, volunteers, resource people, and clients. • All programs, organizations, projects, etc., will submit an outline of their cultural competency plan detailing the goals, objectives/activities, and post-assessment for each of the criteria. Source: Oregon’s Cultural Competency and Gender-Specific Resource Guide, Werber, 2001. OMS provides support, including staff development, training, technical assistance, and program development, to assist OYA facilities in delivering effective, culturally specific treatment services. OMS also helps regions to identify key focus areas for their regional diversity plans, establish measurable outcomes for activities in these focus areas in collaboration with staff from the Community Resources Unit, and advocate for resources. OMS provides consultation, training, and technical assistance for OJJDP and the National Institute of Corrections on cultural competency/diversity, DMC, intensive aftercare programs, and elements of transition. Web site: www.oregon.gov/OYA/oms.shtml. Contact information: (503–737–7205, e-mail oya.info@oya.state.or.us). Culturally Competent Staffing Practices Staffing practices can be a powerful tool for strengthening an organization’s capability to deliver culturally competent services. Juvenile justice agencies can hire, promote, and retain at all levels qualified, culturally competent personnel who belong to the minority DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-18 groups that these agencies serve (see exhibit 5). As part of an effort to develop and maintain a culturally diverse staff, some organizations have established minority internship programs. Agencies also can hire interpreters and translators so that nonEnglish speaking youth and families can participate in juvenile court proceedings. Jurisdictions have taken a number of other steps to improve their juvenile justice systems, including adding juvenile court probation staff in tribal juvenile courts, recruiting members of minority groups to serve on community boards, reducing barriers to advocacy, developing brochures and other materials in languages other than English, and providing better information to parents (especially in languages they can understand) (Hsia, Bridges, and McHale, 2004). Exhibit 5: Characteristics of Culturally Competent Programs and Organizations The literature on culturally competent services has identified a number of principles and practices that characterize successful culturally competent programs and organizations (King, M.A., Sims, A, and Osher, D., 2001). They include the following: • Commitment to providing culturally competent services. Ensure that your organization’s or program’s mission statement expresses a commitment to cultural competency and gender specificity as an integral aspect of all activities. Conduct a comprehensive assessment of your organization’s cultural competency and gender specificity. • Awareness and acceptance of the concept of diversity. Form a committee or task force on cultural competency and gender specificity within your program or organization to foster awareness and acceptance of the concept of diversity. • Efforts to promote staff self-awareness and self-appreciation. Through informal brown bag lunches and other means, engage staff in discussions and activities that give them an opportunity to explore attitudes, beliefs, and values related to cultural and gender diversity. Every fiscal year, dedicate a portion of your program or organizational budget to funding staff development activities such as participation in conferences, workshops, and seminars on cultural competency and gender-specific services. • Understanding of the dynamics of cultural differences and how they influence the development of relationships and interventions. Gather and organize resource materials related to cultural competency and gender specificity for staff in your organization or program. • Understanding of client and community cultural backgrounds, values, and contact with service delivery system. Identify the different cultural, linguistic, and racial and ethnic groups and groups of girls and young women within the geographic area that your organization or program serves. Familiarize yourself with their history, culture, traditions, and values. Assess the degree to which they are accessing services and their satisfaction with services they receive. • Flexibility in adapting methods and skills to match client and community needs and backgrounds. Build and use a network of natural helpers, community informants, and others who are knowledgeable about the different cultural, linguistic, and racial and ethnic groups and groups of girls and young women that your organization or program serves. Translations must be done with great care. Not all terms used in the English language— especially technical words and phrases—have equivalent expressions in other languages. In addition, different dialects may interpret the same term differently. Therefore, DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-19 jurisdictions should periodically engage community residents to assess the accuracy of translated information. Given the distrust that may exist between members of racial and ethnic minority communities and law enforcement personnel, juvenile courts also can consider performing outreach activities that make law enforcement officers more visible in the community and allow more contact with community members. Innovative methods such as neighborhood prosecution can increase trust between community residents and the justice system. For example, the San Diego City Attorney’s Neighborhood Prosecution Unit combines restorative justice practices (discussed earlier) with neighborhood prosecutors who work with police and community partners to develop proactive, community-based solutions to crime problems. Their Downtown Community Court requires misdemeanor offenders who commit crimes downtown to perform community service to repair the harm they have caused. This court is a collaborative effort involving the Downtown San Diego Partnership, San Diego Superior Court, San Diego Police Department, Office of the Public Defender, and other government and community-based organizations. Their Mid-City Community Court focuses on misdemeanor quality-of-life crimes in the mid-city area. A panel that includes community members determines sanctions. Offenders perform community work service, attend rehabilitative and educational programs, and pay fines and fees. The goals of this community court are greater accountability of offenders to the community and greater community involvement in the criminal justice process. Web site: www.sandiego.gov/cityattorney/criminal/npu/index.shtml. Contact information: 619–533–5500, e-mail reception@sandiego.gov, attention: NPU. System Change (DMC factors: differential opportunities for prevention and intervention; differential handling; legislative, policy, and legal factors; indirect effects; and accumulated disadvantage) System change strategies seek to alter the basic procedures, policies, and rules that define how a juvenile justice system operates to address DMC (see exhibit 6). Because such strategies aim to transform the system itself, they have the potential for producing pervasive, fundamental, and lasting change in a system’s ability to respond effectively to minority youth (OJJDP, 2001:37). These strategies are especially critical in jurisdictions where factors influencing minority overrepresentation may be embedded in the cultural, policy, procedural, and legislative framework of the juvenile justice system. Although system change strategies can be tremendously powerful and enduring, they can be challenging to implement for at least two reasons. First, system change often requires coordination among a variety of youth-serving organizations, including child welfare, education, health, and juvenile justice agencies, that may not be accustomed to coordinating and collaborating with each other. Second, some personnel may not understand the need for change and may strongly resist it. Many may be cynical about change or doubt that effective means exist with which to accomplish major system DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-20 change. Others in the system may perceive that proposed changes could undermine values that they hold dear or go against the way they believe things should be done. In addition, different system components may have conflicting goals that pit them against each other. Exhibit 6: Elements of Successful System Change Initiatives To accomplish pervasive, fundamental, far-reaching, and lasting changes, system initiatives must include the following elements (adapted from Hsia and Beyer, 2000): • A “big picture” perspective that system leaders and managers can share. States and local jurisdictions need to be convinced of the benefits of system change. They must envision how they want their juvenile justice system to function, assess how it currently operates, prioritize the changes they want to implement, and persevere over the long term. • Cross-agency partnerships and coordination between stakeholder groups. System change affects juvenile justice professionals, staff from other youth-serving agencies, and members of the community. Therefore, form interagency groups to coordinate planning and implementation of the system change initiative. • Ongoing data collection and effective use of research and evaluation findings. Gather and analyze data to document system problems and identify potential solutions based on available research regarding what may and may not work to reduce DMC. Research data also can provide a sound basis for legislative reforms and funding decisions and establish objective measures of progress for the initiative. • Financial incentives. Set aside adequate funds so that you can provide, if necessary, financial incentives for implementing system change. • Top-down and bottom-up commitment. Agency leaders and staff must both embrace the initiative. Typically, a visionary, persuasive, and consistent “champion” instigates the system change. Then, a change agent usually translates the vision into a realistic and executable plan. Next, implementation generally is best carried out as a team effort. Communications about the system change initiative should flow frequently from the leaders to all staff, and information regarding concerns and frustrations should flow back from staff to the leaders through open forums to maintain their support and commitment to change. Generally, when a jurisdiction initiates a DMC-related system change, it examines the rules by which its juvenile justice system operates to determine if any policies, procedures, or laws place minority youth at a disadvantage. Elements that the jurisdiction may have to review include the following: existing sentencing guidelines, diversion guidelines, minimum standards for equitable treatment and processing of juvenile offenders, detention risk assessments, probation classification systems, release criteria, factors considered in judicial waiver cases, and state and local statutes. Types of system change that can influence DMC include legislative reforms; administrative, policy, and procedural changes; and structured decisionmaking. Legislative Reforms Legislative reforms can be an effective strategy for addressing DMC because they have enormous potential for producing broad-based change in every aspect of the system. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-21 During legislative sessions, it is important to monitor bills concerning juvenile justice issues and children and families to be sure they do not result in statutes that could fuel overrepresentation or bias justice officials’ decisionmaking or reporting (e.g., legal definition of a gang, mandatory juvenile sentencing guidelines, construction of secure facilities). Examples of Laws That Increase Overrepresentation The disparities in federal sentencing laws for crack cocaine and powder cocaine possession and trafficking illustrate how legislation can worsen DMC. Crack cocaine use became prevalent in the mid-1980s and received massive media coverage as a violenceinducing, highly addictive drug plague of inner cities. This media attention led to the quick passage of two federal sentencing laws concerning crack cocaine in 1986 and 1988. The laws created a 100:1 quantity ratio between the amount of crack cocaine and powder cocaine needed to trigger certain mandatory minimum sentences for trafficking and possession. For powder cocaine, a conviction of possessing 500 grams or more with intent to distribute carries a 5-year sentence. For crack cocaine, a conviction of possessing only 5 grams with intent to distribute carries a 5-year sentence. In addition, crack cocaine is the only drug that carries a mandatory prison sentence for first offense possession. A person convicted in federal court of possessing 5 grams of crack cocaine automatically receives a 5-year prison term. A person convicted of possessing 5 grams of powder cocaine probably would receive a probation sentence. Under these laws, crack cocaine users and dealers receive much harsher penalties than users and dealers of powder cocaine. Moreover, most defendants convicted of using and dealing crack cocaine are African American. The result has been a massive sentencing disparity by race, with African Americans receiving longer sentences than the mostly white and Hispanic powder cocaine offenders (The Sentencing Project, no date). Zero tolerance policies are another example of administrative policies that may unintentionally draw more minorities into the system. These types of policies also have been empirically shown to result in disproportionate treatment of minority youth. Zero tolerance laws originated from policies designed to deter students from carrying guns to school and engaging in violence at school. Over time, zero tolerance policies were extended to restrict other behaviors; research indicates that these policies are more likely to exist in predominantly African American and Latino school districts. In addition, minority youths appear to be more likely to be disciplined for minor infractions than their nonminority counterparts (Nellis, 2005; Dunbar and Villarruel, 2003). Examples of Laws That Address Overrepresentation Legislation can also be designed to reduce DMC. In Washington State, the legislature enacted laws to establish standards for decisionmaking at certain stages of the juvenile justice process and also require state agencies to monitor and report annually on how county juvenile courts handle minority youth. The specific legislation includes the following: DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-22 Engrossed Substitute House Bill 1966. Enacted in 1993, the bill mandated that a group of justice officials in the Office of the Administrator of the Courts (the Juvenile Justice Racial Disproportionality Work Group) develop standards for the prosecution of juvenile offenders, review disproportionality in diversion, and review the use of detention in an effort to reduce disproportionality. Prosecutorial standards were adopted in 1995. House Bill 2319 Enacted in 1994, the bill mandated statewide annual monitoring of racial disproportionality in juvenile courts at the county level. The law also provided funds in the 1993–95 biennial budget to “evaluate racial and ethnic disparity within county programs,” including annual reporting on the effectiveness of county-based measures to reduce disproportionality within the state. Further, the law established local juvenile justice advisory committees at the county level to monitor and report annually on proportionality and the effectiveness and cultural relevance of local and state rehabilitative services for juveniles. The committees were also charged with reviewing and reporting on citizen complaints regarding bias or disproportionality within local juvenile justice systems. The committees are required to submit reports annually to the Washington State Sentencing Guidelines Commission, the agency that oversees all aspects of juvenile and criminal sentencing in the state and reports biennially to the legislature. House Bill 2392 Passed in 1996, the bill established an experimental program in two counties to implement and evaluate prosecutor guidelines for reducing racial inequality in the prosecution of juveniles. Two participating prosecutors’ offices developed and implemented a system to track the application of prosecutorial standards in cases brought to juvenile court. The system shows whether the race, gender, religion, or creed of the suspect or victim influenced the decision to prosecute or not prosecute. Administrative, Policy, and Procedural Changes Legislative reforms that significantly alter the way an organization operates often provide the impetus for administrative, policy, and procedural changes that can reduce DMC. In the State of Washington, for instance, the three pieces of legislation described above led to major administrative and procedural changes in the state’s juvenile justice system that were intended to influence DMC. They included the following: adoption of standards for prosecuting juvenile offenders; development of experimental programs implementing prosecutor guidelines to reduce racial inequality in the prosecution of juveniles; a requirement that state agencies supervising youth adjudicated delinquent or convicted in criminal court report annually on minority representation; and establishment of local juvenile justice advisory committees to monitor and report annually on proportionality, and review and report on citizen complaints regarding bias or disparity within local juvenile justice systems (Hsia, Bridges, and McHale, 2004:15–16). DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-23 Structured Decisionmaking One of the most powerful system reforms for reducing and preventing DMC is the implementation of structured decisionmaking using statistical risk classification (i.e., a risk assessment instrument). A risk assessment instrument is an empirically based, standardized, objective instrument for use in evaluating a youth’s background and current situation and estimating the likelihood that the youth will continue to be involved in delinquent behavior (see exhibit 7). Community corrections may use the results of risk assessment to specify the level or intensity of supervision needed; in residential settings, risk assessment results may determine the security level and living unit (Clear and Gallagher, 1983; Wiebush et al., 1995). Exhibit 7: Protocol for Developing and Implementing a Risk Assessment Instrument Gottfredson and Snyder (2005) have identified five steps for developing and implementing a risk assessment instrument: • Step 1: Define the criterion categories of “favorable” or “unfavorable” performance, or “new offense,” or some other event. This step involves defining the behavior or event you will predict and developing procedures for classifying persons based on their performance associated with that behavior or event. Step 1 is of utmost importance because it establishes the standard for selecting predictors and testing the validity of results. In addition, it sets limits to generalization. • Step 2: Select and define the attributes or characteristics on which you may base the predictions. These “predictor candidates” should relate significantly to the criterion categories. • Step 3: Measure the relations between the predictor candidates and the criterion categories in a representative sample of the population for which you will draw inferences. Measure these relations using the appropriate statistical technique (Pearson product moment correlation coefficient, the point biserial correlation, the phi coefficient, or the odds ratio). In most cases, use a random sample from the population to which you will make generalizations to find the representative sample. Any haphazardly identified sample is apt to introduce bias, so procedures for sample selection should ensure that every individual in the population has an equal chance of being included in the sample. • Step 4: Test the prediction procedures in a new sample of the population to verify the relations found in the original sample. Although most researchers omit this verification (referred to as cross-validation), it is a critical step. Without it, one can have little confidence in the utility of a prediction method for any practical application. • Step 5: Apply the prediction method in the types of situations for which the method was developed. You can take this step, provided you supported the stability of predictions in the cross-validation (step 4) and used the appropriate samples. In many juvenile justice systems, practitioners make decisions based on their experience and knowledge of a youth’s background, without using research-based tools. However well intentioned, such clinical predictions can be rife with unintentional racial bias that results in DMC. The absence of structured decisionmaking at any point in the juvenile justice process allows practitioners to base decisions on subjective criteria that may be related to race. For example, Iowa’s assessment research indicated that some officers equated the wearing of certain sports apparel with gang membership, so youth wearing DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-24 such apparel were more likely to be referred to juvenile court instead of diverted (Leiber, 1994). Structured decisionmaking holds the promise of enabling practitioners to objectively classify delinquent youth according to level of risk and to reassess level of risk at different stages in the juvenile justice process. Accurate information about level of risk, in turn, can improve decisionmaking regarding treatment, placement, and court disposition. Statistical risk assessment entails having youth complete a standardized risk assessment instrument, which usually consists of questions related to a small number of factors that research has shown can predict future offending. The screener determines the risk level from the numerical scores assigned to the responses and often divides the summated risk score into categories of low, medium, and high risk. (Some instruments allow the screener to override the instrument’s determination or offer opportunities to mitigate or aggravate the score based on favorable or unfavorable characteristics. Screeners must apply overrides and mitigating and aggravating factors with caution, however, because the value of the screening instrument lies in its objectivity.) Statistical risk assessments are a valuable tool for reducing recidivism because they allow practitioners to accurately reassess the level of risk at various decision points and thus respond more efficiently and effectively to youth in the system. Even structured decisionmaking instruments can contribute to minority overrepresentation unless practitioners take proper care at each step of the development process. An obvious example of pitfalls to avoid is the use of race (and/or ethnicity) as a predictor of recidivism. While recidivism differences may be correlated with race, they are not caused by race. Attributes that are themselves correlated with race (e.g., poverty, school failure, a high proportion of unsupervised time in a day, levels of community disruption, amount of police surveillance in the community) cause the differences in recidivism (Gottfredson and Snyder, 2005). When information on these attributes is unavailable (as is often the case), their predictive power is partially captured by race and invites the erroneous interpretation that race is causally related to recidivism. To address this problem, Gottfredson and Snyder (2005)6 suggest that risk scale developers add a race variable in the early stages of risk scale development but omit this variable from the published instrument. The authors argue that some statistical methods used in the development of risk scales remove the unique (i.e., independent) effect of race from the prediction process. They maintain that unless race is mathematically included in the initial steps of risk scale development, when race correlates with the criterion measure, one cannot remove racial bias from the resulting risk scale; it remains unobtrusively present beneath the surface, influencing each risk scale score. Multnomah County (Oregon) is an excellent example of a jurisdiction that successfully uses a structured decisionmaking instrument to reduce DMC. As in many jurisdictions, minority youth were overrepresented in detention. Introducing an objective screening instrument not only reduced the use of pretrial detention but also substantially decreased the differences between admissions for white and nonwhite youth (Orlando, 1998). (See case study on Multnomah County DMC, later in this chapter.) DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-25 The State of Washington also has implemented a structured decisionmaking instrument (see exhibit 8). The Washington Association of Juvenile Court Administrators and the Washington State Institute of Public Policy jointly developed the Washington State Juvenile Court Assessment, with input from probation line staff, juvenile court administrators, and juvenile justice researchers. The instrument contains highly relevant assessment content and profiles factors that are critical to promoting positive outcomes for youth involved in the justice system. The instrument was designed to accomplish the following: (1) determine a youth’s level of risk for reoffending; (2) identify the risk and protective factors linked to the youth’s criminal behavior so that case managers can tailor rehabilitation to the youth’s assessment profile; (3) develop a case management approach focused on reducing risk factors and increasing protective factors; and (4) allow case managers to determine if targeted factors change as a result of the court’s intervention. Although reducing DMC was not its primary purpose, the instrument imposes a uniform procedure for assessing youth and thus decreases the number of racially biased assessments and recommendations. Exhibit 8: Washington State’s Process for Developing a Risk Assessment Instrument Jurisdictions interested in developing structured decisionmaking instruments may find the process that the State of Washington used to create its State Juvenile Court Assessment (Washington State Institute for Public Policy, 2004) helpful. The Washington Association of Juvenile Court Administrators and the Washington State Institute of Public Policy performed the following 10 steps: 1. Reviewed juvenile delinquency literature, including recidivism prediction literature and instruments, theoretical models for juvenile delinquency, risk and protective factor research, resiliency research, and research on effective juvenile delinquency programs. 2. Drafted a risk assessment instrument based on the literature review. 3. Submitted the draft for review and comment to a group of international experts (Scott Henggeler, Medical University of South Carolina; Mark Lipsey, Vanderbilt University; Patrick Tolan, University of Illinois at Chicago; and David Farrington, Cambridge University, England). 4. Modified the instrument based on reviewers’ comments. 5. Submitted the revised instrument for review and comment to more than 40 juvenile court professionals in the state and convened a series of focus groups with juvenile court professionals. 6. Modified the instrument based on anecdotal information that the state’s juvenile court professionals provided. 7. Delivered a 2-day training to juvenile court professionals on use of the instrument. 8. Pilot tested the instrument in 12 state juvenile courts with 150 youth and modified the instrument based on pilot test results. 9. Implemented the instrument in two stages. Stage 1 entailed administering a prescreen assessment—a shortened version of the full 132-item instrument—to all youth on probation. Stage 2 involved administering the full instrument only to youth assessed as moderate or high risk based on the prescreen. The full instrument generated a profile of risk and protective factors to guide decisionmaking about interventions for these moderate- and high-risk youth. (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-26 Exhibit 8: Washington State’s Process for Developing a Risk Assessment Instrument (continued) 10. Validated the instrument(s). As documented in a report on the instrument’s validity, the assessment produced a valid risk classification, and the risk and protective factors included in the assessment were demonstrated to be empirically associated with recidivism. Therefore, it is appropriate for the courts to use the assessment to place youth in programs that address their risk profiles and to expect that effective programs will reduce risk factors and increase protective factors for these youth. (To access the shorter prescreen assessment and the full Washington State Juvenile Court Assessment, visit the Web site of the Washington State Juvenile Court Administrators at www.wajca.org/risk.cfm.) Case Studies of Jurisdictional Initiatives This section presents case studies of jurisdictional DMC initiatives that have shown encouraging results in reducing minority overrepresentation at various juvenile justice decision points. There have been three major efforts over the last two decades to develop and assess efforts to reduce minority overrepresentation in the juvenile justice system. The first was the OJJDP DMC Initiative initiated in 1991. Under this initiative, OJJDP competitively selected Arizona, Florida, Iowa, North Carolina, and Oregon to receive funds to experiment with approaches to reducing DMC. The Annie E. Casey Foundation established the second in 1992 to demonstrate that jurisdictions can establish more effective and efficient systems to accomplish the purposes of juvenile detention. This initiative was known as the Juvenile Detention Alternatives Initiative (JDAI). The goals of the JDAI were to reduce the number of children unnecessarily or inappropriately detained; to minimize the number of youth who fail to appear in court or reoffend pending adjudication; to redirect public funds toward successful reform strategies; and to improve conditions of confinement. The final effort is that of the Haywood W. Burns Institute (Burns Institute). The Burns Institute leads a group of traditional and nontraditional stakeholders through an intensive data-driven, consensus-based process to reduce the overrepresentation of minority youth in their juvenile detention facilities. Many of the case studies highlighted here are a result of one or more of these initiatives.7 In addition, the initiatives highlighted here are primarily charged with reducing the disproportional representation of minority youth in detention as opposed to the entire juvenile justice system. This irregular distribution is the result of the relatively recent change of the DMC requirement in the JJDP Act of 2002 that modified the definition of DMC from disproportionate minority “confinement” to disproportionate minority “contact.” This broader definition is only now beginning to influence the assortment of programs designed to impact disproportionality in the system. Each case study provides background information on the DMC problem (including identified DMC factors and relevant juvenile justice decision points), a description of strategies used to address the problem, a discussion of available evidence regarding the strategies’ impact on DMC trends, and contact information. Because the research base for DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-27 these initiatives is not sufficiently rigorous, they are described as “developing” initiatives that have a strong theoretical base and for which there is some empirical and anecdotal evidence of effectiveness. These initiatives are, nevertheless, excellent foundations from which other jurisdictions can learn. Bernalillo County, New Mexico Problem Bernalillo County is home to Albuquerque and is located in central New Mexico. According to 2002 census figures, the county had 593,765 residents, 233,565 of Latino origin. In 1999, 72 percent (2,840) of the juveniles arrested in the county were minority youth. This figure is well above the percentage of minority youth in the county’s general population. Detention reform and system change have been a major focus of Bernalillo’s DMC reduction initiative. Detention reform began in earnest in 1998. At that time, the Bernalillo County Juvenile Detention Center was an 80-bed facility with an average daily population of 118 juveniles. At one point in early 1998, the daily population had risen to a high of 143 juveniles, and the Center installed stackable bunk beds to accommodate the additional youth. Because of the severe overcrowding, it was virtually impossible to conduct normal programming, and tension within the facility mounted for both staff and youth. These circumstances drove detention center Director Tom Swisstack and juvenile court Judges Tommy Jewell and Michael Martinez to begin reforming the detention system, with guidance from the Annie E. Casey Foundation. The Foundation provided seed funding for the county to study JDAI pilot sites and plan its own detention reform campaign. Bernalillo established a steering committee of top officials from the juvenile court, the probation department, the district attorney’s office, the public defender’s office, and the Juvenile Detention Center. The committee conducted an extensive review of the county’s current juvenile justice procedures and identified several critical issues. One major problem was the large number of probation violations. Most probation agreements contained dozens of rules and requirements, which made it unlikely that any youth on probation could meet all the terms of the agreements. Curfew violations, failed drug tests, driving without permission, and other minor misbehavior were commonplace. These violations often led officers to revoke probation and return youth to locked detention pending a new court hearing. The system was characterized as a “lobster trap”—it was easy to get in but hard to get out. Another problem was the issuance of an excessive number of bench warrants. The committee found that many youth had plausible excuses for missing their scheduled court dates but were arrested nonetheless and sent to detention until their rescheduled court dates. The committee also found that many youth in detention needed mental health services. According to a 2002 report prepared by the U.S. House of Representatives, Committee on DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-28 Government Reform, Special Investigations Division, 13 of the 14 juvenile detention facilities in New Mexico were housing youth who had serious mental disorders and who had been sent to the facilities for no reason other than to wait for mental health services to become available. The report also indicated that one in seven New Mexico youth in detention was awaiting mental health treatment. In Bernalillo, the figure was even higher—63 percent of the teens referred to the Center suffered from emotional or behavioral health problems. Drawing on the successful experiences of other sites that used the JDAI principles, Bernalillo designed an integrated, community-based continuum of services that consisted of prevention and early intervention services as well as graduated sanctions provided by a cooperative of the county’s education, juvenile justice, and social services organizations. The cooperative’s goal was to intervene early and effectively to help youth stay in school and out of the juvenile justice system. The county developed this new approach specifically to address the issues that surfaced during the procedural review. Bernalillo County, New Mexico: DMC Problem DMC Factors (√ = Identified) Juvenile Justice Decision Points (√ = DMC Identified) • Differential offending • Population at risk • Differential opportunities • Arrest • Differential handling √ • Referral • Legislative, policy, and legal factors √ • Diversion • Justice by geography • Detention √ • Indirect effects • Petition • Accumulated disadvantage • Delinquent findings • Probation √ • Secure corrections • Transfer to adult court Strategies Direct Services The Bernalillo cooperative offers an array of direct services that range from prevention strategies in the public school system, to progressive detention alternatives, to secure detention. Truancy Intervention and Prevention Program In 2001, the Juvenile Detention Center received funding that supported the development of a strong collaboration among the Albuquerque Public Schools (APS), the juvenile justice system, and the community. The need for a Truancy Intervention and Prevention Program (TIPP) was evident in light of several factors. The APS annual dropout rate was DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-29 9.35 percent, compared with the national average of 5 percent. Also, truancy referrals to the children’s court increased 14 percent between 1997 and 1998. (APS defines truancy as 11 full days of excused or unexcused absences per semester.) The Trumbull/La Mesa area was chosen as the pilot site for the TIPP initiative. TIPP receives referrals from two elementary schools. A child is referred when he or she has been absent seven or more times in a semester. The goal is to intervene effectively with children and families, and decrease school absences before the child establishes a pattern of truancy. TIPP staff visit homes and link families with needed services and resources. Within the home and community, TIPP seeks to involve parents in positive interactions that foster healthy attachment to their children, the school, and the neighborhood. In addition, TIPP provides teen mentors who help improve the children’s academic skills, encourage them to attend school regularly, and assist them in overcoming roadblocks to personal and social achievement and growth. Early Interventions Directing juveniles away from further involvement in the juvenile justice system is a priority of the collaborative. To this end, the cooperative implemented several community-based programs for high-risk youth in partnership with the APS, Juvenile Probation and Parole, Children’s Court, ABQ Weed and Seed Projects, Advocacy Inc., Behavioral Health Research Center of the Southwest, Albuquerque Parks and Recreation, Albuquerque Community Centers, Albuquerque Police Department, Bernalillo County Sheriff’s Department, and Albuquerque and Hispano Chamber of Commerce. The partnership provides a wide range of services. These include general education and computer training, substance abuse education and treatment groups, life skills training, employment skills training, mental health services, health and HIV education, alternative sentencing, mentorship, and recreation therapy. Various intervention programs within the Juvenile Detention Center offer education, recreation, substance abuse education and treatment, therapeutic group activities, and community-sponsored activities. Residents who feel that alcohol and other drugs are having a negative impact on their lives can participate in the AYUDA program. Ayuda means “help” in Spanish and is the acronym for “Assisting Youth Using Drugs and Alcohol.” AYUDA provides assessment, counseling, relapse prevention, and case management services; addresses issues including anger management and domestic violence; and uses innovative strategies such as music groups and Tai Chi. AYUDA also makes appropriate referrals for youth who would like to participate in comparable programs upon release. Another program at the Juvenile Detention Center provides both indoor and outdoor recreation activities and other positive activities. This program’s goal is to promote healthy lifestyles. The Center offers art therapy and pet-assisted therapy to improve residents’ morale and behavior while they wait for their hearings. These programs are intended to have beneficial effects that continue after the residents’ release and, therefore, are expected to help decrease recidivism. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-30 Diversion Bernalillo established the Children’s Community Mental Health Clinic (CCMHC), a free­ standing, outpatient mental health clinic for system-involved juveniles and other youth in the community. Rick Miera, an elected member of New Mexico’s state legislature and a substance abuse counselor at the Juvenile Detention Center, championed this innovative diversion strategy. The Center operates CCMHC in collaboration with several health care organizations. The CCMHC service team consists of psychiatrists, nurses, social workers, counselors, and case managers. Available services include evaluation and assessment; individual, family, and group therapy; medication management; substance abuse treatment; case management; and psychological testing. Juvenile justice system practitioners, care providers, parents, and patients can make referrals to CCMHC. Evaluations done at CCMHC are for diagnostic and treatment purposes. A multidisciplinary team designs treatment plans to meet the individual needs of each patient. Detention Alternatives In 1998, Bernalillo was already operating a home detention and electronic monitoring program, but it was used infrequently. On any given day, just four or five youth were enrolled in the program, and no staff had been dedicated to the program. Today, the Community Custody Program (CCP) oversees 30 to 35 young people per day. Each CCP supervisor receives a program manual that outlines four levels of supervision. Some of the youth in CCP also participate in the Youth Reporting Center (YRC) program. This alternative to secure detention offers academic learning, recreation, and other workshops designed to promote and reinforce the acquisition of beneficial skills and socially acceptable behaviors. YRC is located on the grounds of the Juvenile Detention Center and is open from 8 a.m. to 8 p.m. 7 days per week. The Second Judicial District Court or the Children, Youth and Families Department makes referrals in conjunction with the Probation and Parole Office. Training and Technical Assistance None reported. System Change Bernalillo County methodically reorganized its resources, budget, and staff to focus on community-based treatment and innovative policies that cost taxpayers less money. Detention center staff members were assigned to direct services such as electronic monitoring and the day treatment center rather than secure detention. The center hired an expeditor to follow each youth placed there and speed up the cases. Bernalillo’s detention reform leaders meet at 11 a.m. each workday to develop a consensus recommendation for every young person slated for a detention hearing. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-31 Bernalillo’s steering committee also reduced the number of youth detained for violating probation orders or missing court dates. The committee developed and implemented a new sanctions grid prescribing the steps officers should follow before revoking probation. Probation officers are now required to bring any request to revoke probation to the deputy probation director. The result has been a 50-percent drop in the number of youth placed in locked detention for probation violations. Bernalillo also reformed the system to reduce the number of youth confined on bench warrants for failure to appear at their court hearings. Reforms included sending notices in Spanish and making reminder calls. In 2000, before the reforms were implemented, the county admitted 542 youth to detention on bench warrants and typically confined them until their court dates. The county “cleaned house,” thus reducing the number of warrants. That is, Bernalillo reviewed old warrants and used the National Crime Information Center index to identify superfluous warrants. Legislative changes also helped reduce the number of youth in detention. New Mexico changed its children’s code; these changes included reducing petition times and not allowing children younger than age 11 to be detained. Bernalillo County, New Mexico: DMC Strategies DMC Strategies (√ = Implemented) Direct Services • Prevention √ • Early intervention √ • Diversion √ • Alternatives to secure detention √ • Alternatives to secure corrections • Advocacy Training and Technical Assistance • Cultural competency training and program development • Staff recruitment System Change • Legislative reforms √ • Administrative, policy, and procedural changes √ • Structured decisionmaking tools √ Impact on DMC Trends By improving system operations and offering a greater array of programmatic choices, Bernalillo County has made great strides in reducing the number of youth overall— including minority youth—who come into contact with the juvenile justice system. Much DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-32 of this success has come from decreasing the reliance on locked detention. The population of the Juvenile Detention Center has declined from an average daily population of 113 in 1999 to 47 in 2004. Over a 6-year period (from 1999 to 2005), the number of youth booked declined from 4,726 to 3,892, and the number of youth booked and held in confinement dropped from 2,494 to 1,549. Fewer youth are going to detention and those who are placed in detention stay fewer days. The average length of stay fell from 15 days to 13 days. In addition, the juvenile court issues fewer bench warrants because reforms have increased the number of youth who appear for their scheduled court dates and give youth a second chance when they offer reasonable excuses for missing court. Between 1999 and 2005, the number of youth who failed to appear in court decreased from 831 to 480. Alternatives to detention such as the community custody program and the YRC have been successful. Only 5 percent of participants commit new misdemeanors, and another 17 percent are returned to detention after breaking program rules. A large majority (76 percent) complete the programs successfully and appear in court as scheduled without committing new offenses. Regarding the direct effect of reforms on reducing DMC, there is still work to be done. The number of minority youth booked in 2005 decreased to 2,426 (62 percent) compared with 2,840 (72 percent) minority youth booked in 1999. Although this figure is disproportionately high when compared with the number of youth in the general population, it represents a step in the right direction. More data are being evaluated regarding Bernalillo’s full continuum of juvenile justice services and their impact on DMC trends. Contact Information Doug Mitchell Bernalillo County JDAI Coordinator Bernalillo County Juvenile Detention Center 5100 Second Street NW Albuquerque, NM 87107 505–761–6600 demitchell@bernco.gov Resources Children’s Defense Fund. “Chapter 6: Youth development: Poverty and the pipeline to prison.” State of America’s Children, 2005. Downloaded December 15, 2005, at www.childrensdefense.org/site/DocServer/greenbook_2005.pdf?docID=1741 Lubow, B., and Mariscal, R. 2005. “Bernalillo County designated model site: Results, innovation, culture of reform, and regional partnerships cited.” JDAI News: Newsletter of the Juvenile Detention Alternatives Initiative. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-33 Mendel, D. 2003. “And the walls keep tumbling down: A demonstration project has come and gone, but detention reform continues to gather steam.” AdvoCasey. Anne E. Casey Foundation. www.aecf.org/publications/advocasey/spring2003/walls/walls.htm. U.S. House of Representatives, Committee on Government Reform, Minority Staff (Special Investigations Division). 2004. “Incarceration of youth who are waiting for community mental health services in the United States.” Prepared for Rep. Henry A. Waxman and Sen. Susan Collins. Washington, D.C. Available online at http://democrats.reform.house.gov/Documents/20040817121901-25170.pdf Cook County, Illinois Problem In December 1992, the Annie E. Casey Foundation launched a multiyear, multisite project called the Juvenile Detention Alternatives Initiative (JDAI). The purpose of JDAI was to demonstrate that interagency collaboration and data-driven policies and programs can reduce the number of youth behind bars without sacrificing public safety or court appearance rates. Cook County was one of the original JDAI sites. The Cook County JDAI project targeted minority youth in detention. In 1996, more than 90 percent of the county’s average daily detention population of 694 was minority youth (Hoyt et al., 2002). The initiative began with the establishment of a Disproportionate Representation Committee (DRC). A presiding judge chaired the DRC. Committee members included representatives of court services and probation, the public defender’s office, juvenile detention, and, at times, the police department and the state attorney’s office. The committee also included representatives of community-based groups such as the Westside Association for Community Action and advocacy organizations such as Northwestern University Law School’s Children and Family Justice Center. The committee developed specific goals related to DMC, including an agenda for research into reasons for overrepresentation of minorities, and revisions of policies, procedures, practices, and programs, as necessary. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-34 Cook County, Illinois: DMC Problem DMC Factors (√ = Identified) Juvenile Justice Decision Points (√ = DMC Identified) • Differential offending • Population at risk • Differential opportunities • Arrest • Differential handling √ • Referral • Legislative, policy, and legal factors √ • Diversion • Justice by geography • Detention √ • Indirect effects • Petition • Accumulated disadvantage • Delinquent findings • Probation • Secure corrections • Transfer to adult court Strategies Direct Services Cook County’s approach for addressing DMC focused on developing a range of community-based alternatives to detention and changing the way the system dealt with youth who failed to appear in court, violated probation, or committed other minor infractions. The county established a continuum of detention alternatives, including home confinement, electronic monitoring, shelters, and evening reporting centers (ERCs). The most innovative programs are the ERCs, where youth eligible for secure detention receive intensive, individualized supervision during the high-crime hours (3:00 p.m. to 9:00 p.m.) on weekdays and Saturdays. The ERCs are an interim measure to reduce the risk of reoffending. Another goal is to ensure that the youth appear in court while allowing them to continue their schooling and remain at home. Nonprofit communitybased organizations located in high-referral neighborhoods operate the ERCs and employ staff primarily from those neighborhoods. Each ERC supervises up to 25 youth and maintains a 1:5 staff-to-youth ratio. At the ERCs, youth participate in recreational activities, receive tutoring and counseling, and are given referrals for other communitybased services. The length of participation in the ERC program ranges from 5 to 21 days, while delinquency proceedings are pending. The cost of the ERC program is approximately $33 per youth per day. The Sheriff’s Work Alternative Program (SWAP) is another community-based program designed to reduce the detention population by creating alternative sanctions for youth who violate the conditions of probation. Before SWAP was created, youth charged with violating the terms of their probation would spend an average of 21 days in detention. SWAP provides a “detention step-down” alternative to this costly practice. Now, the DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-35 court can choose to detain a youth for 7 days in secure detention and then order that he or she participate in SWAP. Training and Technical Assistance The DRC decided to implement cultural diversity and communication training as a way to help staff in juvenile justice agencies become more culturally sensitive. With assistance from a consultant provided by the Annie E. Casey Foundation, Cook County developed a train-the-trainer model to create within each agency the organizational capacity for delivering training. The model focuses on reducing racial stereotypes and bias in decisionmaking and emphasizes how staff influence the decisionmaking process. Since 1997, six system components—the police department, the judiciary, the public defender’s office, the juvenile detention center, the state’s attorney’s office, and juvenile probation and court services—have trained staff. System Change The DRC developed a new risk assessment tool for detention intake, with a point system that takes into account race-related considerations. For example, factors that previously increased the points for minority youth (such as whether a youth had two parents versus another guardian who could hold the youth accountable) were reweighted. The designers of this tool sought both to improve the outcomes for all youth and to be more sensitive to the life contexts of minority youth by reducing the points allocated for criteria more commonly associated with minority arrestees (such as prior police contacts and whether a youth came from a single-parent household). As a result, prior arrests were not counted in the scoring of the detention screening instrument, only prior referrals to court. Cook County also addressed a gap in the quality of defense that minority youth received through the indigent defense system. In July 1996, the Public Defender’s Office established a Detention Response Unit consisting of two paralegals who interview detained youth prior to their custody hearing to obtain their version of the instant offense. During these interviews, the paralegals give the youth important information about what to expect at the hearing and how to conduct themselves. The paralegals relay the results of the interviews, including risk assessment factors, to the public defender conducting hearings that day. In addition, the paralegals verify community ties and contact families to stress the importance of their presence at the custody hearing. The paralegals also suggest detention alternatives that the lawyers can recommend to the court. In summary, by adding a larger social narrative to the court process—one that resource-poor public defenders often cannot marshal—the Detention Response Unit provides judges with a better understanding of each youth’s circumstances and a greater array of communitybased alternatives where youth can be supervised. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-36 Cook County, Illinois: DMC Strategies DMC Strategies (√ = Implemented) Direct Services • Prevention • Early intervention √ • Diversion √ • Alternatives to secure detention √ • Alternatives to secure corrections • Advocacy √ Training and Technical Assistance • Cultural competency training √ • Cultural competent program development • Staff recruitment System Change • Legislative reforms • Administrative, policy, and procedural changes √ • Structured decisionmaking tools Impact on DMC Trends Between 1996 and 2000, the Cook County DMC initiative reduced the average number of youth in secure detention on any given day from 694 to 478. Considering that more than 90 percent of youth admitted and held in detention were members of minority groups, these figures represent a 31-percent drop in the minority detention population (Hoyt et al., 2002). The proportion of minority youth in detention, however, remained roughly the same before and after the initiative. From 1996 to 1999, African Americans continued to be overrepresented at almost every stage of the Cook County juvenile justice system, including the arrest stage and “each of the stages that directly lead to post-trial confinement in a secure detention or correctional facility” (Illinois Criminal Justice Information Authority, 2003). Disagreements regarding the very definition of the DMC problem among DRC members appear to have led the committee to change the initiative’s focus from addressing overrepresentation to creating a more effective system “for all kids.” In 2003, the Illinois Juvenile Justice Commission launched two new local DMC projects in a renewed effort to reduce minority overrepresentation in Cook County: one in Cook County’s South Suburbs and one in Chicago’s Lawndale community. Both these communities are working with experts from the W. Haywood Burns Institute to reduce the number of minority youth who are detained unnecessarily in their communities. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-37 Lawndale’s new DMC project specifically targets youth who are detained for failure to appear in court. On July 1, 2005, Lawndale’s probation officers and adjudicators introduced new administrative procedures designed to help juveniles make their court date. These procedures include: (1) providing additional written reminders of court dates to youth and their families; (2) delaying the issuance of bench warrants when questions exist about why a youth has failed to appear in court; (3) facilitating quicker post-warrant return of youth who fail to appear in court; and (4) expediting assessment and treatment of youth with drug problems (since drug abuse may contribute to absenteeism/failure to appear in court). Lawndale’s DMC Advisory Board has also launched a number of community-based delinquency prevention programs, such as a parenting program and a job readiness program, which the committee believes will have a long-term impact on minority delinquency rates (Lawndale DMC Advisory Committee, 2006). The South Suburban DMC (SSDMC) project targets youth who are detained for technical violations of probation. The project’s advisory board is now creating an ambitious wraparound services program for such youth, and it eventually plans to offer them a wide array of community-based alternatives to detention, including an evening reporting center, an electronic monitoring program, anger management services, counseling services, community health referrals, family group counseling, peer courts, and balanced and restorative justice programs (SSDMC Advisory Committee, 2006). The Burns Institute DMC pilot sites in Illinois have not yet been evaluated. The sites expect to begin formal evaluations to assess their effectiveness at reducing DMC in the targeted areas sometime in 2006. The results from these evaluations will also assist the Illinois Juvenile Justice Commission in determining if the Burns Institute model should be expanded to additional sites. Contact Information Mike Rohan Director Juvenile Probation and Court Services Circuit Court of Cook County 1100 South Hamilton Avenue Chicago, IL 60612 312–433–6575 Carlus Dee Gully DMC Coordinator Youth Network Council 321½ South Sixth Street Suite 200 Springfield, IL 62701 Phone: 217–522–2663 dgully@youthnetworkcouncil.org DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-38 Resources Hoyt, E.H., Schiraldi, V., Smith, B.V., and Ziedenberg, J. 2002. Pathways to Juvenile Detention Reform: Reducing Racial Disparities in Juvenile Detention. Baltimore, MD: Annie E. Casey Foundation. Illinois Criminal Justice Information Authority. 1999. “Juvenile work program provides alternatives to detention.” On Good Authority 3(1):1–4. Illinois Criminal Justice Information Authority. 2003. A Study of Disproportionate Minority Representation in the Cook County Juvenile System. Chicago, IL: The Juvenile Justice Commission. Lawndale DMC Advisory Committee. 2006. “Lawndale detention reduction strategies.” An unpublished progress report submitted to the Illinois Juvenile Justice Commission. South Suburban DMC Committee. 2006. “SSDMC response to juvenile justice commission.” An unpublished progress report submitted to the Illinois Juvenile Justice Commission. Hillsborough County, Florida Problem In 1990, the Florida State Supreme Court created a special committee to investigate racial and ethnic bias in Florida’s judicial system. This committee, known as the Racial and Ethnic Bias Study Commission, sponsored several ambitious research studies examining the treatment of minorities throughout the state’s justice system. The commission’s final report on the juvenile justice system (Florida State Supreme Court Racial and Ethnic Bias Commission, 1991) found that “minority juveniles are being treated more harshly than nonminority juveniles at almost all stages of the juvenile justice system, including: arrest; referral for formal processing; transfer to the adult criminal system; secure detention prior to adjudication; and adjudication and commitment to traditional state-run facilities.” Based largely on the commission’s findings, OJJDP chose Florida to participate in the two-phase DMC initiative. Phase 1 entailed researching the DMC problem, and phase 2 focused on implementing DMC interventions. During phase 1, Florida’s project team conducted additional research examining racial bias in juvenile justice decisionmaking throughout the state. The research uncovered especially severe minority overrepresentation in Hillsborough County. Primarily for this reason, the state selected Hillsborough as the site of its DMC demonstration intervention. According to phase 1 data, the odds in Hillsborough County of African American youth being involved in the juvenile justice system were the same as, or higher than, those of the state’s juvenile population as a whole at almost every decision point in the juvenile justice process (Caliber Associates, 1996). DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-39 African American youth in the county were more than twice as likely as white youth to have an initial referral to the juvenile justice system. During the 16-month research period, 11 percent of the county’s 19,000 African Americans ages 10 to 17 were referred to the system compared with 5 percent of the white juvenile population. African American youth represented 22 percent of the study population8 but made up 39 percent of youth referred to the system and 63 percent of youth eventually committed. Once referred, African American youth were twice as likely to be detained. When not detained, a white youth would more often receive a “no petition” intake recommendation, whereas an African American would more often receive a “petition” recommendation. Statistics for Hispanics (15 percent of the study population) showed a pattern of referrals and dispositions very similar to that for whites. Florida’s DMC project team focused its efforts on expanding the range of treatment, rehabilitation, and diversion programs available to minority youth. Representatives from the following organizations worked on the DMC initiative: the Juvenile Justice Work Group (including the Tampa-Hillsborough Urban League, Inc., and the Agency for Community Treatment Services, Inc.), the Hillsborough Juvenile Welfare Board, the Juvenile Assessment Center (JAC), and the local Health and Rehabilitative Services. The Hillsborough County DMC initiative included three major components: a coalition of concerned citizens and a core planning group, a system for diverting nonserious minority youth from formal involvement with the juvenile justice system, and cultural sensitivity training for professionals in the system. Hillsborough County, Florida: DMC Problem DMC Factors (√ = Identified) Juvenile Justice Decision Points (√ = DMC Identified) • Differential offending • Population at risk • Differential opportunities √ • Arrest • Differential handling √ • Referral √ • Legislative, policy, and legal factors • Diversion √ • Justice by geography • Detention √ • Indirect effects • Petition √ • Accumulated disadvantage • Delinquent findings • Probation • Secure corrections • Transfer to adult court Strategies Direct Services The DMC initiative implemented three direct service strategies to address DMC. First, the initiative recruited 32 service providers from the coalition membership to offer DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-40 traditional services, such as education assistance, job training and placement, mental health and substance abuse treatment, crisis management, and juvenile arbitration. The initiative required all providers to offer a culturally sensitive environment for treatment programs, supervise community service hours, and provide a structured setting for youth. The DMC initiative especially sought to involve providers not traditionally used in the juvenile justice system to broaden the base of participating service providers from the county’s minority community. Second, the initiative established a juvenile civil citation program that permitted the JAC or the state attorney’s office to issue a sanction of up to 40 hours of community service to a juvenile charged with a nonserious offense without taking the youth into custody. The program sought to divert youth from the “front door” of the justice system at the point of initial contact. This diversion was developed in response to data showing that African American youth were overrepresented at the initial point of contact and the overrepresentation gradually increased at later decision points. The program was intended to have a ripple effect—by reducing the number of African American youth at the point of initial contact, the program would decrease the number of African American youth at subsequent stages in the juvenile justice process. Third, the initiative developed an assessment and diversion advocacy system for minority youth. The basic process was as follows: the state attorney’s office identified youth who were qualified to go to court by virtue of their offenses and prior record. In cases involving minority youth who met certain eligibility requirements, the state attorney’s office, instead of petitioning the court, made an exception and diverted the youth to the minority diversion program. For such youth, a JAC specialist drafted a diversion plan that focused on addressing individual and family functioning issues and made arrangements with appropriate treatment providers. The JAC specialist and a case manager on loan from the Florida Department of Juvenile Justice closely supervised the youth as they completed treatment and other sanctions. The maximum caseload for the minority diversion program was set at 50. Training and Technical Assistance Phase 1 findings indicated that intake workers and members of law enforcement might have biases contributing to DMC. Therefore, development and implementation of a plan to provide cultural competency training was part of the Hillsborough DMC initiative. The plan directed the Urban League to hire a full-time training specialist to design a cultural competency curriculum for juvenile justice and law enforcement personnel and provide them with cultural competency training. To prepare the training plan, the training specialist was directed to follow African American youth through the juvenile justice system and into the community, experiencing firsthand the situations and problems that these youth face. The training specialist also received cultural competency training, conducted research needed to design the training curriculum, and received technical assistance from researchers at the University of South Florida to ensure that the best available methods for cultural competency training were used. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-41 System Change The main system change strategy was the formation of a core planning group to coordinate the planning and implementation of the county’s DMC initiative. Members included representatives of the general community, the minority community, private sector and government service providers, the state attorney’s office, law enforcement, the school system, and the juvenile justice system. Group responsibilities included the following: • Reach consensus on the primary issues to be addressed. • Develop a plan to involve the minority community in resolving the DMC problem. • Involve key neighborhood leaders and utilize the media to build community interest in action. • Reach consensus on goals and methods with a coalition of service providers and develop an action plan. • Implement the plan and evaluate progress. The Hillsborough County State Attorney quickly emerged as a strong leader within the core planning group, and his office’s commitment to offer white and minority youth the same opportunities for rehabilitation and diversion was crucial to the success of the Hillsborough program. Hillsborough County, Florida DMC Strategies (√ = Implemented) Direct Services • Prevention • Early intervention √ • Diversion √ • Alternatives to secure detention • Alternatives to secure corrections • Advocacy √ Training and Technical Assistance • Cultural competency training and program development √ • Staff recruitment System Change • Legislative reforms • Administrative, policy, and procedural changes √ • Structured decisionmaking tools DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-42 Impact on DMC trends The Hillsborough DMC initiative was implemented for a little more than 1 year—from January 1994 to February 1995. An analysis of the 1994–95 delinquency data suggests that the pilot project had a profound effect on the county’s juvenile justice system (Nelson, 1996). Although the DMC initiative was designed to decrease the number of minority youth judicially processed and confined, data indicated that the entire county juvenile justice system benefited from the initiative. An analysis of delinquency cases handled judicially in 1994–95 showed a 4-percent decrease when compared with cases the previous fiscal year. Judicially handled cases involving black youth decreased 5 percent, which made 1994–95 the first year in more than a decade that the number of such cases declined. Judicially handled cases involving white youth decreased 6 percent. In contrast, cases handled nonjudicially in 1994–95 increased 26 percent. Nonjudicially handled cases involving black youth increased 30 percent and nonjudicially handled cases involving white youth increased 21 percent. An analysis of cases referred to the Juvenile Alternative Sanctions Program (JASP) yielded additional positive findings. JASP cases may be handled judicially or nonjudicially. The total number of cases referred judicially to JASP decreased 6 percent compared with the previous year. For both black and white youth, cases referred judicially to JASP decreased 8 percent. Prior to 1994–95, the total number of cases judicially referred to JSAP had increased for each of the past 5 years. From 1993–94 to 1994–95, the number of cases nonjudicially referred to JASP increased 182 percent. Cases involving black youth referred nonjudicially to JASP increased 261 percent, and cases involving white youth increased 134 percent. An analysis of cases placed on community control (a community control sanction requires that the youth appear before a judge) showed an 18-percent decrease in 1994–95, compared to the previous year. For the first time in 13 years, community control cases involving black youth decreased (22 percent). For the first time in 5 years, such cases involving white youth decreased (16 percent). Although the number of delinquency commitment cases increased 2 percent in 1994–95, there was a 1-percent decrease for cases involving minority youth. In addition, the overall number of cases transferred to adult court decreased 16 percent in 1994–95, compared with the previous year. There was an 8-percent decrease in such cases involving black youth and a 32-percent decrease in cases involving white youth. Ted Tollett, director of the Florida Department of Juvenile Justice’s Office of Research and Planning, attributes much of the Hillsborough County demonstration project’s success to the vigorous support of the local state’s attorney, who helped ensure that minorities and whites were “given the same breaks” and opportunities within the juvenile justice system. Without the support and buy-in of this critical leader, it is unlikely that the initiative would have produced significant results. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-43 Contact Information Florida’s DMC Initiative was discontinued in early 1995. The state’s subsequent efforts to institutionalize this approach by implementing similar projects in other sites in the state failed because of budget cutbacks. Individuals interested in obtaining additional information about the project should contact: Ted Tollett Director Office of Research and Planning Suite 100, Knight Building 2737 Centerview Drive Tallahassee, FL 32399–3100 850–487–1213 ted.tollett@djj.state.fl.us Resources Caliber Associates. 1996. Evaluation of the Disproportionate Minority Confinement (DMC) Initiative: Florida Final Report. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Florida State Supreme Court Racial and Ethnic Bias Commission. 1991. Where the Injured Fly for Justice: Reforming Practices Which Impede the Dispensation of Justice to Minorities in Florida. Executive Summary. Tallahassee: Florida State Supreme Court. Nelson, R. 1996. Hillsborough County Disproportionate Minority Confinement Pilot Project Evaluation. Tallahassee: Florida Department of Juvenile Justice. King County, Washington Problem In 2000, King County embarked on an effort to reduce overrepresentation of minority youth in the juvenile justice system, building on lessons learned from DMC initiatives in Multnomah County, Oregon, and Santa Cruz County, California. King County (Seattle and the surrounding area) became the pilot site for the W. Haywood Burns Institute for Juvenile Justice Fairness and Equity, an organization working to reduce the overrepresentation of minority youth in juvenile detention systems nationwide. The first step was to establish a team of local juvenile justice stakeholders who would support and actively participate in the reform process. The team convened an advisory board consisting of political leaders, representatives from all key agencies in the juvenile justice system (police officers, probation department personnel, public defenders, prosecutors, and judges), and interested community members (e.g., business owners, youth). The advisory board based decisions on a consensus-based approach, which meant that everyone in the group had to approve a recommendation before it could be adopted. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-44 This approach ensured that all participating agencies signed on to all agreements and made implementation more likely and smoother. The advisory board collected and analyzed local juvenile crime data by race, offense, location, and time. The Seattle Police Department took the lead and set a tone of cooperation and openness. At the onset of this process, the department agreed to collect and share data on race and ethnicity of youth for both pedestrian and traffic stops. As part of the initial data-gathering in 1999, the board hired youth from three target areas in Seattle to conduct community mapping. The 1999 data indicated that African American youth were disproportionately committing drug offenses and Asian American youth were disproportionately committing vehicle and theft offenses. Further analysis showed that juvenile crime spiked at 3 p.m. and remained high until 10 p.m. African American youth constituted 39 percent of juveniles in detention but only 9 percent of all youth in the jurisdiction. Finally, the length of stay in detention was 14 percent longer for African American youth than white youth. These findings led the advisory board to develop a plan calling for specific steps related to programming and police practices. The plan was implemented in three phases: phase 1 involved reducing disparities at arrest; phase 2 focused on detention; and phase 3 addressed dispositional decisions. King County, Washington: DMC Problem DMC Factors (√ = Identified) Juvenile Justice Decision Points (√ = DMC Identified) • Differential offending √ • Population at risk • Differential opportunities √ • Arrest √ • Differential handling √ • Referral • Legislative, policy, and legal factors • Diversion • Justice by geography • Detention √ • Indirect effects • Petition • Accumulated disadvantage • Delinquent findings √ • Probation √ • Secure corrections • Transfer to adult court DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-45 Strategies Direct Services To reduce DMC, King County made a number of significant changes to existing direct services. The first strategy entailed expanding the alternatives to secure detention and closely monitoring the ethnic and racial composition of the participant population. Initially, the new alternatives increased DMC because white youth were disproportionately sent to these alternatives. Soon after, however, the county rectified this problem by monitoring the racial and ethnic composition of participating youth. Now, the county can ensure that the number of minority youth sent to detention alternatives is, at the very least, in proportion to the number of minority youth in secure detention. The county has also implemented several new intervention services for offender youth, including evidence-based treatment programs specifically for minority youth with dual diagnoses. Reclaiming Futures grants have helped to support this effort. Training and Technical Assistance All current and newly hired juvenile justice personnel are made aware of King County’s intention to reduce and prevent DMC. An overall climate of cultural competency is encouraged and expected throughout the system. In accordance with the Burns Institute process for reducing DMC, King County also hired a part-time site coordinator to guide stakeholders in implementing DMC-related system reforms and to ensure that all involved in the DMC initiative remain focused on agreed-upon priorities. (Without a site coordinator’s help, jurisdictions seeking to address DMC often lose momentum because of changes in stakeholders, inconsistent approaches, and short attention spans.) The site coordinator works closely with the Burns Institute to lead stakeholders through the reform process. The site coordinator meets twice yearly with other Burns Institute site coordinators to share best practices. The Burns Institute Web site notes that “. . . the Burns Institute disproportionality focus has been woven into and is a major component of a broader Seattle juvenile justice reform effort called the Juvenile Justice Operational Master Plan.” System Change King County implemented a number of system changes to address DMC. Strategies included modifying the police booking protocol, developing a detention risk assessment instrument, adding an expeditor, reducing failure-to-appear warrants, and using the guidance of a DMC site coordinator to ensure consistency throughout the system. In accordance with new booking procedures, all police officers in the county carry a wallet-sized card that lists the basic detention intake criteria and gives the detentionscreening unit’s phone number. Police are required to call the detention center first before bringing in a youth. If a youth does not meet the criteria, the officer cannot transport him or her to detention. This change in procedure and the close partnership between detention personnel and police officers have been associated with a decrease in inappropriate use of detention for youth. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-46 King County layered on top of the basic detention intake criteria a risk assessment instrument (RAI) that takes into account risks such as offense severity and prior offense history. The RAI is used to generate an objective recommendation to the court on whether a young person should continue to be detained within 24 hours from the time an officer brings a youth to the detention center. If detention is considered appropriate, an expeditor is charged with ensuring that the judicial system responds quickly. The expeditor contacts the youth’s probation officer and creates a report for the judge. Every 2 weeks, the judge, probation officer, and expeditor conduct a detention review to revisit why the youth was placed in detention initially and why he or she has remained there. Further, the detention review also helps to determine whether the youth still needs to be in detention and whether it is appropriate to seek alternatives. King County also has taken steps to reduce the number of failure-to-appear warrants. An analysis of 2001 data showed that 49 percent of African American youth received failureto-appear warrants, compared with 29 percent of white youth. As a warrant and detention history builds, it becomes more likely that the youth will be detained again. The county gathered information to find out why young people failed to appear and how to address the issue. The county has instituted a reminder call system and is collecting data to determine whether this new initiative is useful. King County, Washington: DMC Strategies DMC Strategies (√ = Implemented) Direct Services • Prevention • Early intervention √ • Diversion • Alternatives to secure detention √ • Alternatives to secure corrections √ • Advocacy √ Training and Technical Assistance • Cultural competency training and program development √ • Staff recruitment √ System Change • Legislative reforms • Administrative, policy, and procedural changes √ • Structured decisionmaking tools √ DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-47 Impact on DMC Trends King County has made great strides in terms of decreasing the use of secure detention. From 1998 to 2004, the average daily population in secure detention dropped 44 percent. This reduction was the same for both minority and white youth. In 1998, before the implementation of reforms, 180 to 190 youth were in detention on a daily basis. In 2004, the average daily number was 100 to 110 youth. What this means for African American youth is that their numbers in detention on a daily basis dropped from 69 to 38 during this period. Although these figures are still disproportionate given the racial composition of the overall population, they represent a step in the right direction. In addition, since implementation of the King County DMC initiative, African American youth are more often being referred to alternative programs rather than to detention. By 2004, 40 percent of youth in detention alternatives were African American, compared with only 27 percent in 2000. Contact Information Teddi Edington Juvenile Detention Reform/DMC Coordinator King County Superior Court 1211 East Alder Street Seattle, WA 98122 Phone: 206–205–9539 Fax: 206–205–9408 Teddi.Edington@METROKC.GOV Resources Building Blocks for Youth Initiative. 2005. No Turning Back: Promising Approaches to Reducing Racial and Ethnic Disparities Affecting Youth of Color in the Justice System. Profile of Seattle (Washington) Site. W. Haywood Burns Institute for Juvenile Justice Fairness and Equity. Retrieved from Burns Institute Web site (www.burnsinstitute.org) on January 27, 2006. Mesa County, Colorado Problem Mesa County is rural, but it is the most populated county on the western slope of the Colorado Rocky Mountains, bordering Utah. It is the only county in the state’s 21st Judicial District, and its major city, Grand Junction, is the only metropolitan area on the western slope. The county’s total juvenile population (ages 10 to 17) is 14,211. The county’s largest minority youth population is Hispanic (12.3 percent in 1994 and 13.2 percent today). Overall, minority youth make up 15.5 percent of the total county youth population ages 10 to 17, and 15.5 percent is the target goal for achieving parity for minority youth in the justice system. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-48 In 1994, the State Advisory Group (SAG) funded a study of DMC in the state’s juvenile justice system, finding disproportionate minority confinement in the five targeted jurisdictions that accounted for 80 percent of the state’s juvenile population. This study showed a high rate of overrepresentation of minority youth in Mesa County at the secure detention and commitment decision points. Specifically, the data indicated that 26 percent of youth in secure detention and 66 percent of committed youth were minorities. These figures were well above the percentage of minority youth in the general population. The SAG presented the study findings to the committees that coordinate alternatives to detention and wraparound services in the county. The committees decided that the best way to address this overrepresentation was to implement early interventions to prevent minority youth from being placed in secure facilities. To ensure a comprehensive, working coalition for this initiative, the Mesa County Minority Over Representation (MCMOR) Committee was formed under the leadership of the judicial district’s chief judge. The committee was charged with coordinating the planning and implementation of the county’s efforts to address DMC. Committee members include representatives from the court, law enforcement, schools, human services, diversion, youth services providers, and the community. The committee conducted public meetings in churches and other community sites to establish strong support; community members supplied the refreshments for these meetings. The coalition jelled after a systemwide commitment was made to objectively collect accurate data at all juvenile justice decision points. The committee continues to meet monthly. The district court directed the committee’s first project, which was funded under Title II (Formula Grants). The committee contracted an attorney to interview all system decisionmakers (e.g., law enforcement officers, prosecuting attorneys, judges, probation officers, school officials, service providers), collect and analyze available data to identify possible DMC contributing factors, and recommend ways to address these factors. This process in 1995 found many contributing factors, which the committee continues to update and clarify through an annual analysis of DMC data, including the need for cultural sensitivity training, lack of knowledge of court/legal procedures, language barriers, poverty, unemployment, academic failure, dropping out of school, frustration with the system, lack of communication, overworked legal counsel/public defenders, and large caseloads. The major recommendations were to review policies and practices that could indicate a lack of skills and knowledge related to working with culturally diverse minority youth, including possible bias in the statewide detention screening tool and to provide an intervention that could reduce the number of minority juveniles committed to the state’s Division of Youth Corrections as a result of technical probation violations rather than serious offenses. Since 2000, the committee has compiled an annual statistical report to track changes in DMC. Data from 2000 to 2004 showed that while minority youth are not overrepresented at the point of arrest, they continue to be overrepresented at decision points after arrest that are more restrictive in nature—detention screening, detention placement, and commitment. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-49 Mesa County, Colorado: DMC Problem DMC Factors (√ = Identified) Juvenile Justice Decision Points (√ = DMC Identified) • Differential offending • Population at risk • Differential opportunities √ • Arrest • Differential handling √ • Referral √ • Legislative, policy, and legal factors √ • Diversion √ • Justice by geography • Detention √ • Indirect effects • Petition • Accumulated disadvantage √ • Delinquent findings • Probation • Secure corrections √ • Transfer to adult court Strategies Direct Services The initial study found that many minority youth failed to comply with court orders and probation conditions, and this noncompliance resulted in secure detention and commitment decisions. The committee established the Minority Family Advocacy Project (MFAP) in coordination with other juvenile justice agencies in Mesa County. MFAP provides case management and advocacy services—delivered by experienced advocates—to help minority youth and their families increase their chances of successfully navigating the juvenile justice system and accessing needed treatment. Mesa County Partners, a nonprofit youth services agency, houses the advocates. Currently, there are two advocate coordinators who provide tracking, advocacy, language interpretation, and case management for 70 high-risk minority youth per year. Each advocate maintains an average caseload of 20 youth. MFAP advocates give families an extra set of eyes and ears to ensure that youth and family members understand the requirements of the court and other agencies and to help them through the court/legal process. MFAP advocates assist youth in completing diversion and other court-ordered requirements. MFAP advocates accompany families to court hearings and other meetings and stay in constant contact with them to encourage follow-through with agencies and services. In addition, families can contact the advocates to ask questions or discuss concerns. MFAP services also include mentoring and incentive components for high-risk youth. The advocates recruit, train, match, and supervise mentors who are paired with minority youth. All youth and mentors are required to spend a minimum of 3 hours per week, oneon-one, working on goals that the youth’s treatment team set. The goals can involve DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-50 activities ranging from tutoring and job hunting to positive recreational activities. Mentors may also work with siblings who may be at risk of delinquent behavior. The incentive component for high-risk youth requires that each participant define and work on accomplishing goals related to education, self-enhancement, and a community project. With the school district, MFAP recently began co-facilitating a gender-specific group for at-risk Hispanic girls to offer them appropriate supervised activity, mentoring, and developmental support. The primary referral source to MFAP is the court-appointed assessment team, which screens all requests for secure detention. The team refers all minority youth to MFAP, whether they are detained or receive alternative services. If a youth receives alternative services, the MFAP advocate ensures that the youth complies with the pretrial release conditions. If a youth is detained, the advocate attends the detention hearing and offers services to the family. If the family is interested, the advocate sets up an intake meeting to complete paperwork and begins designing a treatment plan and assembling a treatment team for the family. If the family is not interested initially, the advocate gives the family a business card and offers the family a chance to receive services at a later date. MFAP services have increased the number of youth released to nonsecure supervision and reduced lengths of stay in detention. Secondary referral sources to MFAP are diversion, probation, and the school district. As previously stated, advocates attend case-planning meetings and assist youth in fulfilling the requirements of diversion contracts and probation, thus decreasing revocations and the possibility of detention, commitment, or greater involvement with the juvenile justice system. In addition, MFAP advocates—in cooperation with the school district—work with minority youth who have been suspended or expelled to ensure that their educational needs are still met, as state law requires. As a result, the overall number of suspensions and expulsions has decreased. Training and Technical Assistance In accordance with needs identified in the MCMOR Committee’s initial study, cultural competency training was provided to the broad spectrum of juvenile justice decisionmakers, including law enforcement officers, juvenile probation officers, and district court judges. Over the years, most agencies have incorporated this training into their standard training agenda. On request, the committee will help any agency obtain training or will provide the training, if possible. The committee will recommend training when data trends suggest it is needed. System Change The core system change strategy was the formation of the MCMOR Committee to coordinate the planning and implementation of Mesa County’s efforts to address DMC. The court’s leadership has been critical to the committee’s success. The committee gained credibility and support by collecting accurate data, involving the judicial system, preparing useful annual reports, and maintaining a broad-based membership. The committee’s main responsibilities were to: DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-51 • Educate the community about the DMC issue. • Continue data collection and tracking activities to better study overrepresentation in the juvenile justice system; and understand why particular decisions are made and whether current intervention strategies are affecting DMC trends. • Ensure that minority youth have equal access to alternatives to detention and formal prosecution, such as the intense diversion program that the district attorney’s office operates. • Continue to offer cultural competency training for all parts of the county’s criminal justice, human services, and school systems and to suggest policy improvements, as needed. • Work with the entire community and appropriate agencies to enhance prevention and early intervention resources for minority youth and their families. An MCMOR subcommittee continues to review randomly selected juvenile justice cases to compare the circumstances, offenses, and criminal histories of youth and determine whether different decisions are made for white and minority youth whose cases are comparable. The full MCMOR Committee discusses the results of these reviews and, as needed, considers new ways to improve the juvenile justice system. Mesa County, Colorado: DMC Strategies DMC Strategies (√ = Implemented) Direct Services • Prevention • Early intervention • Diversion √ • Alternatives to secure detention • Alternatives to secure corrections • Advocacy √ Training and Technical Assistance • Cultural competency training and program development √ • Staff recruitment √ System Change • Legislative reforms • Administrative, policy, and procedural changes √ • Structured decisionmaking tools DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-52 Impact on DMC Trends Countywide efforts to improve racial disparities in the juvenile justice system appear to be producing the desired beneficial effects at many juvenile justice decision points (see table 2). In 1998, the overall figures were promising, although they continue to fluctuate somewhat since the initial DMC reduction strategies were implemented. The most profound effect appears at the detention decision point. The percentage of minority youth in detention has decreased 65 percent, dropping from 26 percent in 1996 to a low of 9.1 percent in 2004. The percentage of minority youth in secure commitment has also declined dramatically (63 percent), dropping from 66 percent in 1996 to 24.1 percent in 2004. Unfortunately, it is not possible to judge definitively how the DMC strategies have affected the arrest and probation decision points because baseline data are lacking for minority youth arrested and placed on probation. Table 2: DMC Trends in the Juvenile Justice System, Mesa County, Colorado % Minority Representation Juvenile Justice Decision Points FY96 FY00 FY01 FY02 FY03 FY04** Minority Youth Population* 15.0 15.5 15.5 15.5 15.5 15.5 Arrest N/A 14.8 14.6 15.1 16.0 15.6 26.0 17.4 24.8 15.5 24.2 9.1 Probation N/A 15.7 11.6 10.8 14.3 26.5 Secure Corrections 66.0 21.3 15.8 28.6 20.7 24.1 Referral Diversion Detention Petition Delinquency Transfer to Adult Court * The minority population in Mesa County consists almost entirely of Hispanics. Detention caps were put in place statewide. ** These figures suggest that the Mesa County DMC initiative is making progress toward its goal of addressing minority overrepresentation, but evidence exists that the task in Mesa has not yet been completed: at various juvenile justice decision points, minority youth are represented at a higher rate than minority youth in the general county population.9 Minority youth were overrepresented at probation (26.5 percent) and secure commitment (24.1 percent) in 2004. The committee is working on plans to reduce DMC at these decision points. The committee also is tracking school data trends (i.e., suspensions, expulsions, and dropout and graduation rates). Constant improvements in the collection of data on issues such as gender differences and length of stay will enhance the analysis of impact and support the development of further interventions. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-53 Contact Information Joe Higgins, Executive Director Luke Archuleta, Advocate Coordinator Mesa County Partners 1169 Colorado Avenue Grand Junction, CO 81501 Phone: 970–245–5555 Fax: 970–245–7411 larchuleta@mesapartners.org Sources This summary is based on grant applications and progress reports submitted to the Colorado Division of Criminal Justice, presentations made by the MCMOR Project, and interviews with Joe Higgins, Executive Director, Mesa County Partners. Multnomah County, Oregon Problem To reduce and prevent DMC, Multnomah County undertook a series of reforms that focused on detention and emphasized the use of data-based strategies. The overall objectives were to ensure that decisionmaking was fair and equitable and that the juvenile justice system’s resources were culturally competent, accessible, and appropriately used for all youth from all racial and ethnic groups. Reforms were instituted beginning in 1994. Multnomah County, home to Portland, Oregon, has a population of more than 670,000 residents, with 76.5 percent white, 5.7 percent black, 5.7 percent Asian, 7.5 percent Latino, and 1 percent Native American. In 1994, prior to DMC-related reforms, Latinos represented 6 percent and African Americans 10 percent of the total youth population in the county. In 1990, Latino youth were more than twice as likely to be detained as white youth (34 percent versus 15 percent), and Asians, African Americans, and Native Americans were detained at rates that were 47 to 60 percent higher than those for white youth. In 1993, an analysis of Oregon’s data for phase 1 of OJJDP’s DMC initiative found that detention processing and police referrals were the major factors contributing to overrepresentation in the county. In that same year, it was noted that the secure detention facility consistently operated at full capacity—96 beds. Pre-adjudicated Multnomah youth, including 70 minority youth, occupied approximately 80 percent of the beds. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-54 Multnomah County, Oregon: DMC Problem DMC Factors (√ = Identified) Juvenile Justice Decision Points (√ = DMC Identified) • Differential offending • Population at risk • Differential opportunities • Arrest • Differential handling √ • Referral • Legislative, policy, and legal factors • Diversion • Justice by geography • Detention √ • Indirect effects • Petition • Accumulated disadvantage • Delinquent findings • Probation • Secure corrections • Transfer to adult court Strategies Direct Services The main direct services strategy was to develop a range of detention alternatives including shelter care, foster homes, home detention, and a day reporting center located in the communities of minority youth. These programs were established to divert youth from detention and from being returned to custody for violating the terms of their release. The initiative decided to use local providers with the hope that they would be more accessible to the youth and their families and more knowledgeable about available social services. For instance, Volunteers of America (VOA), one of the local providers, helped implement the county’s home confinement program and tailored supervision based on each juvenile’s level of risk. Youth who did not need detention but could not be trusted to comply with the rules of house arrest wore an electronic monitor. VOA workers called or showed up at the house or school unannounced several times per day to closely supervise youth who represented minimal risk and therefore were not required to wear an electronic monitor. Another detention alternative was the Reception Center, a nonprofit, community-based organization funded by government crime prevention and private foundations. The Reception Center provided a home-like setting where police could take youth who had committed nondetainable infractions. In the past, youth who broke curfews or were runaways—many of whom were Latino or African American—were sent to secure detention. The center gave professional care and support to children and youth up to 18 years of age until they could transition back to their families, foster homes, or alternative family placements. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-55 Augmenting the defense representation of minority youth was yet another direct service strategy. Public defenders or appointed counsels who are frequently overburdened usually represent minority youth. Youth represented by these overburdened defenders often receive more restrictive outcomes than those who retain counsel. To make the legal system a more level playing field for minority youth, trial assistants were added. These advocates helped the public defenders identify the strengths that could be tapped in each case—within the youth themselves, their families, and the community (including community-based programs). The trial assistants also ensured that the information was disseminated equally among all interested parties—the youth, the defender, and the prosecution. Specifically, they reviewed lists of the young people scheduled for preliminary hearings and obtained discovery about the youth and their charges prior to the hearing, in support of the lawyers. They attended pretrial placement planning meetings, where all stakeholders—the district attorney, the defender, probation personnel, and others—decided the appropriate placement of youth scheduled for preliminary hearings. The trial assistants played an important role in ensuring that the defender had equal access to information that the prosecutor and probation personnel already possessed. This enhanced representation significantly increased the use of alternatives for youth who would have otherwise faced secure detention. Training and Technical Assistance The county made concerted efforts to increase cultural competency throughout the juvenile justice system, in accordance with its priority of reducing and preventing racial disparities. First, the county launched a recruiting and hiring initiative to increase the cultural diversity of staff in all components of the juvenile justice system. In addition, all current staff received educational materials on culturally sensitive principles and practices. Second, Multnomah trained all interested parties in ways to reduce disparities. On a broader level, the county offered training to inform decisionmakers and service providers about the issues related to overrepresentation. Through more targeted training, police officers received information on existing detention alternative initiatives. Now, every police officer carries a card listing the actions that he or she should take for specific levels of infraction. This use of objective criteria promotes fair and equitable decisions about youth. System Change As part of the DMC initiative, Multnomah County also developed, implemented, and evaluated a culturally sensitive risk assessment instrument (RAI) to guide admissions decisions. An interagency team of representatives from the judiciary, public defenders, prosecutors, probation and detention system personnel, school officials, and researchers designed the instrument. The team carefully evaluated individual RAI elements in the context of cultural and racial/ethnic differences. For example, instead of relying on criteria such as “good family structure,” which could be biased against minority youth, the instrument relies on “verifiable community ties,” which determines whether there is an adult willing to ensure the youth’s appearance in court. Likewise, instead of DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-56 exclusively using “school attendance” as a mitigating factor, the concept was expanded to include “productive activity.” The county formed a new detention intake team to gauge the internal reliability of the RAI. The team consisted of six to seven intake workers supervised by a detention intake manager and a pretrial placement coordinator. Each day, the placement coordinator and intake manager reviewed the cases of all youth in detention (not just those admitted that day), considering their risk assessment scores, case status, and suitability for communitybased alternatives. The placement coordinator completed daily quality control checks to ensure that youth were processed expeditiously and that staff faithfully adhered to the RAI when making decisions. Another key reform strategy in Multnomah was to implement other sanctions first to reduce the number of parole violators in detention. Twenty to thirty percent of all admissions to detention were probation or parole violators. Prior to reforms, the county detained many youth who violated probation; such decisions often were made inconsistently and without taking into account the risks that the juvenile posed or his or her needs. For example, a review of the data showed that some probation violators did not appear in court because they did not receive the notification. Strengthening communication between the court and families reduced the need to send such youth to detention. To support this reform strategy, the county created a sanctions grid that, along with detention, provided alternatives for probation violators, such as a warning or community service. Further, supervisors and/or an alternative placement committee had to approve decisions to detain probation violators. Multnomah County, Oregon: DMC Strategies DMC Strategies (√ = Implemented) Direct Services • Prevention • Early intervention • Diversion • Alternatives to secure detention √ • Alternatives to secure corrections • Advocacy √ Training and Technical Assistance • Cultural competency training and program development √ • Staff recruitment √ System Change • Legislative reforms • Administrative, policy, and procedural changes √ • Structured decisionmaking tools √ DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-57 Impact on DMC Trends After Multnomah County implemented the systemwide reforms, detention trends improved overall. The county created a more effective juvenile detention system, reduced the overall detention population, and measurably decreased disproportionality. Three factors have been cited as contributing to the county’s success: the design and implementation of an RAI, the development of alternatives to detention, and the provision of training to raise awareness about overrepresentation. While it is difficult to determine which strategic reforms have been most successful, the implementation of the RAI has had an obvious impact. As previously stated, in 1993 the average daily detention population in the county was 96. When the RAI was introduced in 1995, the average number of youth in detention on any given day began decreasing; by 2000, the figure had dropped to 33, which included remanded youth (18 were pre- and post-adjudicated Multnomah delinquent youth and the remainder were youth remanded to the youth court awaiting adjudication). In 1994, an arrested African American or Latino youth had a 10 to 11 percent greater likelihood of being detained at some point in his or her case than an arrested white youth. By 1995, this disparity was reduced to 6 percentage points, and by 2000 the differentials dropped to 3 percentage points for African Americans (12 percent versus 9 percent for white youth) and to 2 percentage points for Latino youth (11 percent). In addition to significantly reducing disproportionality, Multnomah County’s detention reform strategies have helped to decrease the number of detention admissions. By 2000, the number of youth admitted to detention dropped by half for all youth (from 1,107 in 1994 to 478 in 2000) and by half for both African American and Latino youth. Raising awareness about overrepresentation and addressing the issue of “the right kid, at the right place, at the right time” continue to be the challenges in Multnomah County. Mental health cases in detention, many of which involve minority youth, have increased 10 percent. It has been proposed that the increase is a consequence of shifting the responsibility for overseeing youth with mental health concerns from Child Protective Services to County Health Services. This shift has proven problematic in terms of expediting detention processing in general and reducing DMC in particular. One of the strengths of Multnomah’s approach had been the efficient processing of cases involving detained youth. In the past, Child Protective Services received at 8 a.m. a list of all detainees from the previous night. If a youth in its care made the list, a case manager met with the youth and the district attorney by 11:30 a.m. and had a placement plan to give the judge by 1 p.m. With the recent change of hands, the system is not working as efficiently. As a result, some youth are detained longer than necessary. The task is to educate the new stakeholders in County Health Services. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-58 Contact Information Rick Jensen National Model Site Administrator Annie E. Casey Foundation and Multnomah County Detention Reform Initiative Department of Community Justice Juvenile Services Division 1401 N.E. 68th Avenue Portland, OR 97213 Phone: 503–988–5698 rick.k.jensen@co.multnomah.or.us Resources Multnomah County Department of Community Justice. 2005. Multnomah County Report on Minority Overrepresentation. 2005. Created for distribution to the Multnomah County Systemwide Stakeholders. The Sentencing Project. 2000. Reducing Racial Disparity in the Criminal Justice System: A Manual for Practitioners and Policymakers. Schiraldi, V., and Ziedenberg, J. 2002. “Reducing disproportionate minority confinement: The Multnomah County Oregon success story and its implications.” Center on Juvenile and Criminal Justice. Washington, DC: Justice Policy Institute. Accessed August 26, 2005, at www.cjcj.org/pubs/portland/portland.html. Stanfield, R. No date. The JDAI Story: Building a Better Juvenile Detention System. Pathways to Juvenile Detention Reform, series overview. Baltimore, MD: Annie E. Casey Foundation. Villarruel, F.A., Walker, N.E., Minifee, P., Rivera-Vazquez, O., Peterson, P., Perry, K. 2002. ?Donde Esta la Justicia? A Call to Action on Behalf of Latino and Latina Youth in the U.S. Justice System. Executive Summary. East Lansing, MI: Institute for Children, Youth and Families. Michigan State University. Santa Cruz County, California Problem Santa Cruz County was one of the early sites for the Annie E. Casey Foundation’s Juvenile Detention Alternatives Initiative (JDAI) and is now one of four national model sites for that initiative. Located south of San Francisco, the county has a population of 250,000, with 65.5 percent of residents reporting non-Hispanic white, 27 percent Latino, 3 percent Asian, and 1 percent African American. Of the youth population, more than 33 percent are of Latino origin. In 1994, Latino youth accounted for 76 percent of the population in juvenile hall, California’s version of detention. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-59 A task force composed of the chief probation officer and representatives from the county’s Latino Strategic Planning Collaborative and the Latino Affairs Commission conducted a system-by-system review of the county’s juvenile justice system. The research indicated that minority youth brought to detention were more likely to have more serious offense histories and more risk factors than white youth. Two of the risk factors noted were living in single-parent homes and low-income households. While officials from the Santa Cruz probation department acknowledged that there was an obvious need to improve social and economic conditions in the communities, they wanted to focus on problems that were under the department’s direct control. The department identified many problems of this kind, including multiple points of subjective rather than objective decisionmaking; clients encountering barriers to service or lack of access; many examples of cultural insensitivity; and unnecessary delays in the court process, which contributed to longer stays in detention. The Santa Cruz probation department formed a core working group composed of Latino and other community members, justice system representatives (e.g., police officers, prosecutors, judges), and others from youth-serving agencies. Staff from the Youth Law Center in San Francisco provided technical assistance to guide the working group in its review. The working group examined the juvenile justice process to identify the decision points where minority overrepresentation was most pronounced or where the most minority youth were affected. The working group reviewed this information and developed a work plan to address DMC in the juvenile justice system and a checklist to keep the plan on track. The working group then inventoried the local continuum of services and reviewed each element for cultural competency. Underlying all aspects of the DMC initiative was an emphasis on promoting collaboration between judicial system professionals, the community, and the families of youth on probation. Hiring a coordinator for the initiative helped to unify the needs and requirements of the seemingly disparate groups. Santa Cruz County, California: DMC Problem DMC Factors (√ = Identified) Juvenile Justice Decision Points (√ = DMC Identified) • Differential offending • Population at risk • Differential opportunities √ • Arrest • Differential handling √ • Referral • Legislative, policy, and legal factors • Diversion √ • Justice by geography • Detention √ • Indirect effects • Petition √ • Accumulated disadvantage • Delinquent findings • Probation • Secure corrections • Transfer to adult court DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-60 Strategies Direct Services Santa Cruz implemented a number of reforms that strengthened the juvenile justice system’s ability to quickly move youth out of detention and into detention alternatives and/or placements and programs. One of the direct service strategies was providing diversion programs for Latino youth. A survey of the probation department’s services indicated that Latino youth generally did not participate in diversion programs; instead, the courts used more traditional processing, including detention. Diversion programs often were unsuccessful for Latino youth who participated, leading to a diversion failure that was due in part to the probation department’s inability to provide culturally appropriate interventions. In response, the county established additional programs that were better suited to the needs of Latino youth. The first diversion option was a neighborhood accountability board that enlisted volunteers from the youth’s home community and the victims of crime to assist the youth in repairing the harm his or her behavior had caused. The second program was a partnership between probation, police, schools, and a community-based agency, which included family-strengthening services and competency-building opportunities for youth. Developing culturally appropriate interventions was a key to reforms, more than doubling the number of youth diverted to the new, more targeted programs and preventing youth from moving more deeply into the juvenile justice system. A second direct service strategy was expanding the detention alternatives program and adding a community-based agency as a partner in the probation department’s home supervision and electronic monitoring programs. The probation department considers home supervision and electronic monitoring to be useful and effective alternatives to detention. However, the number of Latino youth in the programs was small. These detention alternatives required parental involvement and supervision—the keys to successful home detention. Major barriers to Latino families in using the programs included language differences, transportation problems, and confusion about court processes and the families’ responsibilities to the court. To remove these hurdles, the community-based agency explained the court system to the families and supported the parents in responding to the court’s expectations. A third direct service strategy developed by the county was a culturally appropriate family preservation program that focused on family strengths and the development of family directed service plans. The probation department’s goal was to foster an atmosphere of trust and cooperation. A user survey and the addition of parent advocates helped to identify service-related barriers and improve relations. The department changed the tone of some of the formal communications and included evening and weekend hours to accommodate working families. To further strengthen communication, the probation department made it a goal to have a Spanish-speaking staff member at every stage in the juvenile justice process. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-61 Training and Technical Assistance The Santa Cruz Probation Department took steps to decrease bias in all areas of personnel, including staff recruitment, hiring, and training. To accomplish this task, the department developed and implemented a cultural competency staffing plan and appointed a cultural competency coordinator to oversee the plan. The goal of the plan was to establish guidelines to ensure that the department hired bilingual and culturally competent key staff to provide services to a culturally diverse client population. The Spanish-language capability was particularly useful at the intake and case management stages and helped to expedite movement of youth out of detention and back to their families. Specifically, the department’s objective was to have at least as many Latino or Spanish-speaking staff as the proportion of such youth in the detention center. The purpose is to ensure that when Latino youth enter the juvenile justice system, their families can talk to an intake worker or other juvenile justice professional who speaks Spanish. System Change Santa Cruz undertook a major system change strategy to develop a culturally competent juvenile detention screening instrument. All key stakeholders were involved in developing the instrument. The Santa Cruz instrument was based on a set of quantifiable risk elements free of criteria that could create unintentional racial biases. The elements included: (1) seriousness of current charge; (2) prior adjudications; (3) current legal status; (4) prior court, detention, and placement history; and (5) other jurisdiction-specific factors. The instrument also provided an override option, but any override was carefully monitored for racial disparities in its use. The structured decisionmaking procedure divided youth into three categories that enabled intake personnel to make appropriate detention decisions: eligible for immediate full release, eligible for placement in nonsecure alternatives, and eligible for placement in secure detention. Care was taken to eliminate factors on the Detention Risk Assessment Instrument that described personal characteristics of the youth and to rely, instead, on factors that could be objectively proven. The probation department generated monthly outcome reports, by ethnicity, to measure whether the instrument was accurately predicting success in the detention alternatives programs as measured by court appearances and lack of rearrest. Further, Santa Cruz created a structured approach for responding to probation violations that took into account degree of risk. A range of sanctions and systems for matching violations with appropriate responses and tracking those responses by ethnicity helped to ensure fair and equitable treatment for probation violators. The most pronounced system changes were those made to the detention system itself. When the county began weekend intake procedures, many more minimum- and mediumrisk youth were released in a more timely fashion. This particular reform had the effect of immediately reducing the number of youth detained. Santa Cruz also made available more detention alternatives in remote parts of the county where police and probation previously had limited options other than detention. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-62 Santa Cruz County, California: DMC Strategies DMC Strategies (√ = Implemented) Direct Services • Prevention • Early intervention • Diversion √ • Alternatives to secure detention √ • Alternatives to secure corrections √ • Advocacy Training and Technical Assistance • Cultural competency training and program development √ • Staff recruitment √ System Change • Legislative reforms • Administrative, policy, and procedural changes √ • Structured decisionmaking tools √ Impact on DMC Trends Countywide strategies to reduce racial disparities and improve detention and disposition practices within the juvenile justice system appear to be producing the desired beneficial effects. Since the county implemented the DMC initiative, the number of Latino youth in detention has decreased yearly. In 1997 and 1998 (before the DMC initiative), Latino youth made up only 33 percent of the general youth population but accounted for 64 percent of the daily juvenile detention population. This figure dropped to 53 percent in 1999 (after the launch of the DMC initiative), 50 percent in 2000, and 49.7 percent in 2001. Before the DMC initiative, the Santa Cruz disproportionate rate index value for Latino youth in detention was 1.9. The index value dropped to 1.4 by 2001 and has remained at that level through 2005. Expressed in actual numbers, in 1998, 33 Latino youth were in detention on any given day; by the end of 2005, the average daily population of Latino youth in detention was 9. New efforts are focusing on substance-use-related offenses. In the county, 45 percent of youth going to juvenile court are in court on a substance-use-related offense (Community Action Board of Santa Cruz County, 2005). Further, 67 percent of youth going to court have a documented drug and alcohol use, abuse, or dependency problem. Reforms that the county juvenile probation department has spearheaded as part of the Reclaiming Futures Initiative (supported by the Robert Wood Johnson Foundation) have involved evaluating and improving the systems that deliver drug and alcohol treatment to county youth. Recent innovations include the addition of an evening treatment center for substance-abusing probation violators, located in a segment of the county that refers the DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-63 majority of Latino youth to detention. At the evening center, youth can receive evidencebased treatment approaches in lieu of detention. This innovation has further reduced racial disparities in detention. Contact Information Judith Cox Chief Probation Officer Santa Cruz County Probation P.O. Box 1812 Santa Cruz, CA 95061–1812 Phone: 831–454–3800 E-mail: judy.cox@co.santa-cruz.ca.us Shalinee Hunter Corrections Consultant Corrections Planning and Programs Division Corrections Standards Authority 600 Bercut Drive, Suite A Sacramento, California 95814 Phone: 916–322–8081 shalinee.hunter@cdcr.ca.gov Information about the Santa Cruz County DMC Initiative can also be found online at the following Web sites: Santa Cruz County Probation: http://sccounty01.co.santa-cruz.ca.us/prb/index.asp Annie E. Casey Foundation: http://www.aecf.org/initiatives/jdai/ Resources Building Blocks for Youth Initiative. 2005. No Turning Back: Promising Approaches to Reducing Racial and Ethnic Disparities Affecting Youth of Color in the Justice System. Santa Cruz, CA. Community Action Board of Santa Cruz County, Inc. 2005. The Status of Poverty in Santa Cruz County—2005. Accessed November 7, 2005, at www.cabinc.org. Cox, J. 2000. “Addressing disproportionate minority representation within the juvenile justice system.” Building Blocks for Youth, Santa Cruz, CA. Accessed November 7, 2005 at www.buildingblocksforyouth.org/issues/dmc/addressingdmc.html. Hoyt, E.H., Schiraldi, V., Smith B.V., and Ziedenberg, J. 2002. Pathways to Juvenile Detention Reform: Reducing Racial Disparities in Juvenile Detention. Baltimore, MD: Annie E. Casey Foundation. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-64 Villarruel, F.A., Walker, N.E., Minifee, P., Rivera-Vazquez, O., Peterson, S., Perry, K. 2002. ?Donde Esta la Justicia? A Call to Action on Behalf of Latino and Latina Youth in the U.S. Justice System. Executive Summary. East Lansing, MI: Institute for Children, Youth and Families. Michigan State University. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-65 Endnotes 1. The Blueprints for Violence Prevention (Blueprints) project, an initiative to identify effective violence prevention programs, is operated by the Center for the Study and Prevention of Violence (CSPV) at the University of Colorado at Boulder. The project has identified 11 prevention and intervention programs that meet a strict scientific standard of program effectiveness, and another 18 programs as promising programs. CSPV bases program effectiveness on an initial review by, and a final review and recommendation from, a distinguished advisory board, comprising seven experts in the field of violence prevention. 2. The OJJDP Model Programs Guide and Database (MPG) gives comprehensive descriptions of approximately 200 evidence-based programs encompassing the entire youth services continuum, from prevention to aftercare. The MPG ranks programs as follows: Exemplary. When implemented with a high degree of fidelity, these programs demonstrate robust empirical findings using a reputable conceptual framework and an evaluation design of the highest quality (experimental or quasi-experimental). Effective. When implemented with sufficient fidelity, these programs demonstrate adequate empirical findings using a sound conceptual framework and an evaluation design of the highest quality (experimental or quasi-experimental). Promising. When implemented with minimal fidelity, these programs demonstrate promising (but perhaps inconsistent) empirical findings using a reasonable conceptual framework and a limited evaluation design (single group pre/post-test) that requires causal confirmation using more appropriate experimental techniques. The MPG contains detailed reviews of most of the jurisdictional DMC initiatives summarized at the end of the chapter. It is located at www.dsgonline.com/mpg2.5/mpg_index.htm. 3. Risk and protective factors are neither causes nor cures. Rather, they are statistical predictors with a strong theoretical link to delinquency. 4. The online version of the DMC database will be searchable by DMC factor. 5. OJJDP’s Model Programs Guide and Database contains literature reviews of more than 17 types of prevention programs. Access the MPG at www.dsgonline.com/mpg2.5/mpg_index.htm. 6. See Gottfredson and Snyder (2005) for a more detailed description of this procedure. 7. The Burns Institute is currently working intensively with 10 local jurisdictions to reduce the overrepresentation of minority youth in their juvenile justice systems. Although a few of these sites are included in this section, the Institute could not provide standardized evaluation data for any of the sites with which they are working at the time this manual was published. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-66 8. The study population was composed of approximately 62,000 records for individual youth ages 10 to 17 who entered the juvenile justice system during the 16-month period between April 1991 and August 1992. 9. The minority population in Mesa County consists almost entirely of Hispanics. References Aniskiewicz, R., and Wysong, E. 1990. “Evaluating DARE: Drug education and the multiple meanings of success.” Policy Studies Rev. 9:727–747. Bazemore, G., Seymour, A., and Rubin, T. 2000. Victims, Judges, and Juvenile Court Reform Through Restorative Justice Reform. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Victims of Crime. Bry, B.H., McKeon, P., and Pandina, R. 1982. “Extent of drug use as a function of the number of risk factors. Journal of Abnormal Psychology 91:273–79. Building Blocks for Youth Initiative. 2005. No Turning Back: Promising Approaches to Reducing Racial and Ethnic Disparities Affecting Youth of Color in the Justice System. Washington, DC: Building Blocks for Youth. Accessed May 15, 2006, at www.buildingblocksforyouth.org. Bukoski, W.J., and Amsel, Z. (Eds.) 1994. Drug Abuse Prevention: Sourcebook on Strategies and Research. Westport, CT: Greenwood Publishing Company. Caliber Associates. 1996. Evaluation of the Disproportionate Minority Confinement (DMC) Initiative: Florida Final Report. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Center for Substance Abuse Prevention (CSAP). 2002. Finding the Balance: Program Fidelity and Adaptation in Substance Abuse Prevention. Washington, DC: Substance Abuse and Mental Health Services Administration. Community Action Board of Santa Cruz County, Inc. 2005. The Status of Poverty in Santa Cruz County—2005. Accessed November 7, 2005, at www.cabinc.org. Cox, J. 2000. “Addressing disproportionate minority representation within the juvenile justice system.” Building Blocks for Youth, Santa Cruz, CA. Accessed November 7, 2005, at www.buildingblocksforyouth.org/issues/dmc/addressingdmc.html. Cross, T.L., Bazron, B.J., Dennis, K.W., and Isaacs, M.R. 1989. Toward a Culturally Competent System of Care: A Monograph on Effective Services for Minority Children Who Are Severely Emotionally Disturbed. Washington, DC: CASSP Technical Assistance Center. Georgetown University Child Development Center. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-67 Davis, K. 1997. Exploring the Intersection Between Cultural Competency and Managed Behavioral Health Care Policy: Implications for State and County Mental Health Agencies. Alexandria, VA: National Technical Assistance Center for State Mental Health Planning. Debruyn, L., Chino, M., Serna, P., and Fullerton-Gleason, L. 2001. “Child maltreatment in American Indian and Alaska Native communities: Integrating culture, history, and public health for intervention and prevention.” Child Maltreatment 6(2):118–129. Devine, P., Coolbaugh, K., and Jenkins, S. 1998. Disproportionate Minority Confinement: Lessons Learned From Five States. Bulletin. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Ellis, R., Klepper, T., and Sowers, K. 2001. “Building a foundation for effective intervention: Understanding Hispanic juveniles and their families.” Journal for Juvenile Justice and Detention Services 16(1):37–52. Federle, K.H., and Chesney-Lind, M. 1992. “Special issues in juvenile justice: Gender, race, and ethnicity.” In Juvenile Justice and Public Policy: Toward a National Agenda, I.M. Schwartz (ed.). New York, NY: Lexington Books. Florida State Supreme Court Racial and Ethnic Bias Commission. 1991. Where the Injured Fly for Justice: Reforming Practices Which Impede the Dispensation of Justice to Minorities in Florida. Executive Summary. Tallahassee: Florida State Supreme Court. Gottfredson, D.M., and Snyder, H.N. 2005. Mathematics of Risk Classification: Changing Data into Valid Instruments for Juvenile Courts. Report. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Hamparian, D., and Leiber, M. 1997. Disproportionate Confinement of Minority Juveniles in Secure Facilities: 1996 National Report. Champaign, IL: Community Research Associates. Hawkins, J.D., Catalano, R.F., and Miller, J.Y. 1992. “Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention.” Psychological Bulletin 112(1):64–105. Hoyt, E.H., Schiraldi, V., Smith, B.V., and Ziedenberg, J. 2002. Pathways to Juvenile Detention Reform: Reducing Racial Disparities in Juvenile Detention. Baltimore, MD: Annie E. Casey Foundation. Hsia, H.M., and Beyer, M. 2000. System Change Through State Challenge Activities: Approaches and Products. Bulletin. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-68 Hsia, H.M., Bridges, G.S., McHale, R. 2004. Disproportionate Minority Contact: Year 2002 Update. Summary. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Hsia, H.M., and Hamparian, D. 1998. Disproportionate Minority Confinement: 1997 Update. Summary. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Illinois Criminal Justice Information Authority. 2003. A Study of Disproportionate Minority Representation in the Cook County Juvenile System. Chicago, IL: The Juvenile Justice Commission. Isaacs, M., and Benjamin, M. 1991. Toward a Culturally Competent System of Care, Programs Which Utilize Culturally Competent Principles. Volume II. Washington, DC: Georgetown University Child Development Center, CASSP Technical Assistance Center. King, M.A., Sims, A., and Osher, D. 2001. How Is Cultural Competence Integrated in Education? Washington, DC: American Institute of Research. Krisberg, B., Currie, E., Onek, D., and Wiebush, R.G. 1995. “Graduated sanctions for serious, violent, and chronic juvenile offenders.” In A Sourcebook: Serious, Violent, and Chronic Juvenile Offenders. J.C. Howell, B. Krisberg, J.D. Hawkins, and J.J. Wilson (eds.).Thousand Oaks, CA: Sage Publishers. Krisberg, B., Neuenfeldt, D., Wiebush, R., and Rodriguez, O. 1994. Juvenile Intensive Supervision: Planning Guide. Program Summary. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Latessa, Edward J. 2004. “From theory to practice: What works in reducing recidivism?” In State of Crime and Justice in Ohio: Corrections. Candace Peters. Columbus, Ohio: Ohio Office of Criminal Justice Services, pp. 170–171. Lawndale DMC Advisory Committee. 2006. “Lawndale detention reduction strategies.” An unpublished progress report submitted to the Illinois Juvenile Justice Commission. Leiber, M.J. 1994. “A comparison of juvenile court outcomes for Native Americans, African Americans, and Whites.” Justice Quarterly 11(2):257–277. Leiber, M.J., and Stairs, J.M. 1999. “Race, contexts, and the use of intake diversion.” Journal of Research in Crime and Delinquency 36(1):56–86. Lipsey, M.W. 1995. “What do we learn from 400 research studies on the effectiveness of treatment with juvenile delinquents?” In What Works: Reducing Reoffending. J. McGuire (ed.). New York, NY: John Wiley and Sons, pp. 63–78. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-69 Lundman, R. 1993. Prevention and Control of Delinquency. Second Edition. New York, NY: Oxford University Press. Macallair, D. 1994. “Disposition case advocacy in San Francisco’s juvenile justice system: A new approach to deinstitutionalization.” Crime and Delinquency 40(1):84–95. Mayer. J.P., and Davidson, W.S. 2000. “Dissemination of innovation and social change.” In Handbook of Community Psychology. J. Rappaport and E. Seidman (eds.). New York, NY: Kluwer Academic/Plenum Publishers. Nelson, R. 1996. Hillsborough County Disproportionate Minority Confinement Pilot Project Evaluation. Tallahassee: Florida Department of Juvenile Justice. Newcomb, M.D. 1995. “Identifying high-risk youth: Prevalence and patterns of adolescent drug use.” In Adolescent Drug Use: Clinical Assessment and Therapeutic Intervention. E. Rahdert, D. Czechowicz, and I. Amsel (eds.). Rockville, MD: National Institute on Drug Abuse. Office of Juvenile Justice and Delinquency Prevention. 2001.Disproportionate Minority Confinement Technical Assistance Manual (Second Edition). Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Orlando, F. 1998. “Controlling the front gates: Effective admissions policies and practices.” Working draft. Vol. 3, Pathways to Juvenile Detention Reform series. Baltimore, MD: Annie E. Casey Foundation. Sampson, R.J., and Lauritsen, J.L. 1997. “Racial and ethnic disparities in crime and criminal justice in the United States.” Ethnicity, Crime, and Immigration 21:311–374. Schiraldi, V., and Ziedenberg, J. 2002. Reducing Disproportionate Minority Confinement: The Multnomah County Oregon Success Story and Its Implications. Washington, DC: Justice Policy Institute, Center on Juvenile and Criminal Justice. Accessed August 26, 2005, at www.cjcj.org/pubs/portland/portland.html. Sheldon, R. 1999. Detention Diversion Advocacy: An Evaluation. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. South Suburban DMC Committee. 2006. “SSDMC response to juvenile justice commission.” An unpublished progress report submitted to the Illinois Juvenile Justice Commission. Stanfield, R. No date. The JDAI Story: Building a Better Juvenile Detention System. Pathways to Juvenile Detention Reform series overview. Baltimore, MD: Annie E. Casey Foundation. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-70 The Sentencing Project. No date. Crack Cocaine Sentencing Policy: Unjustified and Unreasonable. Washington DC: The Sentencing Project. The Sentencing Project. 2000. Reducing Racial Disparity in the Criminal Justice System: A Manual for Practitioners and Policymakers. Washington DC: The Sentencing Project. Villarruel, F.A., Walker, N.E, Minifee, P., Rivera-Vazquez, O., Peterson, S., Perry, K. 2002. ?Donde Esta las Justicia? A Call to Action on Behalf of Latino and Latina Youth in the U.S. Justice System. Executive Summary. Institute for Children, Youth and Families. Michigan State University, East Lansing, MI. Washington State Institute for Public Policy. 2004. Assessing Risk for Re-Offense: Validating the Washington State Juvenile Court Assessment. Olympia, WA: Washington State Institute for Public Policy. Weisheit, R. 1984. “The social context of alcohol and drug education: Implications for program evaluation.” Journal of Alcohol and Drug Education 30:27–35. Welsh, W.N., Jenkins, P.H., Harris, P.W. 1999. “Reducing minority overrepresentation in juvenile justice: Results of community-based delinquency prevention in Harrisburg.” Journal of Research in Crime and Delinquency 36: 87–100. Wiebush, R.G., Baird, C., Krisberg, B., and Onek, D. 1999. “Risk assessment and classification for serious, violent, and chronic juvenile offenders.” In A Sourcebook: Serious, Violent, and Chronic Juvenile Offenders. J.C. Howell, B. Krisberg, J.D. Hawkins, and J.J. Wilson (eds.).Thousand Oaks, CA: Sage Publishers. Zehr, H. 1990. Changing Lenses: A New Focus for Crime and Justice. Scottsdale, PA: Herald Press. Other Resources Demuro, P. 2000. Consider the Alternatives: Planning and Implementing Detention Alternatives. Baltimore, MD: Annie E. Casey Foundation. Pennsylvania Commission on Crime and Delinquency: Juvenile Advisory Committee. Disproportionate Minority Confinement Subcommittee. 2000. Disproportionate Minority Confinement Report. Harrisburg, PA. The Model Systems Project. 2004. A Juvenile Justice Initiative Supported by the MacArthur Foundation. Executive Summary, Pennsylvania Work Plan. DMC Technical Assistance Manual, 4th Edition • Chapter 4: Intervention 4-71 Chapter 5: Evaluation Ashley M. Nellis and Marcia Cohen* Overview Two common approaches to assessing the performance of an intervention are performance measurement and evaluation, both of which are important when examining DMC. OJJDP requires states to submit annual performance measurement data in the area of DMC; these measures can be used in a more extensive evaluation of minority representation in the juvenile justice system. Performance measurement is a subset of evaluation concerned with collecting information to determine whether an intervention is achieving its objectives. It uses output measures and outcome measures to report on program implementation and outcomes and data to improve the operation and/or design of a selected intervention. Output measures are data used to demonstrate the implementation of activities; they include products of activities and indicators of services provided. Outcome measures are data used to assess the achievement of objectives and goals. Evaluation is similar to performance measurement in that it, too, uses output and outcome measures to track progress. Evaluation, however, focuses on how an intervention achieves outcomes. That is, evaluation attempts to determine whether outcomes can be attributed to the intervention itself or to factors external to it. An evaluation assesses the effectiveness of an intervention in achieving its goals and/or objectives, uses methods to determine whether outcomes can be attributed to the program or other factors, and helps determine whether modifications to the program are necessary. Differences between performance measurement and evaluation are summarized in table 1. Sometimes, the same data can be used for both performance measurement and evaluation. However, evaluation entails significantly more extensive analysis, requires more resources, and deals with more complex issues of causality. Before reviewing how evaluation can be undertaken and the types of evaluation designs, this chapter will briefly discuss the requirements of OJJDP’s performance measurement system. * About the authors: Ashley M. Nellis is a Research Associate with the Justice Research and Statistics Association in Washington, DC. Marcia Cohen is Vice President for Research and Evaluation at Development Services Group, Inc., in Bethesda, Maryland. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-1 Table 1: Comparison of Performance Measurement and Program Evaluation Performance Measurement Evaluation What is it? A system of tracking progress in accomplishing specific goals, objectives, and outcomes. A formal scientific process for collecting, analyzing, and interpreting data about how well a program was run (process evaluation) or how effectively the program accomplished desired outcomes (outcome evaluation). Why use it? To improve services, strengthen accountability, enhance decisionmaking, and support strategic planning. To assess program effectiveness and determine whether the program is responsible for any change found. How does it work? Monitors a few vital signs related to program performance objectives, outputs, and outcomes. Comprehensively examines programs using systematic, objective, and unbiased procedures in accordance with social science research methods and research designs. Who does it? Program staff. An experienced researcher (often external to the program) who has formal training in evaluation. When is it done? Periodic intervals (usually annually). Generally one time only (often at the end of the program). Source: ―Performance Measurement Helps Build Results-Driven Programs‖ in Community Prevention: Title V Update, Spring 2003 Issue. Product of OJJDP, Title V Training and Technical Assistance Contract, No. OJP-2000-C-003. Available online at: www.dsgonline.com/Documents/TitleV_Newsletters_Final.pdf. OJJDP’s Performance Measurement Requirements for DMC As stated earlier, performance measurement is a narrower form of tracking progress than program evaluation. It monitors a few vital signs related to program objectives, outputs, and outcomes. A good performance measurement system takes into account certain principles. As noted in the Fairfax County, Virginia, Manual for Performance Measurement (2002), measures should be: Results oriented: Focused primarily on desired outcomes, less on outputs. Important: Concentrated on significant issues. Reliable: Give accurate, consistent information over time. Useful: Yield valuable information to both policy and program decisionmakers and provide continuous feedback on performance to staff and managers. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-2 Quantitative: Expressed in terms of rates or percentages (i.e., they can be calculated). Realistic: Achievable. Cost-effective: Sufficiently valuable to justify the cost of collecting the data. Easy to interpret: Do not require an advanced degree in statistics to understand and use. Comparable: Can be used for benchmarking against other organizations, internally and externally. Credible: Inspire confidence in the validity of the data. OJJDP designed its performance measurement system for the Formula Grants and Title V programs with these principles in mind. OJJDP structured the 34 Formula Grants program areas into four basic categories (prevention, intervention, core requirements, and system improvement) and developed logic models and templates of measures that cross program areas. Although several program areas turn up in more than one category, DMC is unique in that it is included in all four categories. This is because DMC programs can accomplish multiple purposes: direct service programs can serve prevention or intervention populations; activities and strategies can be developed to address the DMC core requirement; or administrative, training, or policy changes can address system-level improvement. Therefore, OJJDP-required DMC performance measures are somewhat more complex than those for other program areas. Logic Models Logic models are an important and valuable strategy for prevention program planning. They provide a way to tie program results to program inputs or resources and are needed for a well-structured program. A logic model is a graphic representation that clearly lays out the logical relationships among the problem, program activities, outputs, and outcomes. It describes how the program theoretically works to achieve benefits for participants and stresses the importance of ensuring that a logical relationship exists among an organization’s goals, what it wants to accomplish, and how it uses resources. The logic models and performance measure matrixes for all 34 Formula Grants program areas and the Title V program areas are based on the logic model format illustrated in the following diagram. OJJDP Logic Model Elements Problem Activities Outputs Outcomes   Goal Short-term Long-term Objectives DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-3 Exhibit 1 provides OJJDP’s definitions of goals, objectives, activities, and outcomes. Exhibit 1: OJJDP’s Definitions of Logic Model Elements Program goal states the overall intent of the program to change, reduce, or eliminate the problem described. The goal is a broad statement about what the program intends to accomplish and, therefore, should be written in general terms. It is also the intended long-term result of the program. Program objectives are derived from the program goals and explain how the grantee will accomplish the program goal. Objectives are well-defined, specific, quantifiable statements of the desired results of the program, and they should include the target level of accomplishment, thereby further defining goals and providing the means to measure program performance. Objectives should be SMART: that is, Specific, Measurable, Attainable, Related to the goal, and Time specific. Activities are the program efforts the grantee will carry out to achieve the identified objectives. Examples include conducting outreach, providing parent training, providing peer mediation services, and providing multisystemic therapy. Remember that activities are not outcomes—―to serve 100 children‖ is not an outcome. Outcomes must state what benefit the participants gained from participating in the program. Performance measures/performance indicators are particular values used to measure program outputs or outcomes. They represent the data/information the grantee will collect at the program level to measure the specific outputs and outcomes a program is designed to achieve. There are two types of performance indicators: Output indicators measure the products of a program’s implementation or activities. They are generally measured in terms of the volume of work accomplished, such as amount of service delivered, staff hired, systems developed, sessions conducted, materials developed, or policies, procedures, and/or legislation created. Examples include the number of juveniles served, the number of hours of service provided to participants, the number of staff trained, the number of detention beds added, the number of materials distributed, the number of reports written, and the number of site visits conducted. Output indicators may also be referred to as process measures. Outcome indicators measure the benefits or changes for individuals, the juvenile justice system, or the community as a result of the program. Outcomes are easiest to remember by the acronym BASK: they may be related to behavior, attitudes, skills, or knowledge. Examples are changes in the academic performance of program participants, changes in the recidivism rate of program participants, changes in client satisfaction level, changes in the conditions of confinement in detention, and changes in the countylevel juvenile crime rate. There are two levels of outcomes: ○ Short-term outcomes are the first benefits or changes participants or the system experience and are the ones most closely related to and influenced by the program’s outputs. They should occur during the program or by the end of the program. For direct service programs, they generally include changes in recipients’ awareness, knowledge, and attitudes. For programs designed to change the juvenile justice system, they include changes to the juvenile justice system that occur during or by the end of the program. ○ Long-term outcomes link a program’s initial outcomes to the longer term outcomes it desires for participants, recipients, the system, or the community. Often they are changes in practice, policy, decisionmaking, or behavior that result from participants’ or service recipients’ new awareness, knowledge, attitudes, or skills or changes in the juvenile justice system. They generally occur within 6 months to 1 year after the program ends. Examples include changes in arrest rates, reductions in truancy, and reductions in substance use. Long-term outcomes are meaningful changes, often in the condition, status, or overall problem behavior that gave rise to the program/ intervention in the first place. They should relate back to the program’s goal, such as reducing delinquency. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-4 DMC Reporting Requirements and Measures In general, the OJJDP performance measurement system requires reporting on all mandatory output and outcome measures, two nonmandatory output measures, and two nonmandatory outcome measures. When reviewing the DMC logic model and grid chart, be sure to note that all mandatory measures are printed in bold; some measures are mandatory for intervention programs only, and these are designated with an *. Mandatory measures for prevention programs are noted with **. Those that are mandatory, if applicable, are noted with a +. Grantees can find the OJJDP DMC logic model at www.dsgonline.com/Program_Logic_Model/LogicModelDec04/10%20DMC%20Logic %20Model.doc. This site is frequently updated; dates of updates are noted on the logic model. Key to DMC Logic Model and Grid Chart: Reporting Requirements: * Bold = Mandatory measure. + = Mandatory only if applicable (if not applicable, choose a different measure). * = Mandatory for intervention programs only. ** = Mandatory for prevention programs only. Output: All mandatory (bold) and two nonmandatory measures. Outcome: All mandatory (bold) and two nonmandatory measures (may be either shortterm or long-term). Output Measures There are 16 DMC output measures. Of these, 3 are mandatory measures and the remaining 13 are nonmandatory measures. The OJJDP reporting requirements for output measures call for reporting on all mandatory measures and two nonmandatory measures. All DMC output measures can be categorized into the following three types of output objectives: Increased organizational/system capacity. Improved planning and development. Improved program activities. Mandatory DMC Output Performance Measures The three mandatory output measures are as follows: Formula Grants or Title V funds allocated or awarded for DMC at the state and local levels. These figures are represented in whole dollars allocated at the state level for the DMC coordinator and at the state and local levels for DMC during the reporting period. Program records are the preferred data source. + Number of programs implemented. Only the state agency provides this number; it should present an aggregate of all DMC-related programs implemented—i.e., the number of state programs in operation at the state and local levels during the reporting period. Formula Grants files are the preferred data source. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-5 + Number of program youth served. This is an unduplicated count of the number of youth that the program served during the reporting period. The definition of the number of youth served for a reporting period is the number of program youth carried over from the previous reporting period, plus new admissions during the reporting period. In calculating the 3-year summary, the total number of youth served is the number of participants carried over from the year previous to the first fiscal year, plus all new admissions during the 3 reporting fiscal years. Program records are the preferred data source. Nonmandatory DMC Output Performance Measures Of the 13 nonmandatory output measures available, OJJDP requires subgrantees to report on 2 measures (see the grid chart in appendix A for definitions of these measures and their reporting format). The DMC nonmandatory performance measures are: Number of full-time equivalent employees funded with Formula Grants funds. Number and percentage of program staff trained. Number of hours of program staff training provided. Number of nonprogram personnel trained. Number of hours of nonprogram personnel training provided. Number of program materials developed. Number of service hours completed. Average length of stay in program. Number of planning activities conducted. Number of assessment studies conducted. Number of data improvement projects implemented. Number of objective decisionmaking tools developed. Number of program/agency policies or procedures created, amended, or rescinded. Outcome Measures There are 11 short-term DMC outcome measures; of these, 6 are mandatory and 5 are nonmandatory. There are seven long-term DMC outcome measures; all are mandatory measures (as applicable). The OJJDP reporting requirements for outcome measures call for reporting on all mandatory measures (as applicable) and two nonmandatory measures. All DMC outcome measures can be categorized into the following six types of outcome objectives: DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-6 Reduce delinquency. Increase accountability. Improve system effectiveness. Increase prosocial behaviors. Increase program support. Reduce DMC. Mandatory DMC Outcome Performance Measures The six short-term and seven long-term mandatory outcome measures are as follows. + Number of state agencies reporting improved data collection systems. The number of state-level agencies that show improved data collection systems as evidenced by an ability to collect data by race; collect data by race with increased accuracy and consistency; report timely data collection and submission, etc. during the reporting period. Data improvement project files are the preferred data source. (short-term and long-term measure) + Number of local agencies reporting improved data collection systems. The number of local-level agencies that show improved data collection systems as evidenced by an ability to collect data by race; collect data by race with increased accuracy and consistency; report timely data collection and submission, etc. during the reporting period. Data improvement project files are the preferred data source. (short-term and long-term measure) + * Number and percentage of program youth who offend or reoffend. The number and percentage of program youth who were rearrested or seen at juvenile court for a new delinquent offense. Appropriate for any youth-serving program. Official records (police, juvenile court) are the preferred data source. Note that this measure is mandatory for intervention programs only. (short-term and longterm measure) + ** Number and percentage of program youth exhibiting desired change in targeted behaviors. The targeted behaviors are substance use, school attendance, antisocial behavior, and family relationships. Note that this measure is mandatory for prevention programs only. See the grid chart in appendix A for definitions of these targeted behaviors and their reporting format. (short-term and long-term measure) + Number and percentage of program youth completing program requirements. The number and percentage of program youth who have successfully fulfilled all program obligations and requirements. Program requirements will vary by program but should be a predefined list of requirements or obligations that clients must meet prior to program completion. Program records are the preferred data source. (short-term measure) + Number of contributing factors determined from assessment studies. Assessment studies are conducted to determine the factors contributing to disproportionality at certain juvenile justice system contact points for certain DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-7 racial/ethnic minority(ies). Count the number of factors in the family, the educational system, the juvenile justice system, and the socioeconomic conditions determined to have contributed to minority overrepresentation at certain juvenile justice system contact points. (short-term measure) + Number and percentage of recommendations from assessment studies implemented. Assessment studies contain multiple recommendations. Count the total number of those chosen for implementation. (long-term measure) + Number of contact points reporting reduction in disproportionality at the state level. Number of contact points reporting significant disproportionality at the state level during the reporting period compared with the last reporting period. Contact points include arrest, referral to juvenile court, diversion, detention, petition filed, found delinquent, probation, secure confinement, and transfer/waiver to adult court. (long-term measure) + Number of contact points reporting reduction in disproportionality at the local level. Number of contact points reporting significant disproportionality at the local level during the reporting period compared with the last reporting period. Contact points include arrest, referral to juvenile court, diversion, detention, petition filed, found delinquent, probation, secure confinement, and transfer/waiver to adult court. (long-term measure) Nonmandatory DMC Outcome Performance Measures Five short-term nonmandatory outcome measures are available: Number of minority staff hired. Number and percentage of program families satisfied with the program. Number and percentage of program youth satisfied with the program. Number and percentage of program staff with increased knowledge of the program area. Number and percentage of nonprogram personnel with increased knowledge of program area. Subgrantees are to report on a total of two nonmandatory outcome measures, either shortterm or long-term (see the grid chart in appendix A for definitions of these measures). Performance Measurement Data Collection To develop a data collection plan, first specify the logical relationships among the problem, program activities, outputs, and outcomes. As stated earlier, this is best accomplished through the development of a logic model, which lays out all of the assumptions about how the subgrantee expects to produce the desired outcomes to the planned DMC activities. For example, let us say that a county is implementing the following three activities/strategies to reduce DMC: DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-8 A new detention alternatives program. A new detention screening instrument. Cultural competency training for all juvenile court workers, including probation officers. Table 2 describes a sample logic model that specifies the relationships between these activities and the selection of specific DMC performance measures that the subgrantee can use to measure the outputs and outcomes. The table specifies the activities that the subgrantee will undertake to implement each of the three DMC initiatives. For the first strategy, implementing a new detention alternative, activities include conducting a needs assessment of current detention alternatives, developing a new alternative, and training staff in using the new alternative. For implementing a new detention screening instrument, activities include hiring a consultant to develop/adapt a new detention screening and needs assessment instrument, training staff in using a new detention screening instrument, pilot testing the new instrument, revising the instrument as necessary, and training all intake staff in using the instrument. For cultural competency training, the activities include selecting a curriculum for the training, hiring trainers, scheduling training, and training all court staff. For each of these activities, mandatory and nonmandatory output measures, short-term outcome measures, and long-term outcome measures have been selected from the DMC Program Area 10 logic model. Table 2: Sample DMC Program Logic Model Using OJJDP DMC Performance Measures Goal: To reduce overrepresentation of minorities in detention and throughout the system. Problem* Inadequate number of alternatives to detention programs available in highminority areas, resulting in minority youth being detained at higher rates than nonminorities. Objective Increase number of detention alternatives for minority youth. Activities Implement one new program to serve as an alternative to detention. Inform court staff of availability of detention alternative. Output Measures Mandatory Formula Grants or Title V funds allocated or awarded for DMC at the local level. Number of programs implemented. Number of program youth served. Short-Term Outcomes Mandatory Number/percentage of youth completing program requirements. Number/ percentage of program youth who reoffend. Long-Term Outcomes Mandatory Number/percentage of program youth who reoffend. Number of contact points reporting reduction in disproportionality at the local level. Number/ percentage of program youth who exhibit change in targeted behaviors. . (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-9 Table 2: Sample DMC Program Logic Model Using OJJDP DMC Performance Measures (continued) Problem* Objective Activities Output Measures Short-Term Outcomes Nonmandatory Number of FTEs funded with Formula Grants funds Nonmandatory Number/percentage of program families satisfied with program. Number of program materials developed. Number/percentage of program youth satisfied with program. Long-Term Outcomes Average length of stay in dayreporting center. Lack of objective detention assessment instrument, resulting in minority youth being placed in detention at higher rates than nonminority youth. Reduce rate of minority placement in detention. Hire consultant to develop/adapt a new detention assessment instrument. Pilot new detention assessment instrument, revise as necessary. Train staff in use of new detention assessment instrument. Use new detention assessment for all youth entering the system. Provide oversight of staff using modified detention tool through modification of policies and procedures. Lack of cultural diversity knowledge, skills, and awareness among court staff, resulting in minority youth staying in detention for longer periods than nonminority youth. Improve the cultural competency of court staff. Hire minority staff. Select curriculum for diversity training for court staff. Hire trainers. Have court staff complete pretraining questionnaire measuring cultural diversity. Mandatory Formula Grants or Title V funds allocated or awarded for DMC at the local level. Nonmandatory Number of objective decisionmaking tools developed. Mandatory Number of contact points reporting reduction in disproportionality at the local level. Nonmandatory Number/percentage of program staff with increased knowledge of program area. Number of staff trained on new detention assessment instrument. Number of program/agency policies or procedures created, amended, or rescinded. Mandatory Formula Grants or Title V funds allocated or awarded for DMC at the local level. Nonmandatory Number of hours of program staff training provided. Number/percentage of program staff trained. Mandatory Number of contact points reporting reduction in disproportionality at the local level. Nonmandatory Number of minority staff hired. Number/percentage of program staff with increased knowledge of program area. Train court staff. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-10 Table 2: Sample DMC Program Logic Model Using OJJDP DMC Performance Measures (continued) Problem* Objective Activities Output Measures Short-Term Outcomes Long-Term Outcomes Have court staff complete posttraining questionnaire measuring cultural diversity. *The problems are identified in a needs assessment that the subgrantee should complete prior to developing the logic model. Table 3 provides a sample data collection plan that specifies in precise, clear, and unambiguous terms the data that the subgrantee must collect and the sources of the data, how the subgrantee will collect the data, where the data are located, and who will collect the data. Subgrantees can obtain performance measurement data from a variety of sources, including individuals involved with a given program or initiative, such as agency staff, and official records (e.g., police reports, court/agency records). The data source(s) the subgrantee chooses will depend on the outcome measures selected and the relative feasibility of getting the data. For example, one of the measures is the “number/percentage of program staff with increased knowledge of the program area”; the most appropriate source of this information is the program staff themselves. Another measure is the number and percentage of juveniles who offend or reoffend. In this case, the subgrantee should obtain the number from official juvenile court or police records. In the output and outcome measures in table 3, BOLD words indicate mandatory measures. The outputs and outcomes designated for performance measurement are derived from the OJJDP logic model for the DMC Program Area 10. Table 3: Data Collection Plan Program Area 10: DMC Example Instrument/ Data Source Formula Grants or Title V funds allocated or awarded for DMC at the local level Annual (at the end of the fiscal year) County controller’s general ledger County controller’s office Assistant to county controller Number of programs implemented Annual (at the end of the fiscal year) Juvenile court director’s office Juvenile court Detention alternatives supervisor Number of program youth served Monthly Juvenile court management information system Juvenile court Detention alternatives supervisor Measures Location of Data Person Who Will Collect Data Frequency of Collection Output Measures DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-11 Table 3: Data Collection Plan Program Area 10: DMC Example (continued) Measures Frequency of Collection Instrument/ Data Source Location of Data Person Who Will Collect Data Number of full-time employees funded with Formula Grants funds Annual Juvenile court director’s office Program files DMC coordinator, program director Number of program materials developed Monthly Program records Program files Program director Average length of stay in program Monthly Program records Program files Program director Number of objective decisionmaking tools developed Annual DMC records DMC files DMC coordinator Number of staff trained on new detention assessment instrument Annual Detention records Detention files Detention staff Number of program/ agency policies or procedures created, amended, or rescinded Annual Detention records Detention files Detention staff Hours of program staff training provided Annual (at the end of the fiscal year) Training files Training office files Training director Number/percentage of program staff trained Annual (at the end of the fiscal year) Training files Training office files Training director Number/percentage of program youth who offend or reoffend Annual (at the end of the fiscal year) Juvenile court management information systems Juvenile court Juvenile court director of research Number/percentage of program youth completing program requirements Annual (at the end of the fiscal year) Juvenile court management information systems Juvenile court Detention alternatives supervisor Number/percentage of program youth exhibiting desired change in targeted behaviors Monthly Special data collection tool Varies Varies Outcome Measures (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-12 Table 3: Data Collection Plan Program Area 10: DMC Example (continued) Instrument/ Data Source Number of minority staff hired Annual (at the end of the fiscal year Personnel files Personnel office Director of personnel Number/percentage of program staff with increased knowledge of program area 1. Pre/post cultural competency training 1. Pre/post cultural competency training survey Juvenile court training department Training director 2. Pre/post detention screening instrument training 2. Pre/post detention screening instrument training survey Number of contact points reporting reduction in disproportionality at the local level Annual (at the end of the fiscal year RRI data reported to the state State juvenile justice planning agency DMC coordinator Number/percentage of program families satisfied with program Monthly Special data collection tool Varies Varies Number/percentage of program youth satisfied with program Quarterly Special data collection tool Detention and detention alternative programs Varies Measures Location of Data Person Who Will Collect Data Frequency of Collection Reporting Periods The Formula Grants performance measurement reporting period is the federal fiscal year, October 1 through September 30. Data reports are due to OJJDP on December 31. The Title V performance measurement reporting period is also the federal fiscal year. Title V data reports are due to OJJDP on November 30. (This allows OJJDP time to prepare the Title V Report to Congress, which is due on March 31.) Systems for Data Entry and Reporting There are two relevant systems for entering and reporting DMC data: the DMC Relative Rate Index (RRI) Reporting System and the OJJDP Data Collection Technical Assistance Tool (DCTAT) for reporting performance measurement data. In both systems, subgrantees can convert reports to PDF format and upload them to the OJJDP Grants Management System (GMS) to fulfill the OJJDP reporting requirement. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-13 DMC RRI Reporting System The RRI Reporting System offers a systematic way to enter RRI data county-by-county and statewide by racial/ethnic group. It automatically calculates RRI at nine contact points: arrest, referral to juvenile court, diversion, secure detention, petition, finding of delinquency, probation, secure confinement, and transfer to adult court based on the atrisk population. The RRI Reporting System Web site produces reports and bar charts of RRI data. It also contains answers to frequently asked questions, reports and other publications, tools, resources, and contacts. The Web site is located at www.dsgonline.com/dmc. OJJDP Data Collection Technical Assistance Tool The DCTAT is the data entry portal for the OJJDP performance measurement system. It is available to states and subgrantees for annual reporting of their performance measurement data. The Web site is located at www.jabg-dctat.org/ojjdpbeta. Performance Measurement Technical Assistance The DMC logic model and all of the other Formula Grants logic models are located at www.dsgonline.com//Program_Logic_Model/fg_pm.htm. OJJDP offers several vehicles for DMC technical assistance, including the OJJDP performance measurement Web site, located at www.dsgonline.com/performance_measures.htm. This site offers slides from all of the OJJDP regional trainings on topics such as principles of effective intervention, developing memoranda of understanding (see appendix C for a model interagency agreement), sharing data, data collection and forms, identifying data sources, and monitoring programs. In addition, OJJDP offers technical assistance in DMC through the Formula Grants Training and Technical Assistance program operated by Development Services Group, Inc. Localities may request assistance from their juvenile justice specialists, who can download a technical assistance request form from www.dsgonline.com/projects_formulagrants.html and submit it to their OJJDP State Representative for approval and further processing. Conducting DMC Evaluations The previous section focused on OJJDP’s performance measurement requirements to report mandatory and nonmandatory output and outcome measures. This section discusses preparation for a data collection and analysis process beyond the steps to meet OJJDP’s requirements—a process that each state can use to develop a more comprehensive understanding of DMC in the state or a locality. Collecting and analyzing data can be a complex undertaking that requires advanced knowledge of research methods and statistics. The following discussion does not reflect an expectation that those working on a DMC intervention should conduct an evaluation DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-14 themselves. Ideally, states and localities will consult with an outside evaluator. (See appendix B for a discussion of the benefits of working with an outside evaluator.) The information provided in this chapter is designed to enhance one’s ability to effectively monitor and advise evaluation planning for the DMC initiative(s) selected. Evaluations can be broken into two broad types: process evaluations and outcome evaluations. Process evaluations focus on how an intervention was implemented and how it operates. They identify the procedures undertaken and the decisions made in developing the intervention, and they describe the services delivered. Like performance measurement, process evaluations address whether the program was implemented according to expectations. However, by additionally documenting the program’s development and operation, they allow an assessment of the reasons for successful or unsuccessful performance and provide information for potential replication. If possible, a process evaluation should be followed by an outcome evaluation. Outcome evaluations are used to identify the results of an intervention’s effort or its long-term outcomes. They seek to answer the question, “What difference did the intervention make?” This type of evaluation typically provides knowledge about: (1) the extent to which the problems and needs that gave rise to the program still exist, (2) ways to ameliorate adverse impacts and enhance desirable impacts, and (3) program design adjustments that may be necessary for future interventions. Outcome evaluations use methods to determine whether achievements can be attributed to the intervention or whether they are attributable to other factors. Such evaluations attempt to control for factors that may influence the outcomes. Outcome evaluations are sometimes referred to as impact evaluations. Successful evaluations are guided by a carefully developed evaluation plan that considers the time required to observe expected results, the staff time and expertise necessary to carry out the evaluation, the funds to be reserved for the evaluation, and the data that will be required. The following section walks the reader through four suggested steps of the evaluation plan: developing an evaluation framework, selecting a research design, developing a plan for data collection, and developing a plan for data analysis and reporting. Once these steps are completed, one can move on to implementing the evaluation. Developing an Evaluation Framework The evaluation framework consists of the theoretical claim(s) to be tested through the evaluation and the logic model. The theoretical claim(s) should be based on the findings of the identification and assessment phases of the DMC strategy. For instance, if one determines that minority overrepresentation is the result of police officers’ disparate treatment of youth based on their race, one might theorize that providing cultural competency training for law enforcement officers will have an impact on DMC. It is essential that some theoretical justification—backed by data obtained in the preliminary phases of the DMC initiative—guide the evaluation design in order to expect that the chosen evaluation will reduce DMC. The evaluation, then, will test whether this theory is supported. Making the theory as specific as possible will aid in the steps that follow. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-15 To logically connect the theory to the activities undertaken and the indicators that will best measure the claims put forth in the theory, the subgrantee should develop a logic model that will visually display the conceptual framework for the evaluation. The concept of the logic model was introduced earlier in this chapter and is suggested for performance measurement as well as evaluation. The components are the same for both, but an evaluation usually requires more measures than are used in performance measurement. This is because an evaluation usually compares outcomes of one group to those of another and requires controls for possible intervening factors that may contribute to the observed outcomes. Recall that the logic model includes the following parts: Problem. Program goals. Objectives. Activities. Performance measures/performance indicators. ○ Output indicators. ○ Outcome indicators (short-term and long-term). Selecting a Research Design Remember that evaluations assess the effectiveness of an intervention in achieving its goals and/or objectives, use methods to determine whether outcomes can be attributed to the program or other factors, and aid in determining whether modifications to the intervention are necessary. Selecting an evaluation design will help to frame the study. There are many research designs that might be appropriate for an evaluation. These designs can be categorized broadly as experimental, quasi-experimental, and nonexperimental. These categories are discussed below. Each type of design has benefits and drawbacks regarding cost, availability of subjects for the study, time required to observe outcomes, and statistical expertise available. Note that performance measurement is a necessary part of each of these research designs; it is essential to measure the extent to which an initiative’s objectives are being met. Also, note that it is unlikely that a jurisdiction will pursue only one DMC intervention. Rather, a DMC strategy will likely comprise several initiatives that are pursued simultaneously, as DMC is usually not the result of only one identified problem. Therefore, the evaluation might include multiple research designs depending on the evaluation needs that are determined. Experimental Designs Experimental designs typically involve the use of random assignment, where a sample is randomly divided into an experimental group (members receive an intervention) and a control group (members do not receive an intervention). The main benefit of experimental designs is the ability to attribute the cause of the observed changes in the experimental group to the intervention rather than to something else. Because of random assignment to the two groups, the two groups are assumed to be equal in all relevant characteristics except the presence of the intervention. Therefore, changes in other DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-16 contributing factors or characteristics (i.e., independent variables) will be equally distributed between the two groups. This design allows one to rule out other factors that may have caused a change in the experimental group. Another benefit of the experimental approach is the ability to expect similar results with repeated trials of similarly situated subjects. In other words, outcomes are typically generalizable to other places. Experimental designs may not always be the best option. One drawback of an experimental design is that it tends to be costly in comparison to the other two research designs covered in this section. Another drawback is that assigning some subjects to a treatment group while withholding treatment from another group is sometimes viewed as unethical. This is especially true if the treatment group is shown to benefit from the intervention. However, remember that the purpose of the evaluation is to determine whether the intervention is beneficial. There are ways around random assignment that preserve the experimental nature of this research design, however. One such approach is the wait-list approach, in which outcomes for the intervention group are compared to outcomes of those on the wait list for the intervention. Once the slots for the designated intervention or program are filled and a wait list of comparable size is generated, researchers compare outcomes between the two groups after the intervention. This design is still considered experimental because the researcher does not preselect which subjects receive the intervention and which do not. Many researchers consider the wait-list approach to be ideal because it greatly reduces the ethical complications that might accompany withholding treatment from control group youths. Quasi-experimental Designs Quasi-experimental designs are a slightly less rigorous type of research design that can deliver high-quality results if designed properly. A quasi-experimental design is ideal when randomization is not possible or is not appropriate. This type of design still includes an experimental group and a control group but does not require random assignment to one or the other. Instead, the researcher or program staff could carefully select subjects who are similar in theoretically important ways (e.g., demographic background, offense history) and place them in either the control group or the experimental group. This is called precision matching. Another commonly used quasiexperimental design is the comparison of outcomes among subjects before an intervention to the outcomes of similar subjects after an intervention. One of the benefits of the quasi-experimental approach is that it is usually more feasible than the experimental approach. Statistical methods that allow controls for possible intervening factors enhance the confidence one can have in the observed outcomes. As with the experimental method, results are typically generalizable for implementation elsewhere. The quasi-experimental design, like the experimental design, requires comparison of two groups, which means that a large enough sample must be available. It is also necessary to have access to two groups that are similar enough to allow comparisons to be drawn. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-17 Non-experimental Designs Non-experimental designs include case studies, surveys, ethnographic studies, and document reviews, among other research strategies. Non-experimental designs are infrequently used in outcome evaluations of interventions because, lacking experimental and control groups, they cannot identify whether outcomes are the result of the intervention. Because these designs can help explain why a condition exists, they are more commonly used in the identification and assessment phases. A non-experimental design is not an evaluation according to the definition provided earlier. It is performance measurement. For instance, to find out what considerations school resource officers use in deciding whether to refer a delinquent youth to school personnel versus the police, indepth interviews of the officers might be conducted. Although such research, if implemented properly, can be informative, outcome evaluations should not rely solely on a non-experimental design. One advantage of a non-experimental design is that it does not require controls for outside factors; instead, one simply observes and carefully records all relevant information. A second advantage is that this type of design typically uses just one group of subjects rather than an experimental group and a control group; therefore, it involves less data collection and is easier to plan for and carry out. Because they only have one group, however, non-experimental designs cannot establish causality. Developing a Plan for Data Collection General Considerations There are a variety of methods for collecting performance measurement and evaluation data, including surveys, interviews, records/documents reviews, and direct observation. Any method often uses a data collection instrument to systematically record the desired data. Data collection instruments can be designed to record information from reports or case records, through interviews (in-person or phone), by observing and recording behaviors, or through questionnaires that targeted groups complete. Regardless of the source of the data and means of data collection, the following considerations apply: Collect all of the data needed to answer the evaluation questions. Before collection, determine how the data will be analyzed, so that data are collected in the right way and no extraneous data are collected. Develop instruments that measure what is intended to be measured (i.e., they are valid). Develop instruments that will produce the same results with repeated use (i.e., they are reliable). Provide clear guidance regarding how to collect data. It is often desirable to use existing instruments, as long as you establish their reliability, validity, and appropriateness for the current evaluation. Use of existing instruments can DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-18 save time and money and could enable you to make valuable comparisons to findings from studies conducted elsewhere. Evaluation plans should include the proposed data collection instruments and the rationale for their use. A newly developed or adapted instrument should be tested to determine its validity and whether using it is feasible. The data collection plan should also specify procedures for ensuring the integrity of the data collection process (including, for example, procedures for training the personnel who will collect, review, interpret, and enter the data), so that the data are complete, accurate, and consistent. To ensure reliability and validity, data collection instruments must also be culturally appropriate. For example, forms and questionnaires should be written in the respondents’ primary language and designed so that respondents of all reading levels can understand the questions. Questions should be culturally appropriate, and care should be taken not to ask questions that respondents might find meaningless, confusing, or offensive. If the data collection process requires direct contact with individuals (e.g., indepth interviews asking residents of high-minority neighborhoods about their perceptions of the juvenile justice system), the persons who collect the data should, as much as possible, reflect the demographic characteristics of the respondents. If appropriate, staff should receive cultural diversity training to aid them in their interactions. Planning is necessary for collecting output, short-term outcome, and long-term outcome data. Long-term outcome data, collected 6 to 12 months after an intervention has ended, requires advanced planning. If evaluators want to collect data from youth, locator forms and special forms indicating consent (passive or active) are needed; see appendix D and appendix E for sample forms. Exhibit 2 presents questions that are helpful in assessing data, sources of data, and methods of data collection. Exhibit 2: Questions To Guide the Development and Assessment of Data Collection Plans Does the data collection plan include all data needed for the output and outcome measures? Are the data requirements clear? Do definitions and sources used comply with the performance measurement definitions, where applicable? Can the data be obtained from the identified source (i.e., is it feasible)? Is the method of data collection appropriate? (Is there evidence that the method of data collection will produce valid and reliable data?) Are the data collection instruments culturally appropriate? Have personnel received necessary cultural competency training? Does the data collection plan include procedures for ensuring data quality through effective management and oversight (e.g., training personnel, data entry, and data ―cleaning‖)? Has adequate planning taken place to collect followup data on program participants 6 months to 1 year after program completion, where applicable? DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-19 Examples of Data Collection Planning Assume that a locality has thoroughly identified the problem, conducted a needs assessment, and prepared for and selected an intervention strategy. It is now equipped to conduct an evaluation. (As noted earlier, it is important to pursue an evaluation wherever possible, because an evaluation will yield a much better understanding of the effectiveness of a DMC strategy.) Consider a locality that has determined through its identification and assessment phases that a lack of available detention alternatives, a biased detention assessment tool, and inadequate cultural competency among court staff are the causes of the disproportionate minority detention rate among its youth population in a high-minority jurisdiction. The locality selects a combination of three interventions to reduce DMC at this decision point and undertakes an evaluation study to measure change. The evaluation should allow researchers to determine whether the observed changes were the result of the intervention strategy. The examples that follow show how the locality might apply different research designs to evaluate its interventions. Example 1 Strategy: Improve detention alternatives in high-minority areas Research Design: Experimental design In a high-minority jurisdiction, first-time, nonviolent juvenile offenders are sent to secure detention because alternatives are lacking. Minority youth are detained at higher rates than nonminority youth and for longer periods of time. It is well established that youth held in detention sever important connections to school, family, peers, and the community, which could result in continued delinquency.† Researchers believe that, in addition to easing overcrowding in detention, completing the requirements at an alternative program will reduce the likelihood that youth will reoffend. Furthermore, preliminary data analyses suggest that an absence of such programming in a highminority area could be the reason so many more minority youth are sent into detention. To divert youth in a high-minority area from detention, officials decide to open a dayreporting center that can accommodate as many as 40 youth. At the day-reporting center, youth participate in job preparation, education, and counseling services. Officials decide to measure the effectiveness of this program using an experimental research design. The evaluators choose a wait-list approach to fulfill the random assignment requirement. The 40 slots are filled quickly, and the next 40 youth who are sent to secure detention serve as the comparison group for the study. An important outcome measure for this intervention is the change in the DMC rate in detention. The expectation is that the DMC rate will drop because minority youth are † See J. Austin, K.D. Johnson, and R. Weitzer, Alternatives to the Secure Detention and Confinement of Juvenile Offenders, Juvenile Justice Practices Series Bulletin, Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention, 2005. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-20 diverted from the system and sent to a day-reporting center. For this reason, perhaps an even more informative measure is the recidivism rate among the experimental and control group youth. The theoretical claim is that if first-time, nonviolent offenders are provided with alternatives to detention, they will be less likely to reoffend. To measure this empirically, evaluators decide to track youth throughout the intervention period of 6 months and then at 12 and 18 months after youth complete the program or their stay in detention. Factors other than reoffending indicate changes that are important to the evaluators. As discussed previously, risk and protective factors indicate a youth’s vulnerability to engaging in delinquency or staying away from it. Based on the risk and protective factor literature, evaluators also include measures of academic performance, attitudes and behaviors, and relationships with family and friends. Other factors that may be of interest to evaluators and administrators are whether length of stay in detention decreases and whether the RRI in detention changes. In summary, the evaluation seeks to test three theoretical claims: (1) alternatives to detention will lower the DMC rate at the detention decision point; (2) alternatives will lower the recidivism rate among program participants compared with similar youth who received detention; and (3) alternatives will lower risk factors associated with delinquency among program participants compared with similar youth who received detention. Given that the only difference between the two groups is that one group received treatment and the other received the standard approach of secure detention, researchers can attribute any statistically significant differences in outcomes to the program. However, although differences are expected to be evenly distributed between the two groups, it is still necessary to document and control for important characteristics about all of the youth in the study and ensure that the groups are similar on these factors. Changes to any of the factors during the course of the intervention should be noted, as they may affect the outcomes. Evaluators hypothesize that youth who complete the program at the day-reporting center will score significantly higher on school, family, community, and peer factors after the program than youth in the comparison group because the program youth have been diverted from the system and received more intensive, early intervention services. Specifically, the following information is recorded from both groups: Background information Age Race Ethnicity Sex Number of prior offenses Living arrangements Family substance abuse history Family offense history DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-21 School factors Academic performance Disciplinary actions Attitude toward school Family factors Traumatic events (e.g., death, divorce, domestic violence) Family attachment Community factors Crime rate Employment rate Poverty rate Peer factors Association with prosocial peers Association with antisocial peers The logic model example provided in the earlier discussion of performance measures can also be used for an evaluation, with modifications for collecting additional data on both the treatment and control subjects. As noted earlier, the main difference between performance measurement and an evaluation is the ability to determine whether observed changes are the result of the intervention. To make that determination, it is necessary to carefully record additional relevant information about the treatment and the control groups. The additional output and outcome measures in the tables that follow will help evaluators determine the effectiveness of the day-reporting center in reducing overrepresentation of minority youth in the area. Example 1: Logic Model for Detention Alternatives Problem Inadequate number of alternatives to detention programs available in highminority areas, resulting in minority youth being detained at higher rates than nonminorities. Objective Activities Output Measures Increased number of detention alternatives for minority youth. Implement one new program to serve as an alternative to detention. Mandatory Formula Grants or Title V funds allocated or awarded for DMC at the local level. Number of programs implemented. Number of program youth served. Short-Term Outcomes Long-Term Outcomes Mandatory Number/percentage of day- reporting center program youth completing program requirements. Mandatory Number/percentage of day- reporting center youth who reoffend. Number/percentage of day-reporting center program youth who reoffend. Number/percentage of day-reporting center youth exhibiting change in targeted behaviors. (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-22 Example 1: Logic Model for Detention Alternatives (continued) Problem Objective Activities Output Measures Short-Term Outcomes Nonmandatory Number of fulltime employees funded with Formula Grants funds. Nonmandatory Number/percentage of program families who are satisfied with day-reporting center program. Number of dayreporting center program materials developed. Number/percentage of program youth who are satisfied with day- reporting center program. Long-Term Outcomes Nonmandatory Number of contact points reporting reduction in disproportionality at the local level. Average length of stay in dayreporting center. Additional Number of detention center youth served. Number of hours spent in education versus total hours offered. Number of hours spent in substance abuse versus total hours offered. Number of hours spent in counseling sessions versus total hours offered. Average length of stay in detention center. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation Additional Academic performance of dayreporting center youth. Additional Academic performance of dayreporting center youth. Academic performance of detention center youth. Academic performance of detention center youth. Disciplinary actions taken against dayreporting center youth. Disciplinary actions taken against dayreporting center youth. Disciplinary actions taken against detention center youth. Disciplinary actions taken against detention center youth. Reported substance abuse of dayreporting center youth. Reported substance abuse of dayreporting center youth. Reported substance abuse of detention center youth. Reported substance abuse of detention center youth. Family attachment among dayreporting center program youth. Family attachment among dayreporting center program youth. Family attachment among detention center youth. Family attachment among detention center youth. Relationships with peers among dayreporting center program youth. Relationships with peers among dayreporting center program youth. (continued) 5-23 Example 1: Logic Model for Detention Alternatives (continued) Problem Objective Activities Short-Term Outcomes Output Measures Long-Term Outcomes Relationships with peers among detention center youth. Relationships with peers among detention center youth. Number/percentage of detention center families reporting satisfaction with detention center. Number/ percentage of detention center families reporting satisfaction with detention center. Number/percentage of detention center youth reporting satisfaction with detention center. Number/ percentage of detention center youth reporting satisfaction with detention center. In addition to knowing what the evaluator wants to measure to determine change between the experimental and control groups of youth, completing a data collection plan assists with organizing and coordinating the data collection process. The “additional” measures in the tables for this example, mentioned above, expand on the original DMC mandatory and optional performance measures. Data are obtained from official sources (school records, police records, detention records, program records, county controller records) and from self-report data obtained through questionnaires administered to the youths and their families. Both types of data sources can produce valid and reliable information. Example 1: Data Collection Plan for Evaluation—Output Measures Output Measures Frequency of Collection Instrument/ Data Source Location of Data Person Who Will Collect Data Mandatory Output Measures Formula Grants funds allocated or awarded for DMC at the state and local levels Annual County controller’s general ledger County controller’s office Assistant to the county controller Number of programs implemented Annual Juvenile court director’s office Juvenile court Detention alternatives supervisor Number of day- reporting center program youth served Monthly Program records Program files Program director Appropriate Nonmandatory Output Measures Number of full-time employees funded with Formula Grants funds developed Annually Program records Program files DMC coordinator, program director Number of program materials developed for day-reporting center Baseline and monthly Program records Program files Program director Average length of stay in program versus program length Monthly Program records Program files Program director (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-24 Example 1: Data Collection Plan for Evaluation—Output Measures (continued) Additional Output Measures Number of hours spent in education versus total hours offered at dayreporting center Monthly Program records Program files Program director Number of hours spent in substance abuse programming versus total hours offered at day-reporting center Monthly Program records Program files Program director Number of hours spent in counseling services versus total hours offered at day-reporting center Monthly Program records Program files Program director Example 1: Data Collection Plan for Evaluation—Short-Term Outcome Measures Short-Term Outcome Measures Frequency of Collection Instrument/ Data Source Location of Data Person Who Will Collect Data Mandatory Short-Term Outcome Measures Number/percentage of day-reporting center youth completing program requirements Monthly Program records Program files Program director Number/percentage of day-reporting center youth who reoffend Monthly Police records Police department Program director Number/percentage of program youth exhibiting desired change in targeted behaviors Monthly Program records Program files Program director Appropriate Nonmandatory Short-Term Outcome Measures Number/percentage of program families satisfied with the dayreporting center program Monthly Postquestionnaire Evaluator files Evaluator Number/percentage of program youth satisfied with the day reporting center program Monthly Postquestionnaire Evaluator files Evaluator Additional Short-Term Outcome Measures Number/percentage of youth in detention who reoffend Baseline and monthly Police records Police department Police staff Number/percentage of detention center youth completing detention requirements Baseline and monthly Detention records Detention files Detention staff Number/percentage of detention center youth exhibiting desired change in targeted behaviors Baseline and monthly Detention records Detention files Detention staff Academic performance of dayreporting center youth Baseline and monthly School records School personnel School staff Academic performance of detention center youth Baseline and monthly Detention records Detention files Detention staff Reported substance abuse of dayreporting center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Reported substance abuse of detention center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Disciplinary actions taken against day- reporting center youth Baseline and monthly School records School personnel School staff Disciplinary actions taken against detention center youth Baseline and monthly Detention records Detention files Detention staff (continued) DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-25 Example 1: Data Collection Plan for Evaluation—Short-Term Outcome Measures (continued) Short-Term Outcome Measures Frequency of Collection Instrument/ Data Source Location of Data Person Who Will Collect Data Levels of family attachment reported by day-reporting center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Levels of family attachment reported by detention center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Involvement with prosocial peers among day-reporting center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Involvement with prosocial peers among detention center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Example 1: Data Collection Plan for Evaluation—Long-Term Outcome Measures Long-Term Outcome Measures Frequency of Collection Instrument/ Data Source Location of Data Person Who Will Collect Data Mandatory Long-Term Outcome Measures Number/percentage of day-reporting center youth who reoffend Monthly Police records Police files Police staff Number of contact points reporting reduction in disproportionality at the local level Annual RRI Detention data; DMC files DMC coordinator Number/percentage of detention youth who reoffend Monthly Police records Police files Police staff Number/percentage of detention center youth completing detention requirements Baseline and monthly Detention records Detention files Detention staff Number/percentage of detention center youth exhibiting desired change in targeted behaviors Baseline and monthly Detention records Detention files Detention staff Academic performance of dayreporting center youth Baseline and monthly School records School personnel School staff Academic performance of detention center youth Baseline and monthly Detention records Detention files Detention staff Reported substance abuse of dayreporting center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Reported substance abuse of detention center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Disciplinary actions taken against day- reporting center youth Baseline and monthly School records School personnel School staff Disciplinary actions taken against detention center youth Baseline and monthly Detention records Detention files Detention staff Levels of family attachment reported by day-reporting center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Levels of family attachment reported by detention center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Involvement with prosocial peers among day reporting center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Involvement with prosocial peers among detention center youth Baseline and monthly Pre/post questionnaire Evaluator files Evaluator Additional Long-Term Outcomes DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-26 Example 2 Strategy: Implement new detention screening instrument Research design: Quasi-experimental design Consider the second identified problem mentioned on page 20: the lack of objective detention assessment tools, resulting in minority youth being placed in detention at higher rates than nonminority youth. Researchers hypothesize that revising the detention assessment tool will result in a lower rate of minority youth entering the system. A consultant is hired to complete this task. Once the tool is revised, all staff who make detention decisions are trained in its use and are carefully supervised to minimize discretionary decisions as much as possible. Staff use the tool to make the detention decision for all youth who enter the system. Similar to the day-reporting center component of this evaluation, measuring change as a result of the modified detention assessment tool requires two samples whose outcomes researchers can compare. For this approach, researchers decide to compare outcomes for a random sample of 200 youth using the old tool with outcomes for a random sample of 200 youth using the newly modified tool. They use demographic information on these two groups to ensure that youth from each sample are matched on relevant variables including race, ethnicity, age, sex, current offense, offense history, and living arrangements. To determine whether the revised instrument has an impact on DMC, researchers decide to compare the detention rates of minority and nonminority youth during the 3 months before the revision to rates 3 months after the revision. Therefore, detention data for 6 months are required for this evaluation. To summarize, the theoretical claim being tested is that the rate at which staff send minority youth to detention will be significantly lower when they use the revised instrument than it was when they used the old instrument. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-27 The logic model for the evaluation component of this intervention is provided below. Example 2: Logic Model for Detention Assessment Tool Problem Lack of objective detention assessment instrument, resulting in minority youth being placed in detention at higher rates than nonminority youth. Objective Reduce rate of minority placement in detention. Activities Output Measures Hire consultant to develop/adapt a new detention assessment instrument. Mandatory Formula Grants or Title V funds allocated or awarded for DMC at the local level. Pilot test new detention assessment instrument, make revisions, as necessary. Train staff in using new detention assessment instrument. Use new detention assessment instrument for all youth entering the system. Provide oversight of staff using modified detention tool through modification of agency policies and procedures. Nonmandatory Number of objective decisionmaking tools developed. Number/percentage of staff trained in new detention assessment instrument. Short-Term Outcomes Long-Term Outcomes Mandatory Number of contact points reporting reduction in disproportionality at the local level. Nonmandatory Number/percentage of program staff with increased knowledge of program area. Number of program/agency policies or procedures created, amended, or rescinded. Additional Number of assessments conducted with modified detention assessment instrument versus total number of assessments. Additional Detention rates of minority youth. Additional Detention rates of minority youth. Detention rates of nonminority youth. Detention rates of nonminority youth. Number of detention overrides. Number of detention overrides. Example 2: Data Collection Plan for Evaluation—Output Measures Output Measures Frequency of Collection Instrument/ Data Source Location of Data Person Who Will Collect Data County controller’s general ledger County controller’s office Assistant to the county controller Mandatory Output Measures Formula Grants allocated or awarded for DMC at the local level. Annual Appropriate Nonmandatory Output Measures Number of objective decisionmaking tools developed Baseline DMC committee records DMC committee files Local DMC coordinator Number/percentage of staff trained in new detention assessment instrument Baseline Detention records Detention files Detention staff Number of program/agency policies or procedures created, amended, or rescinded Annually Detention records Detention files Detention staff Monthly Detention records Detention files Detention staff Additional Output Measures Number of assessments conducted with modified detention assessment instrument versus total number of assessments DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-28 Example 2: Data Collection Plan for Evaluation—Short-Term Outcome Measures Short-Term Outcome Measures Frequency of Collection Instrument/Data Source Location of Data Person Who Will Collect Data Pre/post questionnaire Evaluator files Evaluator Appropriate Nonmandatory Short-Term Outcome Measures Number/percentage of program staff with increased knowledge of program area Baseline and after training Additional Short-Term Outcome Measures Detention rates of minority youth Monthly Detention records Detention files Detention staff Detention rates of nonminority youth Monthly Detention records Detention files Detention staff Number of detention overrides Monthly Detention records Detention files Detention staff Example 2: Data Collection Plan for Evaluation—Long-Term Outcome Measures Long-Term Outcome Measures Frequency of Collection Instrument/Data Source Location of Data Person Who Will Collect Data Appropriate Mandatory Long-Term Outcome Measures Number of contact points reporting reduction in disproportionality at the local level Annual RRI Detention data; local DMC coordinator files Local DMC coordinator Additional Long-Term Outcome Measures Detention rates of minority youth Monthly Detention records Detention files Detention staff Detention rates of nonminority youth Monthly Detention records Detention files Detention staff Number of detention overrides Monthly Detention records Detention files Detention staff Example 3 Strategy: Provide cultural competency training for all juvenile court workers, including probation officers Research design: Non-experimental design The third identified problem was a lack of cultural competency knowledge, skills, and awareness among court staff, resulting in minority youth staying in detention for longer periods than nonminority youth. One simple solution to this problem was to hire more minority staff to more closely match racially and ethnically the youth entering the system. As a second strategy, the DMC committee funds a cultural competency training program for all juvenile court workers, including probation officers. To evaluate the effectiveness of this training, researchers conduct indepth interviews of court staff a month before the training and a month after the training. This is a non-experimental design mainly because the study uses only one group; it does not have an experimental and a control group, as in experimental and quasi-experimental designs. Court staff are asked qualitative questions about their general perceptions of youth from various backgrounds, the factors they consider when deciding the length of detention to assign to youth, and their impression of the degree do which youth respect the juvenile justice system. Of course, evaluators should be aware of the strong possibility that staff participation in the interview before the training may effect the responses they provide in interviews after the training. This is DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-29 called a testing effect, meaning that exposure to the pretest could influence comments a respondent provides in the second interview. Whereas in the first interview the respondents were unaware of the topics to be covered and were therefore likely to be more candid in their responses, they might not be as candid in the second interview. Researchers can reduce such threats to validity by carefully structuring questions that are modified in the second interview but still collect the same information. In addition to the qualitative information collected in the indepth interview, some quantitative data can easily aid in determining the effective of the training on length of stay among minority versus nonminority youth. Therefore, length of detention among youth who entered the system during this time is also recorded as a short- and long-term outcome measure to see whether cultural competency training has an impact on the duration of detention. In summary, the theoretical claim examined here is that participation in cultural competency training will reduce the average length of detention stays among minority youth in the area. The logic model for providing cultural competency training is provided below. Example 3: Logic Model for Cultural Competency Training Problem Lack of cultural diversity knowledge, skills, and awareness among court staff, resulting in minority youth staying in detention for longer periods than nonminority youth. Objective Improve the cultural competency of court staff. Activities Hire minority staff. Select curriculum for cultural competency training for court staff. Hire trainers. Have court staff complete pretraining questionnaire measuring cultural diversity. Train all court staff. Have court staff complete posttraining questionnaire measuring cultural competency. Output Measures Short-Term Outcomes Mandatory Formula Grants or Title V funds allocated or awarded for DMC at the local level. Nonmandatory Number of hours of program staff training provided. Number of program staff trained. Additional Number of prediversity training in-depth interviews completed. Number of postdiversity training indepth interviews completed. Long-Term Outcomes Mandatory Number of contact points reporting reduction in disproportionality at the local level. Nonmandatory Number of minority staff hired. Number/percentage of program staff with increased knowledge of program area. Additional Length of stay in detention among minority youth. Additional Length of stay in detention among minority youth. Length of stay in detention among nonminority youth. Length of stay in detention among nonminority youth. Perception of minority youth. Perception of minority youths’ attitude toward the juvenile justice system. Factors included in detention length decision. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-30 Example 3: Data Collection Plan for Evaluation—Output Measures Output Measures Frequency of Collection Instrument/ Data Source Location of Data Person Who Will Collect Data County controller’s general ledger County controller’s office Assistant to the county controller Mandatory Output Measures Formula Grants funds allocated or awarded for DMC at the local level Annual Appropriate Nonmandatory Output Measures Number of hours of program staff training provided Post training Evaluator records Evaluator files Evaluator Number of program staff trained Post training Evaluator records Evaluator files Evaluator Number of pre-diversity training indepth interviews completed Pre training Evaluator records Evaluator files Evaluator Number of post-diversity training indepth interviews completed Post training Evaluator records Evaluator files Evaluator Additional Output Measures Example 3: Data Collection Plan for Evaluation—Short-Term Outcome Measures Short-Term Outcome Measures Frequency of Collection Instrument/ Data Source Location of Data Person Who Will Collect Data Appropriate Nonmandatory Short-Term Outcomes Number of minority staff hired Monthly Court personnel records Court personnel files Court personnel staff Number/percentage of program staff with increased knowledge of program area Pre and post training In-depth interview protocol Evaluator files Evaluator Length of stay in detention among minority youth Monthly Detention records Detention files Detention staff Length of stay in detention among nonminority youth Monthly Detention records Detention files Detention staff Perception of minority youth Pre and post training Indepth interview protocol Evaluator files Evaluator Perception of minority youth attitudes toward the juvenile justice system Pre and post training Indepth interview protocol Evaluator files Evaluator Factors included in detention length decision Pre and post training Indepth interview protocol Evaluator files Evaluator Additional Short-Term Outcomes Example 3: Data Collection Plan for Evaluation—Long-Term Outcome Measures Instrument/ Data Source Location of Data Annual Local DMC coordinator records Detention data; local DMC coordinator files Local DMC coordinator Length of stay in detention among minority youth Monthly Detention records Detention files Detention staff Length of stay in detention among nonminority youth Monthly Detention records Detention files Detention staff Long-Term Outcome Measures Frequency of Collection Person Who Will Collect Data Appropriate Mandatory Long-Term Outcomes Number of contact points reporting reduction in disproportionality at the local level Additional Long-Term Outcomes DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-31 Developing a Plan for Data Analysis and Reporting Analysis The analysis of data will depend largely on the data collected and the information sought. It is advisable to have someone with experience in research methods carry out the necessary analyses for the higher level of investigation involved in an evaluation. Using someone with an advanced degree in criminal justice (or a related field), research methods, or statistics is strongly encouraged because working with data is a complex undertaking. Beyond reporting numbers, a locality that undertakes an evaluation will want to test the theoretical claims based on existing data by measuring pre- and postintervention data. Although evaluations require more time, money, and expertise than performance measurements, they ultimately yield a more reliable product and may be a more efficient investment of resources because they allow greater depth in analyzing minority overrepresentation. States that work in conjunction with a university or a statistical analysis center (SAC) are often in a much better position to conduct more demanding analyses, because they have access to an economical source of expertise. (SACs are state agencies that collect, manage, analyze, and disseminate justice data. A list of state SACs is available on the Justice Research and Statistics Association Web site at www.jrsa.org. Reporting Once the evaluation of an initiative, a strategy, or a program is complete, findings should be reported. The following suggestions can facilitate the reporting process: Be aware in advance that the report’s findings may well generate controversy. Try to anticipate issues and plan responses. To distance the findings from the study’s working group members and stakeholders, consider having an outside source (a reputable, objective organization such as a university that is not affiliated with the juvenile justice system) conduct the evaluation. Remember that the report will be more credible if findings are considered objective rather than driven by a particular agenda. Make stakeholders aware of important findings along the way, to avoid any surprises at the end. Write reports in a way that the public, juvenile justice administrators and personnel, and the media can easily understand. Avoid statistical jargon and language that would be unfamiliar to anyone outside the juvenile justice system. After stakeholders have had a chance to address study findings, release the findings to the media. This will help to ensure that controversial findings are not ignored or denied and will keep issues in the open, so that positive change remains a priority. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-32 Reassessing Intervention Logic Once the evaluation is complete, the project team should ask: Is this what we expected to happen? Are we satisfied with the results? How do the outcomes we observed relate to the intervention? Reassessing intervention logic allows the team to review its original intentions, the actual activities, and the outcomes to determine whether the outcomes met the expectations. Some additional questions worth asking include the following: Which objectives have been accomplished? Which have not, and why not? Are there data to suggest why certain objectives were not accomplished? Do certain objectives or activities need to be modified? Does the goal need to be redefined? Answers to these questions may lead to the adoption of modified measures to assess effectiveness, which may, in turn, require new data collection techniques. This reassessment is central to implementing evaluation as an ongoing process that includes intervention development, assessment, and revision. Overcoming Obstacles to Evaluations States and localities may face some obstacles to progress in their evaluations. A review of DMC studies over the past several years reveals a few commonly mentioned obstacles. These obstacles are related to identifying the correct target population of the DMC strategy, securing stakeholder support, investing resources in evaluation, and relying on intervention-level staff to carry out the evaluation. Although most juvenile justice interventions so far have targeted youth as the focus of change, DMC interventions should not rely solely on programs for youth. In other words, to reduce minority overrepresentation, a jurisdiction may have to examine closely (and then change) its systemwide policies and procedures that determine how all youth are handled within the juvenile justice system. Those whose work involves juvenile justice programs that serve youth may be unfamiliar with evaluations of system-level changes, but with careful planning and data collection, it is possible to evaluate interventions, programs, and strategies. Models such as the Annie E. Casey Foundation’s Juvenile Detention Alternatives Initiative can assist in planning a systemic approach to DMC reduction. States and localities may be reluctant to invest limited resources in evaluation if the benefits of this investment are not immediately apparent. However, spending resources to appropriately identify the problem(s), use evidence-based programs/strategies/ interventions, and assess outcomes will help ensure that resources are used wisely and efficiently. States and localities are encouraged to devote resources to the critical step of performance measurement and evaluation whenever possible, as findings will inform future planning and funding decisions. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-33 OJJDP recognizes that conducting analyses beyond measuring performance may present an excessive burden on states. Although OJJDP encourages states and localities to devote additional time and resources to studying minority overrepresentation in their juvenile justice systems beyond the above requirements, this may not be feasible. However, once data are routinely collected, and if resources and expertise are available, states should further explore the causes and persistence of DMC because such efforts are likely to result in a more thorough understanding of this complex problem. States and localities may overcome the financial burden associated with evaluation by partnering with university researchers and graduate students, who are often interested in opportunities to conduct these types of analyses. Finally, performance measurement and evaluation of DMC reduction strategies require the involvement of staff in the geographic area under study as well as state-level staff. For instance, if a state is implementing an intervention to reduce overrepresentation, intervention-level staff may be tasked with counting youth and recording race and ethnicity data on a routine basis. Intervention-level staff may also submit necessary data to the state’s DMC coordinator and may conduct some analyses of these data as well. Regardless of the strategy chosen, the DMC coordinator is actively involved in all aspects of the DMC reduction intervention and should be aware of data collection efforts and activities. Summary This chapter introduced the topics of performance measurement and evaluation, identifying similarities and differences between these types of assessments. Performance measurement is a necessary part of evaluation, but evaluation extends what we know about the effectiveness of an intervention by allowing us to determine whether outcomes observed were the result of the intervention or something else. Evaluations accomplish this by considering, through statistical means and the use of multiple groups, the possibility that outside factors contributed to the observed results. This chapter also discussed OJJDP’s DMC performance measurement requirements and spelled out the mandatory and nonmandatory indicators that one should be familiar with for this program area. Next, using three examples of mechanisms leading to DMC, the chapter discussed possible intervention strategies and the steps involved in the transition from performance measurement to evaluation of these strategies: developing an evaluation framework, creating a research design, developing a plan for data collection, and developing a plan for data analysis and reporting. The chapter also touched on some of the obstacles that DMC studies commonly experience and suggested ways of overcoming these obstacles. One final reminder: Although the evaluation is best left to those with substantial expertise in this area, the whole DMC team should be aware of the steps involved and participate in planning the evaluation. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-34 Appendix A: Disproportionate Contact Performance Measure Definitions and Reporting Format Performance Measure Key Short Term: Long Term: Annual Term: Bold: Bold*: Bold**: +: Occurs during or by the end of the program. Occurs 6 months to 1 year after program completion. Occurs once a year. Mandatory measure. Mandatory for intervention programs only. Mandatory for prevention programs only. Mandatory only if applicable (if not applicable, choose a different measure). Output Performance Measures # Output Measure Objective Definition 1 Formula Grants or Title V funds allocated or awarded for DMC at the state and local levels Increase organizational/ system capacity The amount of funds in whole dollars allocated at the state level for the DMC coordinator and awarded for DMC at the state and local levels during the reporting period. Program records are the preferred data source. Dollars allocated and awarded for DMC 2 Number of full-time equivalents (FTEs) funded with Formula Grants dollars Increase organizational/ system capacity The number of DMC coordinators funded with Formula Grants funds, as measured through the number of FTEs, working on DMC at the state or local levels during the reporting period. To calculate FTE, divide the number of staff hours that the program uses and divide by 2080. Number of FTEs DMC coordinators paid with Formula Grants funds 3 + Number of programs implemented Increase organizational/ system capacity Only the state agency provides this number, and it should present an aggregate of all DMC-related programs implemented. The number of state programs in operation at the state and local levels during the reporting period. Formula Grants files are the preferred data source. Number of DMC-related programs in operation during the reporting period 4 Number/percentage of program staff trained Increase organizational/ system capacity The number and percentage of program staff trained on DMC-related issues such as improving staff’s understanding of cultural differences, cultural context, cultural diversity, cultural awareness, bias, multicultural workplaces, etc. during the reporting period. The number is the raw number of staff who receive any formal training relevant to the program or their position as program or state-level staff. Include any training from any source or medium received during the reporting period as long as you can verify receipt. Training does not have to have been completed during the reporting period. To get the percentage, divide the raw number by the total number of program staff. Training records are the preferred data source. A. Number of staff who participated in training DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation Reporting Format B. Total number of staff C. Percentage (A/B) 5-35 Output Performance Measures (continued) # Output Measure Objective Definition Reporting Format 5 Number of hours of program staff training provided Increase organizational/ system capacity The number of DMC-related training hours provided to program staff during the reporting period of the program. Training includes in-house and external trainings, conducted and available to staff. Number of DMC-related hours of training provided to staff 6 Number of nonprogram personnel trained Increase organizational/ system capacity The number of nonprogram people trained on DMC-related issues such as improving understanding of cultural differences, cultural context, cultural diversity, cultural awareness, bias, multicultural workplaces, etc., during the reporting period. The number is the raw number of nonprogram people from law enforcement, courts, other related agencies, or community members who participate in training, conferences, or workshops. Although DMC program staff may also participate in such training (e.g., statewide or local DMC conferences), do not count them here. Count them under #4. Number of nonprogram people who participated in training 7 Number of hours of nonprogram personnel training provided Increase organizational/ system capacity The number of DMC-related training hours provided to nonprogram people during the reporting period. Include DMC training, conferences, and workshops conducted not just for DMC program staff only but for juvenile justice system personnel in general (e.g. law enforcement, court, etc.) and other related agencies and community members. Number of DMC-related hours of training provided to nonprogram personnel 8 Number of program materials developed Increase organizational/ system capacity The number of DMC-related materials developed during the reporting period. Include only substantive materials such as cultural competency or DMC curricula, brochures, videos about DMC, etc. Do not include program advertisements or administrative forms such as sign-in sheets or client tracking forms. Count the number of pieces developed. Number of program materials developed during reporting period 9 + Number of program youth served Improve program activities An unduplicated count of the number of youth that the program served during the reporting period. Definition of the number of youth served for a reporting period is the number of program youth carried over from the previous reporting period, plus new admissions during the reporting period. In calculating the 3-year summary, the total number of youth served is the number of participants carried over from the year previous to the first fiscal year, plus all new admissions during the 3 reporting fiscal years. Program records are the preferred data source. Number of program youth carried over from the previous reporting period, plus new admissions during the reporting period. 10 Number of service hours completed Improve program activities The number of hours of service that program youth completed during the reporting period. Service is any explicit activity (such as program contact, counseling sessions, course curriculum, community service, etc.) that program staff or other professionals dedicated to completing the program requirements delivered. Program records are the preferred data source. Total number of program youth service hours DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-36 Output Performance Measures (continued) # 11 Output Measure Average length of stay in program Objective Improve program efficiency Definition The average length of time (in days) clients remain in the program. Include data both for clients who complete program requirements prior to program exit and for clients who do not. Program records are the preferred data source. Reporting Format A. Total number of days between intake and program exit across all clients served B. Number of cases closed C. A/B 12 Number of planning activities conducted Improve planning and development The number of DMC-related planning activities undertaken during the reporting period. Activities include number of memoranda of understanding developed, number of DMC subcommittee meetings held, etc. Number of planning activities undertaken 13 Number of assessment studies conducted Improve planning and development The number of DMC assessment studies undertaken during the reporting period to determine factors contributing to DMC. Number of assessment studies undertaken 14 Number of data improvement projects implemented Improve planning and development The number of data improvement projects funded at the state or local levels specifically to improve the quality and completeness of DMC data. Number of projects funded during the reporting period 15 Number of objective decisionmaking tools developed Improve planning and development Report whether any objective decisionmaking tools, such as detention risk, risk assessment, needs assessment, mental health assessment, were developed to determine the supervision needs of the youth. Number of tools developed 16 Number of program/agency policies or procedures created, amended, or rescinded Improve planning and development The number of DMC-related policies or procedures created, amended, or rescinded during the reporting period. Policies or procedures can be developed at the state or local levels. A policy is a plan or specific course of action that guides the general goals and directives of the program or agency. Include policies that are either relevant to the topic area of the program or policies that affect program operations. Number of program/agency policies or procedures created, amended, or rescinded DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-37 Outcome Performance Measures # Outcome Measure Reporting Term Objective Definition Reporting Format Short Long 1 + Number of state agencies reporting improved data collection systems Improve system effectiveness The number of state-level agencies that show improved data collection systems as evidenced by an ability to collect data by race, collect data by race with increased accuracy and consistency; report timely data collection and submission, etc. during the reporting period. Data improvement project files are the preferred data source. Number of improved state-level data collection systems during the reporting period. √ √ 2 + Number of local agencies reporting improved data collection systems Improve system effectiveness The number of local-level agencies that show improved data collection systems as evidenced by an ability to collect data by race, collect data by race with increased accuracy and consistency; report timely data collection and submission, etc. during the reporting period. Data improvement project files are the preferred data source. Number of improved local-level data collection systems during the reporting period. √ √ 3 Number of minority staff hired Improve system effectiveness The number of staff of a specific minority group hired during the reporting period. The number of minority staff hired √ 4 + * Number/ percentage of program youth who offend or reoffend Reduce delinquency The number and percentage of program youth who were rearrested or seen at juvenile court for a new delinquent offense. Appropriate for any youthserving program. Official records (police, juvenile court) are the preferred data source. A. Number of program youth with a new offense √ √ √ √ 5 + ** Number/ percentage of program youth exhibiting desired change in targeted behavior Improve prosocial behaviors Select as many as apply from 5A–5D 5A Substance abuse Improve prosocial behaviors The number and percentage of program youth who have exhibited a decrease in substance abuse. Self-report or staff rating are the most likely data sources. B. Number of youth in program C. Percentage (A/B) A. Number of program youth with the noted behavioral change B. Number of youth in program C. Percentage (A/B) DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-38 Outcome Performance Measures (continued) # 5B Outcome Measure School attendance Reporting Term Objective Improve prosocial behaviors Definition The number and percentage of program youth who have exhibited an increase in school attendance. Self-report or staff rating are the most likely data sources. Reporting Format A. Number of program youth with the noted behavioral change Short Long √ √ √ √ √ √ B. Number of youth in program C. Percentage (A/B) 5C Family relationships Improve prosocial behaviors The number and percentage of program youth who have exhibited improved family relationships. Self-report or staff rating are the most likely data sources. A. Number of program youth with the noted behavioral change B. Number of youth in program C. Percentage (A/B) 5D Antisocial behavior Improve prosocial behaviors The number and percentage of program youth who have exhibited a decrease in antisocial behavior. Self-report or staff rating are the most likely data sources. A. Number of program youth with the noted behavioral change B. Number of youth in program C. Percentage (A/B) 6 7 8 + Number/ percentage of program youth completing program requirements Number/ percentage of program families satisfied with program Number/ percentage of program youth satisfied with the program Increase accountability Increase program support Increase program support The number and percentage of program youth who have successfully fulfilled all program obligations and requirements. Program obligations will vary by program but should be a predefined list of requirements or obligations that clients must meet prior to program completion. Program records are the preferred data source. A. Number of program youth who exited the program having completed program requirements The number and percentage of program families who report being satisfied with the program in areas such as staff relations/expertise, general operations, facilities, materials, and service. Self-report data collected using program evaluation or assessment forms are the expected data source. A. Number of program families who report being satisfied with the program The number and percentage of program youth who report being satisfied with the program in areas such as staff relations/expertise, general operations, facilities, materials, and service. Self-report data collected using program evaluation A. Number of program youth who report being satisfied with the program DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation √ B. Number of youth who left the program C. Percentage (A/B) √ B. Number of program families who returned the surveys C. Percentage (A/B) √ B. Number of program youth who returned the surveys C. Percentage (A/B) 5-39 Outcome Performance Measures (continued) # Outcome Measure Reporting Term Objective Definition Reporting Format Short Long or assessment forms are the expected data source. 9 10 11 Number/ percentage of program staff with increased knowledge of program area Number/ percentage of nonprogram personnel with increased knowledge of program area + Number of contributing factors determined from assessment studies Increase program support Increase program support Reduce DMC The number and percentage of program staff who gained a greater knowledge of DMC and DMC-related topics through trainings or other formal learning opportunities. Appropriate for any program whose staff received programrelated training. The program need not have provided the training. Selfreport data collected using training evaluation or assessment forms are the expected data source. A. Number of program staff trained during the reporting period who report increased knowledge The number and percentage of nonprogram personnel, such as representatives from law enforcement, courts, referral agencies, or community members, who gained a greater knowledge of DMC and DMC-related topics through trainings or other formal learning opportunities. The program need not have provided the training. Self-report data collected using training evaluation or assessment forms are the expected data source. A. Number of nonprogram personnel trained during the reporting period who report increased knowledge Assessment studies are conducted to determine the factors contributing to disproportionality at certain juvenile justice system contact points for certain racial/ethnic minority(ies). Count the number of factors in the family, the educational system, the juvenile justice system, and the socioeconomic conditions determined to have contributed to minority overrepresentation at certain juvenile justice system contact points. Number of contributing factors determined from assessment studies. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation √ B. Number of program staff trained during the reporting period and returning surveys C. Percentage (A/B) √ B. Number of nonprogram personnel trained during the reporting period and returning surveys C. Percentage (A/B) √ 5-40 Outcome Performance Measures (continued) # Outcome Measure Reporting Term Short Long Objective Definition Reporting Format 12 + Number of contact points reporting reduction in disproportionality at the state level Reduce DMC Number of contact points reporting significant disproportionality at the state level during the reporting period compared with the last reporting period. Contact points include arrest, referral to juvenile court, diversion, detention, petition filed, found delinquent, probation, secure confinement, and transfer/waiver to adult court. Number of contact points (arrest, referral to juvenile court, diversion, detention, petition filed, found delinquent, probation, secure confinement, and transfer/waiver to adult court) reporting significant disproportionality at the state level during the current reporting period. √ 13 + Number of contact points reporting reduction in disproportionality at the local level Reduce DMC Number of contact points reporting significant disproportionality at the local level during the reporting period compared with the last reporting period. Contact points include arrest, referral to juvenile court, diversion, detention, petition filed, found delinquent, probation, secure confinement, and transfer/waiver to adult court. Number of contact points (arrest, referral to juvenile court, diversion, detention, petition filed, found delinquent, probation, secure confinement, and transfer/waiver to adult court) reporting significant disproportionality at the local level during the current reporting period. √ 14 + Number/ percentage of recommendations from assessment studies implemented Reduce DMC Assessment studies contain multiple recommendations. Count the total number of those chosen for implementation. A. Number of recommendations chosen for implementation √ B. Number of recommendations made C. Percentage (A/B) DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-41 Appendix B: Hiring an Outside Evaluator Interpreting racial data is difficult. Though cost may be an issue, hiring an outside evaluator can facilitate DMC studies. Having a trained evaluator is worthwhile, particularly if the evaluator is sensitive to racial factors in juvenile justice or the criminal justice system. It is essential that the evaluator be on board from the beginning, attend steering committee meetings, be familiar with the juvenile justice system and program evaluation, and be aware of the tasks for which he or she will be responsible. For information on recommended qualifications and reasonable expectations for outside evaluators, see the Juvenile Justice Evaluation Center’s Hiring and Working with an Evaluator (www.jrsa.org/jjec/about/briefing_evaluator.html). Although many projects charged with identifying and reducing DMC will want to consider hiring an outside evaluator to complete a formal evaluation of their strategies, basic knowledge about evaluation and performance measurement issues as they relate to minority overrepresentation is essential for all staff working on a DMC project. One reason is that even if outside evaluators are used, various juvenile justice staff will probably collect the data. In addition, performance measurement can be conducted before or along with evaluation, and juvenile justice staff working on a DMC reduction initiative may choose to make statements about how closely the activities follow expectations or about what outcomes have been achieved. This DMC Technical Assistance Manual will be a useful resource for those conducting their own studies of DMC as well as for those who wish to hire an outside evaluator. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-42 Appendix C: Model Interagency Agreement—Juvenile Information Exchange Office of the Attorney General, State of Colorado General Provisions Each of the parties agrees to: 1. Promote a coordinated effort among agencies and staff to achieve maximum public and school safety, while at the same time maintaining the appropriate level of confidentiality of information. 2. Participate in interagency planning meetings, as appropriate. 3. Adopt and abide by a set of common definitions applicable to the agreement. 4. Assign staff, as appropriate, to participate in information-sharing activities undertaken pursuant to the agreement, and to assess and develop plans for at-risk youth and those involved in the juvenile justice system. 5. Jointly plan and/or provide information and access to training opportunities, when feasible. 6. Develop internal policies and cooperative procedures, as needed, to implement the agreement as effectively as possible. 7. Periodically review all procedures and policies affecting the goals of the agreement, and implement changes as needed. 8. Comply with relevant state and federal law and other applicable local rules that relate to records use, security, dissemination of information, and retention/destruction of records, and request and disseminate information pursuant to the agreement only for purposes authorized by law. 9. Develop and disseminate appropriate internal written policies to ensure that confidential information, including education information and juvenile or criminal justice information, is disseminated only to appropriate and authorized personnel. 10. Develop and maintain a method and procedure of transmitting information pursuant to the agreement that reasonably minimizes any possibility the information will become known to unauthorized persons. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-43 11. Designate specific personnel to respond to requests for information made pursuant to the agreement and to receive information pursuant to the agreement. The name of the designated personnel will be made available to all parties to the agreement. 12. Advise all personnel within the agency who may become aware of information covered by the agreement of the duties and responsibilities articulated in the agreement, and provide them with a means to assure information reaches the designated personnel responsible for responding to requests made pursuant to the agreement. 13. Designate and identify for all parties to the agreement a person who is ultimately responsible for the execution of the agreement. 14. Maintain a record of all information exchanged pursuant to the agreement. 15. Execute the agreement uniformly with respect to all persons, without regard to any person’s race, color, religion, ancestry, or nation origin. DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-44 Appendix D: Sample Consent Forms Denver Juvenile Network Common Informed Consent Form Authorization for Release of Confidential Information I, ____________________________________, on behalf of myself and/or my children and/or wards, ______________________________________ ______________ __________________, (Name of child/ward) (date of birth) (Soc. Sec. No.) ______________________________________ ______________ __________________, (Name of child/ward) (date of birth) (Soc. Sec. No.) ______________________________________ ______________ __________________, (Name of child/ward) (date of birth) (Soc. Sec. No.) ______________________________________ ______________ __________________, (Name of child/ward) (date of birth) (Soc. Sec. No.) hereby authorize the agencies listed in the attachments to release and share among themselves confidential information checked below on a need-to-know basis; for investigatory purposes and case management purposes as defined in the Colorado Children’s Code. _____ Child Welfare Information, e.g., social worker case file; medical, psychological and education, consultation reports; court reports; relinquishment and adoption records. _____ Juvenile Justice Information, e.g., arrest and criminal records, probation records, social and clinical studies, law enforcement records in general. _____ Mental Health Information, e.g., psychiatric and psychological diagnoses, reports and evaluations, treatment recommendations. _____ Education Information, e.g., to include standardized test scores, grades, report cards, attendance, IEPs, counseling, special education, learning disability and diagnoses related thereto, disciplinary, health, and social work records and reports. _____ Medical Information, e.g., records and reports of patient history, diagnoses, evaluations, treatment, including those related to developmental disability (with the exception of HIV and AIDS-related information). DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-45 _____ Vocational Rehabilitation Information, e.g., records and reports of disabilities, evaluations, and recommendations. Other: __________________________________________________________________ with the exception of the following: __________________________________________ ________________________________________________________________________ Purpose of Release: For interagency coordination and case management among those with a need to know, to facilitate investigations, assessments, evaluations, care and treatment, supervision, education, protection, proper disposition or placement of the subject person(s), and other services incidental to the administration of the respective agency programs and in the best interests of the subject person(s). The information exchanged may not be used as evidence in a criminal proceeding nor be used to investigate or prosecute a suspected crime, unless such documents are subpoenaed. This consent automatically ends one year from the date I sign this form, or when the sharing of information is no longer needed to manage or provide services to me, my child(ren), or wards, or when I revoke my consent, whichever is sooner, except to the extent that the program or person authorized to make the disclosure has already acted in reliance on this consent. I understand I may revoke this authorization at any time by signing the revocation statement below and provide this document to the appropriate agencies. Agencies and providers who request information under this release may use a copy or facsimile (FAX) of this form in place of the original signed consent form. This Authorization for information sharing has been explained to me. I have read it (or it was read to me) and understand its provisions. I have been given a reasonable amount of time to ask questions and consider whether to permit sharing of this information. I hereby willingly agree to share of information as described above. Dated: ______________________________ Dated: ___________________________ ____________________________________ Signature of Youth _________________________________ Signature of Parent, Guardian or Authorized Representative Also known as: ______________________ Also known as:___________________ Soc. Sec. of child____________________ Child’s date of birth:________________ ________________________________ (Staff person facilitating authorization) Title/Agency_____________________ DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-46 NOTE: If you choose to modify or revoke this Authorization, you must sign below and provide to the appropriate agency (agencies). I hereby revoke my authorization and consent for release of information to the parties listed on this form. Signed: ____________________________________________________ Date: ______________________________________________________ For questions regarding this form, please call the Denver Juvenile Treatment Network (303–893–6898) DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-47 Denver Juvenile Network Common Informed Consent Form Authorization for Release of Confidential Information (Alcohol and Substance Abuse) I, ____________________________________, on behalf of myself and/or my children and/or wards, ______________________________________ ______________ ______________, (Name of child/ward) (Child’s date of birth) (Soc. Sec. No.) ______________________________________ ______________ ______________, (Name of child/ward) (Child’s date of birth) (Soc. Sec. No.) ______________________________________ ______________ _____________, (Name of child/ward) (Child’s date of birth) (Soc. Sec. No.) ______________________________________ ______________ ______________, (Name of child/ward) (Child’s date of birth) (Soc. Sec. No.) hereby authorize the _____________________________________________________ to (name of drug and/or alcohol treatment program) hereby authorize the agencies listed in the attachments to release and share among themselves confidential information checked below on a need-to-know basis; for investigatory purposes and case management purposes as defined in the Colorado Children’s Code. _____ Alcohol/drug use history reports _____ Consultation reports _____ Transition plan _____ Treatment discharge summary _____ Court history and reports summary _____ Assessment/evaluation _____ Progress/compliance reports _____ Family history; social information _____ Alcohol/drug abuse treatment _____ Psychiatric history and treatment summary _____ Urinalysis results _____ Medical history and treatment summary (except HIV and AIDS-related information) _____ Other: _______________________________________________________________________ with the exception of the following: __________________________________________ _______________________________________________________________________ _______________________________________________________________________ DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-48 Purpose of Release _____ To facilitate alcohol/drug treatment _____ To coordinate medical care _____ To provide follow-up information _____ To comply with court-imposed conditions _____ To inform relative/friend of status _____ To provide information to the Court _____ For interagency coordination and case management among those with a “need to know,” to facilitate investigations, assessments, evaluations, care and treatment, supervision, education, protection, proper disposition or placement of the subject person(s), and other services incidental to the administration of the respective agency programs and in the best interests of the subject person(s). _____ Other: _____________________________________________________________________ The information exchanged may not be used as evidence in a criminal proceeding nor be used to investigate or prosecute a suspected crime unless such documents are subpoenaed through a court order. I understand that my records are protected under the federal regulations governing Confidentiality of Alcohol and Drug Abuse Patient Records, Part 2 of Title 42 of the Code of Federal Regulations and that recipients of this information may share it only in connection with their official duties. This consent automatically ends one year from the date I sign this form, or when the sharing of information is no longer needed to manage or provide services to me, my child(ren), or wards, or when I revoke my consent, whichever is sooner, except to the extent that the program or person authorized to make the disclosure has already acted in reliance on this consent. I understand I may revoke this authorization at any time. Agencies and providers who request information under this release may use a copy or facsimile (FAX) of this form in place of the original signed consent form. This Authorization for information sharing has been explained to me. I have read it (or it was read to me) and understand its provisions. I have been given a reasonable amount of time to ask questions and consider whether to permit sharing of this information. I hereby willingly agree to share information as described above. Dated: ________________________________ I _______ consent _______ do not consent (check one) that this information may be shared with my parent or guardian. __________________________________ Signature of Youth _________________________________ Signature of Parent, Guardian or Authorized Representative DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-49 Also known as: _______________________ Also known as: _________________ Soc. Sec. #____________________________ Date of Birth :__________________ ___________________________________ ___________________________________ (Staff person facilitating this Authorization) (Staff person facilitating this Authorization) Title/Agency ________________________ Title/Agency _______________________ NOTE: If you choose to modify or revoke this Authorization, you must sign below and provide to the appropriate agency (agencies). I hereby revoke my authorization and consent for release of information to the parties listed on this form. Signed: ____________________________________________________ Date: ______________________________________________________ DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-50 Denver Juvenile Treatment Network Common Informed Consent Form Public System Member Agencies Colorado State Department of Human Services, including: Division of Alcohol and Drug Abuse Services Division of Child Welfare Services Division of Developmental Disabilities Services Division of Mental Health Services Division of Vocational Rehabilitation Services Division of Youth Corrections Colorado State Judicial, including: Denver Juvenile Court and Probation Denver Juvenile TASC Denver Juvenile Community Assessment Center Denver City Attorney’s Office Denver County Court and Probation Denver Department of Human Services Denver District Attorney’s Office Denver Department of Safety, including: Denver Police Department Denver Sheriff’s Department Denver Fire Department Denver Public Schools Office of the State Public Defender DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-51 Denver Network Service Providers Date: ____________________________________ List private agencies name, telephone and fax # To add agencies not listed above, note date added and initial (guardian/child). Other Agencies: Initial _______ _______ _______ _______ Agency Name ________________________________________ ________________________________________ ________________________________________ ________________________________________ DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation Date Added _____________ _____________ _____________ _____________ 5-52 Common Informed Consent Form Request for Information Records being requested by: Agency Name:___________________________________________________________ Address: _________________________________________ Telephone: ____________ Contact: __________________________________________ Date: ________________ This consent to release information is limited to information that your agency generates only. (Please state the name or title of the individual or the name of the organization, with the address, which is the custodian of the records being requested.) TO: (Please identify the person who is the subject of the request, with identifying information): RE: (Name) DATE OF BIRTH: In accordance with the attached consent to share information form, please forward the following types of information concerning (please be specific): The information will be used for the purpose(s) of (please be specific): The person named above or his/her legal representative has been notified of this request by mail/phone/in person (please circle one) on this date: __________________________. (TO BE COMPLETED BY THE CUSTODIAN OF THE RECORD ) Action Taken: Date: ____________________________ Signature: _________________________ DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-53 Appendix E: Sample Participant Locator Form Participant Locator Information Full Name Birthday SS Number Program Start Date End Date Mother Name Address City State Zip code Phone number Father Name Address City State Zip code Phone number Name, address and telephone number of a relative or close friend who does not live with you and who will always know how to contact you. Name Address City State Zip code Phone number DMC Technical Assistance Manual, 4th Edition • Chapter 5: Evaluation 5-54 Chapter 6: Federal, State, and Local Partnerships To Reduce Disproportionate Minority Contact Heidi Hsia, Michael Wilson, Kim Wilson, and James Frabutt* Lesson 6 in the introduction to this manual stresses that DMC reduction requires strong partnerships. This chapter describes ways to form and strengthen partnerships among federal, state, and local DMC reduction efforts. OJJDP’s Role OJJDP, as the federal agency charged with implementing the Juvenile Justice and Delinquency Prevention (JJDP) Act of 2002, has undertaken the activities described below to help states that participate in the Formula Grants Program meet the DMC core requirement in Section 223(a)(22) of the Act. Determining States’ Compliance With the DMC Core Requirement All states and territories, except for Puerto Rico (which the U.S. Census Bureau has exempted from reporting racial statistics), shall submit DMC identification spreadsheets as part of the DMC compliance plan in their 3-year plans. When a state determines that DMC exists, it shall provide a DMC compliance plan in its 3-year plan and in plan updates. A state’s annual DMC compliance plan must discuss the status of and progress made for each of the planned activities in the prior compliance plan. States with significant local DMC reduction efforts should complete this section by locality as well. The DMC compliance plan should also include a plan for the following 3 years (as in a 3year plan) or year (as in a plan update). The plan should include specific activities in data collection, data system improvement, assessment, programmatic and systems improvement strategies, evaluation, and monitoring activities, as appropriate. The plan must also specify the timeline, funding amount, and funding source(s) designated to conduct each of the planned activities. Moreover, OJJDP requires states to submit updated DMC data in their 3-year plan for at least three jurisdictions with the highest minority concentrations or, preferably, the localities with focused DMC reduction efforts. The 3-year plan should also discuss the Relative Rates Indexes (RRIs) obtained, compare the updated data with data obtained in earlier years, and illustrate how the data inform/guide the state’s 3-year DMC compliance plan. * About the authors: Heidi Hsia, Ph.D., is the Disproportionate Minority Contact Coordinator at OJJDP. Dr. Hsia wrote the section on federal support to states. Michael Wilson and Kim Wilson of the North Carolina Governors Commission and James Frabutt, Ph.D., of the University of North Carolina at Greensborough wrote the section on North Carolina’s DMC initiative. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-1 OJJDP’s annual determination of states’ compliance with the DMC core requirement is based on the completeness of their DMC compliance plans; the demonstration of actual, systematic, continuing, and good-faith implementation of their planned activities; and the progress reported each year. OJJDP has standardized the compliance determination process in recent years. In its review of the current plan, OJJDP compares the state’s report of progress made with its prior year plan. OJJDP staff also determine whether the data obtained drive the plan and whether the state has designated adequate resources for the planned DMC reduction activities. Each determination letter, signed by the OJJDP Administrator, outlines the state’s accomplishments in the prior year, the planned activities for the following year (or 3 years), and recommendations for enhancements in the area of DMC reduction. A DMC compliance determination letter, therefore, is not the end of OJJDP’s work with the state regarding DMC for that year but is used as a technical assistance tool for the beginning of a continuous followup with the state regarding its DMC reduction efforts throughout the year. Holding States Accountable for Noncompliance The JJDP Act of 2002 stipulates that OJJDP will reduce a state’s Formula Grant allocation by 20 percent for each core requirements for which the state was found to be not in compliance in the previous year. During the Formula Grant application review process, OJJDP works diligently with states whose DMC compliance plans the Office deems are inadequate, with the goal of improving their plans. If a state does not meet the required standards by September 30 of that year, OJJDP will make a final determination of noncompliance, specifying a reduction of 20 percent of that state’s Formula Grant allocation in the subsequent year. OJJDP determined that two states in FY 2004 and one state in FY 2005 were noncompliant with the DMC core requirement and reduced their formula allocations in the following year as a consequence. Providing Financial and Technical Assistance Support to Annual DMC Conferences The Coalition for Juvenile Justice (CJJ), with cooperative agreement awards from OJJDP, has emphasized DMC in its training activities. In 1996, CJJ held the first National DMC Planning and Strategy Meeting and, with OJJDP financial support, has since made the DMC conference an annual event. Three of the 10 conferences so far have focused on a specific ethnicity: American Indian youth in 2000, Hispanic youth in 2001, and African American youth in 2002. With OJJDP support, CJJ published and broadly disseminated two conference reports to augment the 2000 and 2001 conferences: Enlarging the Healing Circle: Juvenile Justice for American Indian Youth and Esperanza: Awakening to the Needs of Latino Youth. In addition, these annual conferences have provided opportunities for State Advisory Group members and state juvenile justice staff to learn about the DMC reduction work of the Juvenile Detention Alternatives Initiative from the Annie E. Casey Foundation, the Burns Institute’s approach to reducing disproportionality at detention, the Youth Law Center’s Building Blocks for Youth, and the Graduated Sanctions Project of the National Council of Juvenile and Family Court Judges (also funded by OJJDP), among others. OJJDP’s DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-2 leadership has consistently supported these conferences with opening remarks. In recent years, OJJDP staff have also participated in the conference planning committees and made presentations at the conferences. Providing Training and Technical Assistance to State and Local DMC Reduction Efforts OJJDP, through its training and technical assistance contract providers, offers onsite technical assistance at no cost to the states and localities. OJJDP responded to 30 onsite training and technical assistance requests in FY 2005, a significant increase from 17 in FY 2004. In addition, OJJDP provides phone and e-mail consultations in the DMC area as a cost-effective way to maximize access to technical assistance. DMC has also become an integral part of OJJDP-sponsored regional and national training conferences in recent years. The one-to-one consultation sessions on data collection, interpretation, and use offered during the training conferences have been welcome opportunities for conference participants to address their unique concerns. Developing and Implementing Performance Measures To Assess DMC Reduction Activities, Providing Technical Assistance/Tools To Increase State/Local Evaluation Capacity As part of its development of a performance measurement system for the Formula Grants program, OJJDP has created a DMC logic model outlining goals, objectives, activities, and output and outcome measures of these activities. Chapter 5 of this manual discusses and links to the logic model tool. Further, the Juvenile Justice Evaluation Center of the Justice Research and Statistics Association, under contract with OJJDP, produced a guidebook, Seven Steps To Develop and Evaluate Strategies To Reduce Disproportionate Minority Contact, in January 2005. This publication is available at www.jrsa.org/jjec/about/dmc_guidebook.html. Establishing and Maintaining a DMC Coordinators’ Listserv OJJDP has long promoted focused DMC efforts at the state level through the establishment of state-level DMC coordinators to guide and support local efforts in their respective states. Based on the position descriptions from California, Colorado, Iowa, Kentucky, Missouri, Montana, and Pennsylvania, OJJDP compiled a sample state DMC coordinator position description in 2001 to facilitate establishment of the position in other states. More than 30 state-level DMC coordinator positions (the number changes with staff turnover) and a number of local-level DMC coordinator positions have been created around the country. OJJDP has established and maintains a DMC coordinators’ listserv to facilitate communication with and among the DMC coordinators. When OJJDP communicates with state-level DMC coordinators, all state juvenile justice specialists are copied to ensure that they are informed even if their state has not designated a DMC coordinator. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-3 Establishing and Maintaining a DMC Web Page An important part of OJJDP’s support to state and localities is its DMC Web page (www.ojjdp.ncjrs.gov/dmc/), a one-stop shop that brings together DMC-related information for the field. The site contains a DMC chronology, various tools, training videos and other resources, a library of significant state DMC reports, DMC-related publications, state and (if available) local DMC contacts, and links to other large-scale DMC activities, such as the Building Blocks for Youth, Annie E. Casey and Multnomah County Detention Reform Initiative, and the W. Haywood Burns Institute for Juvenile Justice Fairness and Equity. The DMC Web page was launched on July 27, 2001. The average number of visits per month to the page has nearly doubled from 1,491 in 2001 to 2,904 in 2005. The average number of unique visitors per month also has increased dramatically, from 544 in 2001 to 765 in 2005.1 Developing a DMC Web-Based Data Entry System To Be Used With the DMC Technical Assistance Manual This Web-based data entry tool enables localities and states to enter raw data concerning the volume of activities by race and ethnicity at different juvenile justice decision points to calculate the existence and extent of DMC as expressed by the relative rate indexes. It provides a central repository of state and local data across the country and facilitates within the state or within localities comparisons of DMC changes over time. States and localities can access the data entry system at www.dsgonline.com/dmc. Publishing the DMC Technical Assistance Manual (3rd Edition) Electronically The manual incorporates lessons learned in the past years to provide up-to-date guidance for ongoing DMC reduction efforts nationwide. The manual content was featured in the all-day DMC preconference at the 2005 OJJDP National Conference. (Interested parties can access the DMC preconference slides and its video recording from OJJDP’s DMC Web page at www.ojjdp.ncjrs.gov/dmc.) The manual, available through the DMC Web page, can reach a large readership and will be updated periodically. Making Direct Awards To Promote Innovative Local DMC Reduction Projects In FY 2004, OJJDP made a direct award to the Youth Law Center’s Washington, DC, office to address two problems that have proven difficult for the states: collecting accurate data on Hispanic youth in the juvenile justice system and reducing DMC at critical decision points in the system for these youth. The Center chose two sites for this project over a 3-year period: Travis County, Texas, and Reno, Nevada. The first site focuses on DMC and Hispanic youth in the juvenile justice system; the second site focuses on DMC and Hispanic and African American youth. In both sites, the project gives particular attention to Hispanic youth and the accuracy of data on these youth. With the closing of its Washington office in February 2006, the Youth Law Center’s headquarters (and only office) in San Francisco took over administration of this award. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-4 Instituting Conference Calls Among DMC Coordinators In response to requests from DMC coordinators, and modeled after the bimonthly juvenile justice specialists’ and compliance monitors’ conference calls, OJJDP instituted DMC coordinators’ conference calls in March 2006 on a bimonthly and as-needed basis. These calls provide a regular forum for DMC coordinators to share information, strategies, and concerns and to problem-solve. Planned Activities OJJDP’s planned activities that will soon be available include: An Annual Summary of DMC Reduction Efforts by State OJJDP’s three publications, Disproportionate Confinement of Minority Juveniles in Secure Facilities: 1996 National Report, Disproportionate Minority Confinement: 1997 Update, and Disproportionate Minority Confinement: 2002 Update,2 all summarized DMC reduction efforts nationally; the latter two publications provided case studies of two states’ multiyear, systematic efforts to address DMC. In response to states’ requests, OJJDP will make available an annual summary of DMC reduction activities by state in the summer of each year beginning with 2006—after OJJDP has reviewed the DMC compliance plans in the states’ Formula Grants applications. States can reference this document throughout the year and consult their peers regarding strategies of interest. Training of Trainers on a DMC Community Planning Curriculum Increasingly, OJJDP recognizes the need to target DMC reduction efforts at the local level and will make available quality training to interested sites. State and local DMC coordinators are perfectly positioned to provide such training to sites within their states. Many of them have expressed a desire to receive the necessary training to become trainers for their states. In addition, OJJDP is looking at training consultants to assist sites in states that have not designated DMC coordinators (spring of 2007). Training of Trainers on a DMC Curriculum for Juvenile Justice Professionals This curriculum intends to sensitize juvenile justice professionals about the existence of DMC and what they, in their respective roles, can contribute to DMC reduction. OJJDP has developed this curriculum to facilitate state and local efforts in this regard. OJJDP expects that increasing state and local capacity to provide such training within their jurisdictions will produce a positive and broad impact in the juvenile justice field over time. OJJDP plans to field test this curriculum in the summer of 2007, with the first training targeted for the early spring of 2008. The remainder of this chapter presents a case example describing how North Carolina used federal and state resources to support its local DMC reduction efforts (vertical partnerships). The case example also demonstrates important horizontal partnerships established at the state and local levels. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-5 North Carolina’s DMC Reduction Initiative Partnerships at the State Level In 2001, the Governor’s Crime Commission (GCC) created a permanent DMC Committee that recently became a subcommittee of North Carolina’s State Advisory Group, the Juvenile Justice Planning Committee (JJPC). The DMC Subcommittee, which is staffed by a full-time DMC coordinator, provides recommendations to the JJPC regarding efforts to reduce DMC. The subcommittee is composed of representatives from local law enforcement agencies; the state’s Department of Juvenile Justice and Delinquency Prevention (DJJDP), Department of Corrections, Department of Public Instruction, Administrative Office of the Court, and Department of Health and Human Services; and citizen appointments. The group recently invited new members so that each minority group identified in the federal JJDP Act is represented. Current subcommittee strategies to address DMC in North Carolina include the following: Demonstration projects: working with demonstration projects in four counties to provide resources, technical assistance, and ongoing monitoring and evaluation of programs and activities designed to reduce DMC in these jurisdictions. This is described in more detail in the next section. Collaboration: collaborating with the North Carolina DJJDP to develop a system to collect accurate data disaggregated by race. The state has identified incomplete and inconsistent data and the need for improved juvenile justice information systems as key challenges to addressing DMC. The GCC has worked in conjunction with the DJJDP to develop a statewide system (NC-JOIN) that allows the state to collect statistical data in a more uniform and consistent manner. Data collected from NC-JOIN enables the GCC to compute the relative rate index of minority youth representation at all juvenile justice system contact points, as OJJDP requires. The GCC’s statistical analysis center is also determining how geographic information system (GIS) mapping can be used to target DMC efforts in specific jurisdictions. DMC awareness: increasing the awareness of DMC in the juvenile justice system and educating the public, juvenile justice professionals, as well as the GCC. This is accomplished through conference presentations, development and dissemination of materials, and technical assistance resources provided by OJJDP. State’s Partnerships With Demonstration Counties The DMC Subcommittee determined that to reduce DMC, the state must work with local representatives. As part of the state’s FY 2003 plan to address DMC, the subcommittee was to partner with four counties to address the issue of minority overrepresentation in the juvenile justice system. The criteria the subcommittee used to choose the counties included minority arrest rates, minority youth detention admission rates, youth development center admission data, suspension and expulsion rates for minority youth, DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-6 and geographic distribution. The subcommittee also considered current resources in each jurisdiction and whether each county had existing prevention or intervention programs for youth. The subcommittee selected New Hanover, Union, Guilford, and Forsyth counties to partner with the GCC in developing county-specific plans to address minority overrepresentation. Each county established a steering committee that serves as the primary point of contact with the GCC. Meetings are held in each jurisdiction to discuss how each would address issues specific to its jurisdiction. The DMC coordinator, with the assistance of the juvenile justice specialist, provides technical assistance to these demonstration sites on grant writing, strategic planning, and mission development. During FY 2004, the state awarded the counties planning grants so they could mobilize stakeholders and begin the process of analyzing their specific DMC issues. Each county now has a working DMC steering committee, with membership drawn from the community, law enforcement, courts, the school system, and private citizens. Each steering committee met monthly and was charged with developing a comprehensive DMC reduction plan in a process that included gathering data and assessing the extent of minority overrepresentation in the county. The planning grants ended in June and November 2005, and each county established a comprehensive, research-based DMC reduction plan in 2005. The intent of the comprehensive plan is to provide the county with clear direction on how it will implement DMC reduction activities in the following year. With guidance from the GCC, the counties are now using current grant funds to implement the specific strategies outlined in their plans. Partnerships and Activities in Demonstration Counties Although activities in the four demonstration counties vary, they all share three important traits: Each has designated a local DMC coordinator who will build and maintain local partnerships and maximize their benefits. Each has signed and publicized a memorandum of understanding (MOU) of all partner agencies to demonstrate the strong commitment of the partner agencies and ensure the longevity of the local partnership. Each local partnerships includes a local university, for its research expertise and perceived objectivity in the eyes of data providers such as schools, law enforcement, and juvenile justice. Brief descriptions of the four demonstration counties’ DMC initiatives follow. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-7 New Hanover County New Hanover County is located in the coastal area. Among its 172,780 residents, 16,218 are between the ages of 10 and 17 years. The median annual household income is $40,170, with 13 percent of the population living in poverty. New Hanover County’s DMC steering committee currently consists of 24 representatives from community and state agencies, institutions, and the faith community. Once the committee represented all the key stakeholders needed to effectively address DMC, it drafted an MOU and held a public signing of the document to acknowledge that DMC is an issue in the county. Further, the MOU showed the commitment of the DMC steering committee to reduce minority overrepresentation. This public MOU signing was featured in one of the county’s local newspapers. New Hanover County has hired a DMC project coordinator to lead and coordinate the DMC demonstration project. In addition, the DMC steering committee has contracted with the University of North Carolina-Wilmington to identify the appropriate instruments needed to collect DMC-specific data at the various decision points in the county’s juvenile justice system. The data collection team also evaluated existing instruments to determine their effectiveness in collecting the necessary statistics to measure minority overrepresentation. Using the data collected from the research team, the DMC steering committee developed a comprehensive county plan that includes specific strategies and activities the county will implement to reduce the number of minority youth involved in the system. This plan also includes an evaluation of existing programs within the county, with recommendations for enhancement, redirection, and the institutionalization of diversionary programs. For more information, contact: Patricia Melvin DMC Project Director Assistant County Manager New Hanover County 320 Chestnut Street Wilmington, NC 28401 910–341–7184 pmelvin@nhcgov.com or Nequan Peartree DMC Coordinator 718 S. Third Street Wilmington, NC 28410 910–342–2512 npeartree@nhcgov.com DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-8 Union County Union County is located in the southwest area of the state. Among its 158,000 residents, 17,560 are between the ages of 10 and 17 years. The median annual household income is $50,640, with 8 percent of the population living in poverty. Union County’s objectives are similar to those of New Hanover County, where the primary focus of its planning process included mobilizing a steering committee and developing a data system that would provide baseline DMC-specific data. The county DMC reduction plan is a direct result of the information obtained from the county’s data collection efforts. Union County has developed a DMC steering committee that is a subcommittee of the county’s Juvenile Crime Prevention Council.3 The county has also hired a DMC project coordinator who oversees the county’s efforts to reduce DMC. The county held a public signing of its MOU, at which representatives from the police department, juvenile probation department, public school system, district attorney’s office, community-based organizations, and social services agencies gathered to show their commitment to addressing DMC in the county. The county has focused the majority of its efforts on educating itself about the issue of DMC and possible factors that may have led to minority youth being overrepresented in the juvenile justice system. The county has also focused on identifying existing resources and creating a local data mapping system that will help the steering committee target where new services are needed to reduce the number of minority youth entering the system. Union County has partnered with Wingate University to coordinate a data collection system that it will use as a baseline evaluation system for the county. The data associated with juvenile arrests, adjudication, and disposition have been used to determine where disparities in decisionmaking may exist. The county will change policies and procedures that may have put minority youth at a greater risk for entering the juvenile justice system. For more information, contact: Shawn Keith DMC Project Director Juvenile Court Counselor P.O. Box 1091 Monroe, NC 28111-1091 704–289–4169 Shawn.keith@ncmail.net or Rebecca Smith DMC Coordinator Juvenile Court Counselor P.O. Box 1091 Monroe, NC 28111-1091 704–289–4169 Becky.smith@ncmail.net DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-9 Guilford County Guilford County, as part of the 11-county Piedmont Triad region of North Carolina (population: 1.27 million), is centered along the Piedmont Industrial Crescent stretching from Raleigh to Charlotte. Guilford County has the third largest population in North Carolina, with 438,520 residents in 658 square miles; 48,460 of its residents are age 10 to 17 years. The median annual household income is $42,620, and 10.6 percent of residents live in poverty. Guilford County has two major cities—Greensboro and High Point. The racial breakdown of the county is: non-Hispanic white, 62.9 percent; black or African American, 29.3 percent; Hispanic/Latino, 3.8 percent; Asian, 2.4 percent; other, 1.8 percent. North Carolina, and Guilford County in particular, are undergoing major demographic shifts as increasing numbers of immigrants and refugees settle in the state. The Triad region was built on a manufacturing economy of tobacco, textiles, and furniture. Today, however, medicine, technology, banking, and higher education fuel the region’s rapid growth. In the future, commercial biotechnology is poised to become a key driver in the region’s economic transition. Guilford County has created a DMC steering committee and has partnered with the Center for Youth, Family and Community Partnerships at the University of North Carolina at Greensboro (UNCG). The center has hired a part-time DMC project coordinator to oversee day-to-day operations of the steering committee. For the planning phase of this project, representatives from UNCG, the project coordinator, and the DMC steering committee collected and analyzed local DMC data to develop intervention and prevention activities. Guilford County developed a data mapping system to assess the extent of DMC and a county plan to address minority overrepresentation. Guilford County also held a public signing of its DMC MOU, which received media coverage in one local newspaper. The Guilford County DMC steering committee hosted a training session, ―Undoing Racism,™‖ for its members to facilitate common understanding of the institutional issues that impact minority overrepresentation in the system. The county’s goals for its DMC project are similar to the other counties’ in that they are focused on achieving a consensus on a locally meaningful definition and identification of the DMC issue in order to plan prevention and intervention activities for the upcoming year. Guilford County’s DMC planning process includes issue definition and awareness, assessment and data analysis, and prevention/intervention activities. The DMC steering committee held focus groups, interviews, and stakeholder visits to identify factors that contribute to DMC in the county. This process was important in identifying potential resistance to understanding and acceptance of the DMC issue by some. The county has also completed an inventory of youth-serving resources to identify possible gaps in prevention and intervention services. The DMC steering committee has also completed a comprehensive suspension and expulsion report entitled Suspensions in Guilford County Schools, 2003-2004: Using DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-10 Rates to Examine Race and School Effects (www.news-record.com/legacy/ news/indepth/susrate.pdf). This report measures the extent of disproportionate minority suspensions in county schools using the relative rate index. For each school, data included the total student body membership, total number of students of each race or ethnicity (American Indian, Asian, black, Hispanic, multiethnic, and white), total number of short- and long-term suspensions, and the total number for each racial category. The data in this report revealed that some schools had great disparity in the suspension rates for minority and nonminority students, whereas other schools had little disparity. The county plans to assess all of the data collected; share it with the university, DJJDP, county schools, and law enforcement partners; and develop a portfolio of intervention and prevention efforts for local implementation. See this chapter’s appendix for an indepth profile of the Guilford County DMC reduction initiative. For more information, contact: James Frabutt, Ph.D. DMC Project Director 330 South Greene Street Suite 200 Greensborough, NC 27402 336–217–9736 jmfrabut@ung.edu or Mary Kendrick DMC Coordinator 41 McNutt Building, UNCG Greensboro, NC 27402-6170 336–317–6451 marykendrick1@yahoo.com Forsyth County Forsyth County is located in the central Piedmont area. Among its 324,360 residents, 36,497 are between the ages of 10 and 17 years. The median annual household income is $42,100, with 11 percent of the population living in poverty. Forsyth County began addressing overrepresentation of minority youth more than 10 years ago but was unable to put a strategic focus in place. Three years ago, the county’s Juvenile Crime Prevention Council began a 3-year initiative to identify community systems and local data that would reduce DMC. In an effort to use community input to address DMC, Forsyth County conducted a series of youth focus groups to identify how people at the local level view the issues surrounding minority overrepresentation. The focus groups discussed concerns related to the family and the community and how each has an integral role in the involvement of youth in the juvenile justice system. The county has incorporated information from these focus groups into its overall county plan, which shows the importance of conducting DMC reduction efforts at the local level. Forsyth County has partnered with a researcher from Winston-Salem State University to analyze the youth who are currently confined in the state’s youth development centers. The research identified variables that contributed to confinement and developed an DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-11 intervention/prevention plan based on these variables, with the goal of reducing the rate at which youth are entering the system. For more information, contact: Tonya Atkins or Project Director Forsyth Council for Children and Families 601 North Cherry Street Suite 250 Winston-Salem, NC 27101 336–724–2831 Tonya@fccf.us Tennille Pratt DMC Coordinator Forsyth Council for Children and Families 601 North Cherry Street Suite 250 Winston-Salem, NC 27101 336–724–2831 Tennille@fccf.us Future Plans for North Carolina’s DMC Initiative As each county moves forward with its individualized DMC reduction plan, it is important to this initiative that activities are measured, evaluated, and eventually replicated. OJJDP requires all states addressing DMC to report on specific performance measures to ensure that DMC reduction activities are data driven and outcome based. Each county is currently charged with measuring the performance of its activities and documenting how each implemented strategy has resulted in a reduction of minority youth who have contact with the juvenile justice system. As the demonstration counties move into the evaluation and monitoring stage of their DMC reduction plans, the DMC Subcommittee expects to replicate their strategies and activities, sharing them with other counties in the state. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-12 Endnotes 1. Visits are defined as all the activities, from beginning to end, of one visitor to a Web site. Unique visitors are individuals who have visited a Web site at least once in a fixed time. 2. See H.M. Hsia, G.S. Bridges, and R. McHale, Disproportionate Minority Confinement: 2002 Update, Program Summary, Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. This summary and other DMC-related publications are available online at www.ojjdp.ncjrs.gov/dmc/pubs/index.html. 3. Each county in North Carolina has an established Juvenile Crime Prevention Council that receives funds from the North Carolina Department of Juvenile Justice and Delinquency Prevention. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-13 Appendix: Guilford County’s DMC Reduction Initiative A Community-Academic Collaboration To Drive the DMC Project The Guilford County1 DMC project began with the convening of a group of representatives from the local Juvenile Crime Prevention Council (JCPC) in November 2003. This group invited the University of North Carolina at Greensboro’s Center for Youth, Family, and Community Partnerships (CYFCP), which has been involved with juvenile justice issues in Guilford County over the past several years (Forsbrey, Frabutt, and Smith, 2005; MacKinnon-Lewis and Frabutt 2001; Shelton, Frabutt, and Arbuckle, 2003), to write a DMC planning proposal to the North Carolina Governor’s Crime Commission. Since the mission of the CYFCP is to build the capacity of families, service providers, researchers, teachers, and communities to ensure the health and well-being of children, the aims and scope of the project were a natural fit with the Center’s experience and existing portfolio of initiatives. Immediately, CYFCP recommended that the working group expand to become a larger committee for the DMC process. The group now includes representatives from county organizations that address concerns of children and youth (police departments from High Point and Greensboro, the Guilford County Sheriff’s Office, Guilford County Schools, the county’s Department of Social Services and Department of Mental Health and Public Health, district court judges, nonprofit organizations, and parents of youth involved with the juvenile justice system). Since July 2004, a cohesive, representative, and action-oriented committee has executed the Guilford County DMC Planning Grant (see table 1 for a listing of participating agencies). A project management team (including a half-time project coordinator, a quarter-time project director, and a graduate research assistant) based at CYFCP has served as the central convening, organizing, and planning arm for the project. This team issues monthly reports to the Guilford County JCPC on DMC committee activities. As one of its first project activities to raise awareness of the societal context of DMC issues, the committee participated in ―Undoing Racism‖™ training. The People’s Institute for Survival and Beyond, based in New Orleans, Louisiana, provided the training, which uses dialog, reflection, role-playing, strategic planning, and presentations. The intensive process challenges participants to analyze the structures of power and privilege that hinder social equity and prepares them to be effective organizers for justice. More than 20 DMC Committee members attended this event in June 2004. Another milestone event occurred in September 2004, when members of the DMC Committee joined with the county school superintendent, the police chiefs of High Point and Greensboro, the county sheriff, the chief juvenile court counselor for the Department DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-14 of Juvenile Justice and Delinquency Prevention, the chief district court judge, the public defender, the district attorney, and the director of the Department of Social Services in signing a memorandum of understanding (MOU). This document formally and publicly acknowledged their collective commitment to identifying and addressing DMC in Guilford County. Moreover, signatories agreed to give serious consideration to subsequent recommendations following an analysis of agency data. Table 1: DMC Committee Members and Agency Affiliations Alcohol and Drug Services Black Child Development Center for Youth, Family, and Community Partnerships Department of Juvenile Justice and Delinquency Prevention Department of Social Services District Court Judges Faithworks Ministries Family Life Council Family Services of the Piedmont Governor’s Crime Commission Greensboro City Council Greensboro Education and Development Council Greensboro Housing Authority Greensboro Lifeskills Center Greensboro Parks and Recreation Greensboro Police Department Guilford Center Guilford County Department of Juvenile Court Alternatives Guilford County Manager’s Office Guilford County Schools Guilford County Sheriff’s Department Guilford Education Alliance Guilford Technical Community College High Point Parks and Recreation High Point Police Department Juvenile Crime Prevention Council NC A&T University North Carolina Office of the Juvenile Defender One Step Further Parent Representatives United Way Win-Win Resolutions Youth Focus YWCA of Greensboro A critical working group derived from the overall DMC Committee was the Resource and Needs Subcommittee, chaired by a parent advocate and representative. The Resource and Needs Subcommittee was tasked with identifying and cataloging a local continuum of services—both prevention and intervention—that could reduce DMC. Moreover, by identifying the array of existing services, the subcommittee would gain a better understanding of services that were needed but were currently unavailable in the county. The listing was intended as a resource for parents, service providers, school staff, law enforcement, and youth-serving organizations seeking appropriate referrals. Drawing from resources such as United Way’s 211 listing of community-based and governmental services, existing program documentation, and committee members’ suggestions, the subcommittee produced a 29-page list of supportive services for youth. The list of services and resources is organized according to major categories, such as afterschool programs, mentoring, counseling, and substance abuse. The document contains contact information for each program or service, a short description of program content, the intended audience, and active hyperlinks for programs or services that have Web pages. The document was posted to Web sites of agencies serving youth throughout the county (e.g., Guilford Education Alliance [http://guilfordeducationalliance.org/links.htm] and UNCG Center for Youth, Family, and Community Partnerships) and will be routinely distributed to parents, school administrators, law enforcement agencies, juvenile court counselors, and other youth service providers in both hard copy and electronic format. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-15 The second major focus of the county’s efforts during the planning grant year was to collect, analyze, and summarize relevant data that would inform the county’s understanding of the dynamics surrounding juvenile justice decision points impacting DMC. Over the course of the planning grant, the Data Subcommittee worked in concert with the Focus Group Subcommittee to compile both quantitative and qualitative data that would provide insight into the community’s understanding of the DMC issue. The following section reviews those data sources (i.e., schools, law enforcement, juvenile justice, and focus groups) in more detail. A Data-Driven Process To Inform Action School Data Numerous investigations have documented the link between school suspensions and subsequent entry into the juvenile and criminal justice systems (e.g., Mendel 2003; Wald and Losen 2003). Therefore, one of the DMC Committee’s key goals was to measure the extent of disproportionate black suspensions in Guilford County schools. One way to do that was to compare black student suspensions with white student suspensions. Such comparisons can be done in various ways. The measure that OJJDP uses to assess disproportionate minority contact within the juvenile justice system is called a Relative Rate Index (RRI). The RRI is preferable to other measures that are affected by the relative size of minority youth populations and the number of different minority populations to be compared. The RRI method reduces statistical bias, allows accurate comparisons, and can be used to compare multiple racial and ethnic groups. For these reasons, the DMC Committee chose to use this method to analyze suspensions in the county’s school system. The Chief Student Services Officer for the county school system provided the DMC Management Team with suspension data that the Data Warehouse for Guilford County Schools had compiled during the 2003–2004 school year. Data were organized by school, race, and ethnicity. For each school, the team reviewed data that included total student body membership, total number of students of each race or ethnicity (categorized as American Indian, Asian, black, Hispanic, multiethnic, and white), total number of shortand long-term suspensions for the school, and total number for each racial or ethnic group within the school. Importantly, data were unduplicated, which means every count represented a different student rather than multiple suspensions for individual students. These data allowed the team to examine disparities in black and white suspension rates for each school. The team calculated suspension rates and relative rate indexes for every school in the county, comparing the short- and long-term suspension rates of black students with those of white students. The team calculated suspension rates for each race by dividing the number of short- or long-term suspensions for each race by the total number of students of that race and multiplying by 100. Next, the team calculated the RRI for each school by dividing black short- and long-term suspension rates by white short- and long-term suspension rates. For instance, if school XYZ had a black short-term suspension rate of DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-16 15.1, the team would divide that suspension rate by the white short-term suspension rate of 5.8 and arrive at an RRI of approximately 2.6. This means black students were shortterm suspended at 2.6 times the rate of white students; or, for every white student suspended, 2.6 black students were suspended. The team presented the findings in several tables, arranged by school level, beginning with data for elementary schools, then middle schools, high schools, and other schools like middle colleges and multilevel schools (see table 2 for sample school data). School names and total student membership were listed on the left, and short- and long-term suspension rates were broken down for white and black students in the cells of the tables. RRIs for short- and long-term suspensions were provided for each school in the far right columns. To protect the privacy of individual students, the team reported only rates, rather than frequencies, for each school. Table 2: Guilford County High School’s Short- and Long-Term Suspension Rates, 2003–2004 White Rates Schools (student membership) Black Rates Relative Rate Index Short Term Long Term Short Term Long Term Short Term Long Term 4.4 0.6 22.7 3.5 5.2 5.8 18.3 3.3 30.4 6.0 1.7 1.8 Grimsley (1,738) 1.1 0.1 20.0 4.4 18.2 44.0 Southeast (1,230) 12.7 0.8 19.7 2.5 1.6 3.1 9.3 0.4 27.0 2.7 2.9 6.8 Andrews (1,166) Eastern (891) Western (1,299) A review of the tabular data indicates that although some schools have relatively high rates of black suspensions compared with white suspensions, other schools have very little (if any) problem with disproportionate black suspensions. Discrepancies between black and white suspensions were seen at all school levels, including elementary, middle, and high school. Although no students were long-term suspended from elementary schools, elementary schools reported some of the largest racial disparities in short-term suspension rates. Observations such as these indicate that routinely calculating an RRI will enable schools to proactively monitor racial and ethnic disparities in suspension rates and take steps to address imbalances before they become larger problems. Moreover, schools that do not currently exhibit issues with disproportionate black suspensions will benefit from monitoring their RRI as much as schools that are actively working to reduce disparities. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-17 Law Enforcement Data Review of law enforcement decision points is a critical step in identifying those pathways into the juvenile justice system that most impact DMC (Cox and Bell 2001). As Hoyt and colleagues noted, ―Racial disparities in juvenile detention begin at the arrest stage. In fact, it is these disparities that set the stage for disproportionality at all the other decision points in the court process‖ (Hoyt et al. 2002, 68). Other sites have discovered that variability in police practices (e.g., arrest and transport) resulted in differential treatment of minority youth (Pope, Lovell, and Hsia, 1996). Given these observations, throughout the planning phase, the county’s DMC Committee collaborated with the Greensboro Police Department, the High Point Police Department, and the Guilford County Sheriff’s Department to examine locally relevant decision point information. In particular, the DMC Committee requested that each law enforcement agency provide juvenile (ages 6 to 15) arrest information for calendar year 2004. Table 3a indicates that the Greensboro Police Department recorded 1,853 juvenile arrests in 2004, with African American youth representing 81 percent of all youth arrested. Table 3b shows that nearly 59 percent of all juvenile arrests involved an African American male, and nearly 23 percent of all arrests involved an African American female. The most common charges reported in Greensboro were runaway, larceny, simple assault, and disorderly conduct. Table 3a: Greensboro Police Department Juvenile Arrests, 2004 (N = 1,853) Race (n) Percentage of Total Gender Blacks (1,509) 81.44% 72.10% male 27.90% female Whites (253) 13.65% 52.57% male 47.43% female 3.89% 73.61% male 26.39% female Other (72) Table 3b Greensboro Police Department Juvenile Arrests, 2004 (N = 1,853) Race and Gender (n) Percentage of Total Black males (1,088) 58.72% Black females (421) 22.72% White males (133) 7.18% White females (120) 6.48% Other males (53) 2.86% Other females (19) 1.03% DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-18 A nearly identical pattern emerged in the analysis of the High Point Police Department’s juvenile arrest data for the same time period. Table 4a shows that African Americans accounted for 81 percent of the 742 juvenile arrests recorded. As indicated in table 4b, African American males (57 percent) and African American females (23 percent) were most represented among all arrested juveniles. The most common charges reported in High Point were affray/simple assault, disorderly conduct, larceny, and breaking and entering. Table 4a: High Point Police Department Juvenile Arrests, 2004 (N = 742) Race (n) Percentage of Total Gender Blacks (596) 81.20% 70.97% male 29.03% female Whites (138) 18.80% 61.59% male 38.41% female 1.08% 87.50% male 12.50% female Other (8) Table 4b: High Point Police Department Juvenile Arrests, 2004 (N = 742) Race and Gender (n) Percentage of Total Black males (423) 57.01% Black females (173) 23.32% White males (85) 11.46% White females (53) 7.14% Other males (7) .94% Other females (1) .13% DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-19 Figure 1 below is a map that uses color intensity to illustrate geographic patterns of juvenile offending that the High Point Police Department provided. The areas of deep red/orange on the map denote locations with a high density of juvenile arrests. As the DMC Committee reviewed this map, it became clear that the highest arrest densities were areas clustered around school addresses (e.g., High Point Central High School, Ferndale Middle School, Andrews High School).2 Just as it has been documented at other sites (e.g., Wald and Losen, 2003), this local information has been critical to the DMC Committee’s clearer understanding of the obvious link between behavioral issues at school and entry into the juvenile criminal justice system.3 Figure 1: Density Map of 2004 Juvenile Arrests in High Point DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-20 Juvenile Justice Data The purpose of this component of the team’s data-gathering efforts was to measure the extent of disproportionate minority contact in Guilford County’s juvenile court system. One way to do that is to compare white youth’s contacts in the system with African American youth’s contacts. Such comparisons can be done in various ways. As noted earlier, the measure that OJJDP uses to assess disproportionate minority contact within the juvenile justice system is the RRI. To produce analyses consistent with those conducted at the state level, the team also used this method to analyze minority contacts in the county’s juvenile court system. Using NC-JOIN, an online data system maintained by North Carolina’s Department of Juvenile Justice and Delinquency Prevention, the department’s data analyst provided data by age, race, and gender for key decision points in the county’s juvenile court system for calendar year 2004. Among the data reviewed were total complaints received, complaints approved, complaints not approved, complaints adjudicated, complaints disposed, and complaints dismissed. These data were organized by race and by offense severity. Likewise, the DMC Committee reviewed counts of county admissions to juvenile detention, Juvenile Crime Prevention Council (JCPC) programs, and youth development centers. In 2004, most complaints against youth were not for serious or violent offenses. In fact, as table 5 shows, the five most common complaints were for misdemeanor offenses, which is not surprising considering the most common charges reported by law enforcement agencies. Table 6 provides an example (complaints received) of the type of descriptive information that the team reviewed for each juvenile justice decision point. For each race, the table reports simple frequencies for each category offense (grouped according to severity). For example, 47 complaints were received for Asian youth in 2004: 2 violent, 10 serious, 26 minor, and 9 status. Of note in table 6 is the observation that of 3,013 total complaints received in 2004, 2,196 (73 percent) involved black youth and 633 (21 percent) involved white youth. No other racial category exceeded 2 percent of the total. Table 5: Most Common Complaints, Guilford County Juvenile Court System, 2004 Complaint Frequency Simple Assault 335 Larceny 250 Simple Affray 163 Breaking and Entering 133 Disorderly Conduct by Engaging in Fighting 126 DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-21 Table 6: Complaints Received, by Offense Class, Guilford County Juvenile Court System, 2004 Race A-E (Violent) F-I, A1 (Serious) 1-3 (m) (Minor) Infraction Status Totals 9 47 Asian 2 10 26 Black 46 507 1,507 6 130 2,196 Latino 1 9 32 1 6 49 Multiracial 10 17 5 32 Native American 4 3 2 9 Other 2 22 3 27 10 9 1 20 Unknown White 8 110 460 8 47 633 Totals 57 662 2,076 15 203 3,013 The team calculated incidence rates and RRIs for each decision point in the county’s juvenile court system, comparing rates of occurrence for black youth with those of white youth. First, the team calculated incidence rates for each race by dividing the number of incidents for each race by the total county youth population for that race and multiplying by 100. Next, the team divided black incident rates by white incident rates to calculate an RRI for each decision point. Table 7 (next page) provides an example of incidence rates and relative rates at the point of complaints received. For example, in the minor complaint category, the incidence rate was 1.7 for white youth and 8.6 for black youth. Stated another way, out of all the white youth in Guilford County (27,593), 1.7 percent received a minor complaint in 2004. In contrast, of all the black youth in Guilford County (17,426), 8.6 percent received a minor complaint during the same time period. Also note that for each level of severity, the incidence rate is higher for black youth. Continuing the above example, to calculate the RRI, divide the incidence rate of 8.6 for minor complaints received for black youth by the incidence rate of 1.7 for white youth to arrive at an RRI of 5.058 or approximately 5.1. This means minor complaints were reported against black youth at 5.1 times the rate of white youth; or, for every minor complaint involving a white youth, 5.1 minor complaints involved black youth. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-22 Table 7: RRI Complaints Received, by Offense Class, Guilford County Juvenile Court System, 2004 Number of White Youth Population at risk (ages 10–17) Rate of Occurrence (White Youth) Number of Black Youth Rate of Occurrence (Black Youth) Relative Rate Index 27,593* ---------- 17,426* ---------- Complaints received (minor) 460 1.7 1507 8.6 5.1 Complaints received (serious) 110 .4 507 2.9 7.3 8 .03 46 .3 10.0 Complaints received (violent) *Population numbers were derived from Puzzanchera, C., Finnegan, T. and Kang, W. (2005). “Easy Access to Juvenile Populations.” Online: www.ojjdp.ncjrs.gov/ojstatbb/ezapop/. Focus Groups The DMC Committee was commited to seeking the voices and experiences of several groups directly involved with DMC issues. Therefore, in the spring of 2005, the committee conducted four focus group sessions with caregivers of court-involved youth, court-involved males, court-involved females, and juvenile court counselors. The committee convened the focus groups to explore participants’ personal experiences and seek their insights into ways the juvenile justice system could address the issues around DMC in the county. Each session was audiorecorded and transcribed. Findings and interpretations represent major themes and perspectives of the sessions as summarized by multiple observers and readers. New Directions As the DMC project moves beyond its first year, a set of four interrelated focus areas will guide its work. First, sites across the country concur that data review and decision-point mapping consistently emerge as critical initial steps in DMC reduction efforts because these processes often set the stage for fundamental change (Nellis 2005). ―Data identify how DMC looks and how it operates in a system, without resorting to anecdotes or emotionally charged debates over individual bias‖ (Hoytt et al. 2002, 14). Given those observations, the project’s first focus area is DMC data management and utilization. That is, each quarter, the team will track school, law enforcement, and juvenile justice DMC trends. An emphasis on cultural competence used in DMC decisionmaking is the second focus area. Cultural competency training for key juvenile justice decisionmakers is a promising strategy across multiple sites (Nellis 2005). Misunderstandings about cultural differences and racial stereotyping frequently contribute to differential sentencing decisions for black DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-23 and white youth who have committed similar crimes. Some culturally based expectations of youth that influence judgments and tend to vary across racial and ethnic groups include a ―proper‖ display of respect toward officials and an appropriate expression of remorse for delinquent behavior (Bridges and Steen 1998). Because ―cultural competence is not a fixed characteristic of an agency; rather, it is an ongoing developmental process that agencies and individuals engage in to address diversity in the community-service area,‖ institutionalized cultural competency training teaches agency officials to recognize and minimize the influence of cultural differences on their decisionmaking processes and to empower youth to more successfully negotiate the juvenile justice system (Cox and Bell 2001, 38). When Devine, Coolbaugh, and Jenkins (1998) examined the strategies that five states used to successfully reduce DMC, they found that multiple-approach, rather than singleapproach, strategies were most effective. Specifically, strategies that incorporate family and youth advocacy, coalition building among youth-serving agencies, and targeted resource development appear to lower DMC rates. Similarly, Cox and Bell (2001) maintain that DMC increases when communities lack sufficient and appropriate detention alternatives, fail to identify and address gaps in needed services, or exclude caregivers and family members from decisionmaking processes. Therefore, the third focus area is the promotion and utilization of targeted prevention services. The fourth major focus area is to develop and implement agency policy, procedure, and practice modifications to impact DMC. Small, cost-free changes in policies, procedures, or practices have demonstrated powerful impacts on DMC. For instance, some agencies have reduced DMC by changing detention eligibility criteria so that they detain only the highest risk offenders (Orlando 1999), others developed race-neutral assessment instruments for law enforcement officers to use in the field to guide their decisions to detain or release youth (Rust 1999), and still others modified their operating hours to accommodate the schedules of working parents (Cox and Bell 2001). Opportunities and Challenges Guilford County embarked on this DMC work 15 years ago, during the early 1990s. What came of that effort was little more than a plan that was promptly shelved and received no further attention. This time, the county has committed itself to making its DMC efforts goal driven and action oriented. It is significant that the county invited the university to take a lead role in the first place. This is a result of several years of working with the community service providers and establishing relationships. In many communities, service providers and/or the system itself would be host for this kind of project, with university participation being ancillary. In this instance, the grant was awarded to the university, not to a community-based organization or collaborative. Unlike the last time the county undertook DMC activities, the CYFCP has served as a centralized, objective, convening organization that has taken a clear managerial lead (grant management, logistics, research, agency coordination) in the project. Having an engaged, university-based center involved in project management was useful for several other aspects of the project. For example, the project’s commitment to impartially and objectively receiving and sharing data from schools, law enforcement, and juvenile DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-24 justice contributed greatly to project success. Likewise, the center’s faculty and graduate student expertise was brought to bear on the collection and analysis of quantitative and qualitative data and immediate dissemination through reports, fact sheets, handouts, and presentations. Appendix Endnotes 1. As part of the 11-county Piedmont Triad region (population: 1.27 million) of North Carolina, Guilford County is centered along the Piedmont Industrial Crescent stretching from Raleigh to Charlotte. Guilford County has the third-highest population in the state at 421,000. 2. Another area with a high arrest density was observed at the Oak Hollow Mall. 3. Although not reproduced here, the Greensboro Police Department provided a similar geographic map that showed patterns of juvenile offending clustered near schools. Appendix References Bridges, G.S., and Steen, S. 1998. ―Racial disparities in official assessments of juvenile offenders: Attributional stereotypes as mediating mechanisms.‖ American Sociological Review 63(4):554–570. Cox, J.A., and Bell, J. 2001. ―Addressing disproportionate representation of youth of color in the juvenile justice system.‖ Journal of the Center for Families, Children, and the Courts 3:31–43. Devine, P., Coolbaugh, K., and Jenkins, S. 1998. Disproportionate Minority Confinement: Lessons Learned From Five States. Bulletin. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Forsbrey, A.D., Frabutt, J.M., and Smith, H.L. 2005. ―Social isolation among caregivers of court-involved youth: A qualitative investigation.‖ Journal of Addictions and Offender Counseling 25:97–113. Frabutt, J.M., Cabaniss, E., Arbuckle, M.B., and Kendrick, M.H. 2005. Reducing disproportionate minority contact in the juvenile justice system: Promising practices. Paper presented at the 2005 Society for Prevention Research Annual Conference, Washington, DC. Frabutt, J.M., Kendrick, M.H., Arbuckle, M.B., and Cabaniss, E. 2005. Reducing disproportionate minority contact in the juvenile justice system, Guilford County demonstration project: Final report and Guilford County DMC reduction plan. Greensboro, NC: Center for Youth, Family, and Community Partnerships, University of North Carolina at Greensboro. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-25 Hoytt, E.H., Schiraldi, V., Smith, B.V., and Ziedenberg, J. 2002. ―Pathways to juvenile detention reform: Reducing racial disparities in juvenile detention.‖ Pathways to Juvenile Detention Reform Series, 8. Baltimore, MD: Annie E. Casey Foundation. MacKinnon-Lewis, C., and Frabutt, J.M. 2001. ―A bridge to healthier families and children: The collaborative process of a university-community partnership.‖ Journal of Higher Education Outreach and Engagement 6(3):65–76. Mendel, D. 2003. ―And the walls keep tumbling down: A demonstration project has come and gone, but detention reform continues to gather steam.‖ Advocasey 5(1):18–27. Nellis, A.M. 2005. Seven Steps To Develop and Evaluate Strategies To Reduce Disproportionate Minority Contact (DMC). Juvenile Justice Evaluation Center Guidebook Series. Washington, DC: Juvenile Justice Evaluation Center. Orlando, F. 1999. ―Controlling the front gates: Effective admissions policies and practices.‖ Pathways to Juvenile Detention Reform Series, 3. Baltimore, MD: Annie E. Casey Foundation. Pope, C.E., Lovell, R., and Hsia, H.M. 2002. Disproportionate Minority Confinement: A Review of the Research Literature from 1989 through 2001. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Rust, B. 1999. ―Juvenile jailhouse rocked: Reforming detention in Chicago, Portland, and Sacramento.‖ Advocasey 1(3):1–16. Shelton, T.L., Frabutt, J.M., and Arbuckle, M.B. 2003. University-community engagement to support youth, family, and community development. Paper presented at Outreach Scholarship Conference 2003: Excellence through Engagement, Madison, WI. Wald, J., and Losen, D. 2003. ―Defining and redirecting a school-to-prison pipeline.‖ In Deconstructing the School-to-Prison Pipeline: New Directions for Youth Development, No. 99..San Francisco, CA: Jossey-Bass. DMC Technical Assistance Manual, 4th Edition • Chapter 6: Federal, State, and Local Partnerships 6-26 Chapter 7: Strategies for Serving Hispanic Youth1 Maria F. Ramiu and Dana Shoenberg* Introduction Empirical research has demonstrated that youth of color are overrepresented throughout the justice system (Leonard, Pope, and Feyerherm, 1995). The experience of youth of color in the juvenile justice system often differs from that of their white counterparts when objective criteria of offense and offense history are held constant (Poe-Yamagata, and Jones, 2000; Villarruel et al., 2002). Differential treatment of youth of color sometimes manifests as a higher likelihood of incarceration or increased length of incarceration (Poe-Yamagata and Jones, 2000). African American, American Indian, and Hispanic youth are the three populations that are most often identified as having disproportionate contact with the juvenile justice system (Hsia, Bridges, and McHale, 2003). This report focuses on DMC issues related to Hispanic youth. In most jurisdictions, current approaches to collecting and accessing data are inadequate to measure overrepresentation and disparate treatment of Hispanic youth in the juvenile justice system (Villarruel et al., 2002) and have not consistently tracked the ethnicities of youth of color in their care. Persons of Hispanic or Latino ethnicity can be of any race; some also may choose to identify their race as Hispanic or Latino/a. Systems that recognize only the races but not the ethnicities of youth and their families lack essential information about the number of Hispanic youth they serve, thus creating an ―invisible minority‖ in juvenile justice. For example, when systems ask youth to check on a form whether they are black, white, Hispanic, Asian, or ―other,‖ they may force youth to choose between identifying their race or their ethnicity. This approach leads to undercounting of Hispanic youth who report their race rather than their ethnicity when forced to choose (Villarruel et al., 2002; Holman, 2001). Where Hispanic youth are undercounted because they are reported as white, the true extent of Hispanic overrepresentation cannot accurately be measured (Villarruel et al., 2002). Furthermore, if the data collection ends with inquiries only about race and ethnicity, juvenile justice decisionmakers have incomplete information about the youth and families they serve. Information regarding the youth’s primary language, English-language proficiency, language spoken in the home, parents’ English-language proficiency, national origin, household composition, and other family characteristics may help support provision of culturally and linguistically competent interventions. * About the authors: Maria F. Ramiu is Staff Attorney, Youth Law Center, San Francisco, California. Dana Shoenberg is Senior Staff Attorney, Center for Children’s Law and Policy, Washington, DC. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-1 Cultural and linguistic competence is essential to create a fair system with meaningful rehabilitative interventions. For example, parental involvement may be hampered when language or cultural communication gaps exist between juvenile justice professionals and the youth’s family. Asking youth to translate conversations between their parents and professionals in the juvenile justice system is a recipe for inaccurate translation and misunderstanding, since youth (a) may not be sufficiently proficient in English and therefore may not completely understand the meanings of words these professionals are using, and (b) may be motivated to translate incorrectly. Yet systems with insufficient information about their bilingual staffing needs sometimes lack sufficient culturally and linguistically competent staff and inappropriately rely on youth to translate. In addition, families who come to the United States from countries with autocratic or corrupt legal and law enforcement systems may not understand or trust the juvenile justice system. Culturally and linguistically competent staff are essential to help families access appropriate services and support their children’s successful completion of rehabilitation programs. Adequate race, ethnicity, and language proficiency data are important for assessing racial and ethnic disparities and developing, implementing, and monitoring appropriate interventions. Project Overview This report summarizes lessons learned from a 2-year cooperative agreement award from the Office of Juvenile Justice and Delinquency Prevention (OJJDP) to the Youth Law Center. This project was originally funded from September 1, 2004, to August 31, 2006, and extended to August 31, 2007. The purpose was: (1) to develop new and accurate data collection methods for Hispanic youth in the juvenile justice system to accurately assess and identify appropriate strategies to reduce Hispanic DMC, and (2) to reduce DMC for Hispanic and other youth at critical decision points in the system. Because most decisions that determine whether individual youth enter and penetrate the juvenile justice system are made at the local level, project staff chose to work at the county level, engaging court personnel, community members, advocates and other stakeholders involved in their juvenile justice systems. The project was structured using a site-based approach through which Hispanic data collection issues and DMC interventions could be tested. The project was designed to bring together key juvenile justice system stakeholders in each county site to collaboratively develop and implement a data informed plan to improve data collection and create interventions to better address the needs of Hispanic youth in each jurisdiction. The site selection process focused on identifying significant or growing Hispanic populations in jurisdictions that did not collect and/or disaggregate accurate data on Hispanic youth at one or more major decision points in the justice system. The juvenile justice system personnel and community leaders of those jurisdictions needed to be committed to improving data collection and reducing overrepresentation and racial disparities affecting youth of color. The objectives of the site work were to develop new data collection methods, use the data collected to identify gaps in services and resources, and develop services and resources to fill them. Two jurisdictions, Washoe County, Nevada, and Travis County, Texas, participated as project sites for the development of comprehensive data collection methods for Hispanic DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-2 youth in the justice system and the reduction of DMC at critical decision points indicated by the data on disproportionality. In Washoe County, the DMC reduction work focused on confinement in secure correctional facilities which function as the State’s training schools. In Travis County, work focused on detention of youth for probation violations. The work at each site is described below. Washoe County, Nevada Washoe County, Nevada, has a significant and growing Hispanic population. Nevada’s Hispanic population tripled from 1990 to 2000. The State has the fifth largest percentage of Hispanics in the United States. In the Nevada public schools, 27 percent of all students in grades 6–12 are children of immigrants. Washoe County is Nevada’s second largest metropolitan area, anchored by the city of Reno. Nevada’s rapidly growing Hispanic population is reflected in the juvenile justice system. Since 2000, Hispanic youth in the Washoe County juvenile justice system have increased from less than one-quarter to more than one-third of the juvenile detention and secure confinement populations. In addition to its significant and growing Hispanic population, Washoe County’s data indicated racial disparities or minority overrepresentation in several or all contact points in the juvenile justice system. Nevada’s DMC data reporting system uses the mixed racial/ethnicity ―Hispanic‖ category when collecting data on Hispanic youth. However, 2003 data for Washoe County revealed low Relative Rate Indices (RRIs) of overrepresentation of Hispanic youth at most stages of the juvenile justice system, except at the disposition to secure confinement stage, where the RRI for Hispanic youth was 5.08.2 The Hispanic population and DMC data issues made Washoe County an attractive candidate as a potential site. Washoe County also scored high on the level of commitment to the project’s DMC goals. Over the last few years, Washoe County has undertaken several juvenile court reform efforts, including initiatives on detention, mental health, education, and model courts and has committed to considering racial disparities at every stage of the juvenile justice system. Thus, addressing DMC issues, particularly closely related to the growing Hispanic population, was compatible with the County’s ongoing reform efforts. The stakeholders were particularly interested in the potential of the project to focus on DMC at the secure disposition stage. The focus on the later processes of the system would supplement other reform work focusing on the earlier processes of the system. The initial site work began with the project staff meeting with key stakeholders, led by the presiding juvenile court judge, and the director of juvenile justice services. The county formed a small advisory group to guide the project and developed a project workplan consisting of three phases: assessment and research, recommendations development, and implementation. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-3 Data Collection During the project’s initial assessment process, stakeholders identified data collection issues. They felt that it was important to look beyond categorizing youth by race and ethnicity, as Washoe County used the ―mixed race/ethnicity‖ category for Hispanic youth in accordance with State DMC data reporting requirements. The option of using two questions on race/ethnicity on group surveys raised the issue of whether such a change across data collection systems would increase the accuracy of data on Hispanic youth. The State-legislated DMC reporting system uses a mixed racial/ethnic category for ―Hispanic;‖ a change in race/ethnicity categories would require devoting resources at the State level but would result in little gain for the local DMC work. In Nevada, the legislature is part-time and only meets biannually. An effort to change State law would have diverted resources from and delayed the progress of the local work. Other questions arose regarding the inclusion of other ethnic categories and the utility of having detailed census-level racial and ethnic data. The costs of adding a second question for race/ethnicity outweighed the benefits. Stakeholders decided to proceed with the categories that best reflect system perceptions and self-identification for youth and families. They chose to place the focus on improved data collection across racial/ethnic categories that would provide more useful information to further the DMC work. Washoe County’s available data included automated information capable of producing aggregate data as well as case record information contained in individual files. The automated data included much of the demographic and offense data, but other information regarding risk and protective factors of family, language, culture, and service interventions was contained only in individual case files. Thus, the county chose a case review methodology as the primary component of the project’s assessment phase. A case review was developed, conducted, and analyzed through a joint effort among the Washoe County Department of Juvenile Services staff, a project research consultant, and a professor at the University of Nevada, Reno. The case review was designed to learn more about the characteristics of youth committed to secure confinement and the effect of juvenile justice system interventions before commitment to secure confinement. The case review focused on youth who received suspended commitment dispositions in calendar years 2004 and 2005. The original disposition of almost every youth from Washoe County confined in the State training schools was a suspended commitment to the State training school that was subsequently revoked. The case review consisted of 110 cases of youth receiving suspended commitment dispositions in calendar year 2004, comparing differences between youth who were confined within 2 years (55 cases) and those who were not (55 cases). The research team conducted an additional review of 24 cases of youth who received a suspended commitment in calendar year 2005 and were confined in the State training school within 1 year of the suspended commitment disposition. The team compared the 2005 cases with a sample of cases of confined youth from the suspended commitment dispositions reviewed from the calendar year 2004 group. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-4 The information collected from the case review included the following: Youth demographics (age, race/ethnicity, gender, native-/foreign-born, and nonEnglish language). Referral and offense history (number, type, and age at first referral). Commitment offense. Youth health, education, and employment status (disabilities, mental health and substance abuse histories, school enrollment, attendance, and special education status); social relationships (gang, delinquency involvement, and parenting status); and abuse/neglect history. Parent demographics/characteristics (native-/foreign-born status, English proficiency, residence, incarceration history, and socioeconomic status). Household composition. Court-ordered interventions. The case review examined the characteristics of and differences between youth with suspended commitment dispositions who did and did not progress to commitment in the State training school to learn more about the characteristics of the youth who were committed to secure confinement. The analysis of the case review provided a number of findings to support the recommendations and interventions in the project’s second and third phases. Findings The significant findings regarding youth committed to the State training school include the following: Committed youth were significantly more likely to have two parents who did not speak English (58 percent of committed youth compared to 44 percent of noncommitted youth). Committed youth were significantly more likely to have fathers who were incarcerated (22 percent of committed youth and none of the noncommitted youth). Committed youth also were significantly more likely to have both mental health and substance abuse issues (71 percent of committed youth compared to 56 percent of noncommitted youth). Only 4 percent of committed youth were employed, compared to 17 percent of noncommitted youth. There was no statistically significant difference in employment DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-5 history (i.e., having a job) between committed and noncommitted youth. No African American youth in the sample had employment histories or were currently employed. Committed youth in the suspended commitment disposition were more likely than noncommitted youth to have had orders for individual, family, and substance abuse counseling; mandatory school; and search and seizure for drugs, drug paraphernalia, alcohol, and graffiti. Committed youth were less likely than noncommitted youth to have orders for drug court or a diversion program in the suspended commitment disposition. Virtually all parents (100 percent of mothers and 93 percent of fathers) of youth in the sample resided in the United States. The case review also revealed certain significant findings with respect to Hispanic youth in particular: Hispanic youth were more likely to not be attending school at the time of the last arrest preceding commitment. Hispanic youth whose commitment was initially suspended were more likely than white youth to receive an order for mandatory school attendance. Hispanic youth whose commitment was initially suspended were less likely than white youth to be ordered to undertake family or individual counseling. Hispanic youth were more likely (25 percent) to have repeated one or more grades. Hispanic youth (77 percent) were more likely than white youth (43 percent), African American (33 percent), or Asian youth (33 percent) to be involved with delinquency and/or gangs. Hispanic youth were somewhat less likely (66 percent) to have a documented mental health problem than white youth (80 percent). A higher percentage of Hispanic youth were born outside the United States (29 percent compared to 5 percent of white youth and no African American or Asian youth). Hispanic youth were somewhat more likely to have a translator than were white, African American, or Asian youth. Hispanic youth (12.60) and African American and Asian youth (12.44) had a lower average number of referrals than white youth (13.08). DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-6 Hispanic youth (34 percent) were more likely to live with both parents than were white youth (23 percent) or African American youth (0 percent). Hispanic youth (9 percent) were more likely to be parents, compared to 3 percent of white youth and 0 percent of African American and Asian youth. The case review process revealed a number of issues regarding data collection: Data on English-language proficiency of youth are not collected. Youth who need translation services in court are tracked in the automated system. Few youth received translation services in court (6 percent of committed youth and 4 percent of noncommitted youth), yet the majority of Hispanic youth had non-English-speaking parents. Translators were only used for youth who did not appear to speak any English. The use of court translators by youth is a very limited proxy for the youth’s ability to communicate in English in written and spoken form. School district data indicated that almost 20 percent of students are identified as LEP (limited English proficient) students.3 Data on parent language, immigrant status, family structure, youth health, education, and employment status are not put in the automated data system and thus no aggregate data are available on these elements through juvenile justice services. Complete data on youth participation, compliance, and completion rates with ordered services and interventions is not collected. Recommendations Data Collection The inability of the case review process to reach certain findings regarding intervention intensity, language proficiency, and intervention outcomes—as well as specific findings regarding education status, gang involvement, family demographics, and parental status—suggested the need for improvements in data collection methods. The following recommendations were made regarding data collection: Data on limited English proficient youth and their parents should be better assessed, and language proficiency data should be collected in the automated data system. The automated data system should include variables that indicate whether a youth participated, completed, withdrew from, or failed to complete an intervention and at what point in time the youth discontinued participation. Data collection methods need to be modified and improved, and certain existing data contained in the individual case files need to be incorporated into the automated data system, specifically, those data elements contained in the case review instrument that are not in the system: the youth’s language, school status, parents’ languages, household composition, and interventions ordered and completed. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-7 Language Access Hispanic youth were more likely than other youth to have parents who did not speak English and to report that both parents did not speak English. These findings highlight the importance of language-accessible and culturally appropriate services for Hispanic youth and their families that maximize parents’ understanding of the legal process, to support their participation as fully as possible in planning for the youth, and provide families with the services they need. The findings regarding the immigrant and linguistic status of youth and their parents suggested the following needs: Improvement in the assessment of language access needs of youth and their families, particularly the Hispanic population. Improvement in access to services that meet linguistic needs of youth and their families. Education The findings indicated that Hispanic youth might be at an increased risk of poor educational performance. Compared to white, African American, and Asian youth, Hispanic youth were less likely to be enrolled and attending school. Compared to their counterparts, they were also more likely to be enrolled but not attend school or not to be enrolled in any school. In addition, Hispanic youth were more likely to have repeated one or more grades than were white, African American, or Asian youth. One of the most frequently court-ordered interventions for Hispanic youth was mandatory school attendance. These findings suggest that Hispanic youth may be particularly at risk educationally. Project staff recommended development of interventions that would focus on the following: Addressing barriers to educational success for Hispanic youth including language proficiency issues, availability of educational advocacy assistance, and lack of academic supports. Improving relationships of youth and their parents with the school and probation systems through providing bilingual probation staffing, services, and information to ensure that youth are enrolled in and attend school. Mental Health and Substance Abuse Status Committed youth were significantly more likely to have both mental health and substance abuse issues. Additionally, the findings indicated racial/ethnic disparities in mental health status and in ordered interventions that implicated behavioral health issues. Accordingly, the recommendations included the following: Further assessment of the racial/ethnic differences with behavioral health issues and the delivery and effectiveness of accessible, culturally appropriate services. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-8 Coordination of current mental health and substance abuse treatment services for those youth who are dually diagnosed. Delinquency and Gang Involvement Hispanic youth were more likely than white, African American, or Asian youth to be involved with delinquency and gangs. This finding suggested the need for ongoing development and implementation of effective interventions regarding negative social relationships. Parenting Status Although few youth had children, Hispanic youth were more likely to be parents. The parenting status finding (the small percentage notwithstanding) suggested the need for interventions for Hispanic youth with parenting issues. Interventions The intervention implementation phase of the project focused on language access, education, and the additional data collection recommendations as described below. Language Access Washoe County improved data collection methods for Hispanic youth, specifically regarding language proficiency, which is now assessed and included as a component in the collection system. A language proficiency survey (shown in appendix A of this chapter) was initially developed for Washoe County as a preliminary assessment of language proficiency by the education transition coordinator (described in the education interventions below). The survey instrument information was incorporated into the service assessment process at intake. Education Washoe County has established an education transition program for youth exiting the detention center. The county designed the program to ensure the seamless transfer of youth and their educational records from detention to community schools, other detention centers, and community and employment agencies. The ultimate goal of the program was to reduce the number of students who recidivate and are sentenced to State juvenile and adult corrections facilities, and to increase the number of students who are successfully and continuously engaged in school, work, and community activities. The components of the program include the following: Establishing a transition coordinator position. Developing individualized transition plans. Developing and implementing a student education passport. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-9 Establishing a seamless transfer of educational records and services. Increasing interagency cooperation and communication. Establishing a youth tracking system. Ensuring that the schools identify, assess, and serve limited English proficient (LEP) youth. To assist youth with the transition to community programs, other education interventions were implemented in the detention center: Identifying LEP students. Instituting English language development classes for LEP students. Providing computer programming modules in the school computer lab. Training all detention center school teachers in ―sheltered instruction,‖ an instructional strategy aimed at the language needs of LEP students. Out-of-custody youth who have not been participants still benefit from the resources developed for the transition program and indirectly benefit from their improved relationships with the schools and community programs. The juvenile services case workers are able to access the resources and improved relationships to help reconnect youth, particularly Hispanic youth, with schools, community programs, and other support systems. The transition program and other education interventions have not been in operation long enough to assess their effectiveness or impact on disproportionate minority contact with the juvenile justice system. However, the county has reached several preliminary conclusions that Hispanic youth are receiving improved services. The collaboration of the probation department with the school district through the detention school and with its administrative offices has improved educational data collection. By establishing new information-sharing protocols, including allowing the detention school to access the school district’s database, services to Hispanic youth were improved, including access to LEP services in detention and community schools, assistance in school enrollment, and connecting with community services that met their language and cultural needs. Additional Data Collection Improvement of data collection on interventions was imperative in tracking the impact of the work of the project and evaluating all juvenile agency services. Data on interventions, participation, and completion rates are now collected in a standard format in the youth’s file. Although the automated system contains basic intervention information, the County DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-10 continues to modify the system so it can produce comprehensive data reports that facilitate a thorough evaluation of intervention effectiveness. Continuing Intervention Development In addition to the continued work on the interventions already described, the Washoe County site used this work as the basis for ongoing development and implementation to address Hispanic disproportionate minority contact with the juvenile justice system. A symposium on Hispanic youth was held to foster community partnerships to provide new services to address the education, employment, behavioral health, and social services needs identified in the project. Washoe County has also established two gender-specific Evening Reporting Programs as well as a pilot mentoring program for Hispanic boys. The County has modified case management practices as a result of the case review findings. Case managers have a better understanding of some of the differences in the educational, employment, family, behavioral health status and other factors associated with the path to confinement of Hispanic youth in secure facilities, so they can now focus services in those areas. Travis County, Texas Travis County, Texas, which includes Austin, the State capital, has a large and established Hispanic population. Of the population of youth ages 10 to 16 (those eligible for the juvenile justice system), 39 percent are Hispanic. Twelve percent of the youth are African American, 45 percent are white, and 4 percent are Asian. A medium-sized county with a significant Hispanic population and a receptive climate for conducting careful analysis, Travis County was an ideal location to address the disproportionate minority contact of Hispanic youth. Through the leadership of the Travis County Juvenile Probation Department (TCJPD), probation staff and representatives of the judiciary, school police, public defenders, district attorneys, community organizations, and service providers all participated in Travis County’s examination of racial and ethnic disparities in the juvenile justice system. These stakeholders shared data, recommended key contacts to be interviewed or included in the project, and participated in fact-gathering interviews. TCJPD staff devoted extensive effort to the project, including data analysis, organizing files for review, arranging and participating in interviews, and planning an intervention strategy. The RRI analysis revealed that youth of color in Travis County were most overrepresented at the point of referral into the juvenile justice system.4 Hispanic youth were 3.85 times more likely than white youth to be referred to juvenile court. The analysis also revealed that once Hispanic youth were involved in the juvenile justice system, they generally were not disproportionately represented in later stages, with one notable exception: The rate at which Hispanic youth were securely detained for probation violations was 1.42 times greater than the rate of detentions of white youth. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-11 As is true of urban poverty, rates of referral to the juvenile justice system are probably affected by a complex set of factors. The struggle for decisionmakers in this project, which was funded for only 2 years, was whether to try to address the multiple (often systemic) contributing factors to juvenile court referrals or to choose a more limited project that did not require the engagement of systems outside the juvenile court. Decisionmakers determined that they were more likely to complete an analysis and implement interventions if the project was focused on disproportionality of detentions for probation violations rather than on referrals to juvenile court. A later phase of DMC reduction work could address disproportionality of referrals to juvenile courts if enough agencies and other stakeholders were interested in coming together to examine the causes and find solutions. Thus, the project focused on examining the differential rates of detention of youth of color for technical violations of probation and on designing appropriate interventions. Data Collection The juvenile probation system in Travis County uses a database developed at the State level for probation departments. TCJPD maintains computer records of juveniles and their families by using the Caseworker system provided by the Texas Juvenile Probation Commission (TJPC). This system is designed for county-level recordkeeping and meeting mandatory State reporting requirements. The race/ethnicity field in the Caseworker system is structured to report race/ethnicity in a manner that conforms to other Texas law enforcement entities, including the Texas Department of Public Safety. It combines two aspects of an individual—race and ethnicity—into a single category; a user can designate the individual as African American, American Indian, Asian-American, Hispanic, white, other, and unknown. This method is required for reporting to the State government, and TCJPD is required to use this field to report its data to TJPC. To be consistent with its other reports to TJPC, TCJPD uses the race/ethnicity field in Caseworker for all data reporting. TJPC’s approach also provides consistency when transferring information between TCJPD and other law enforcement entities, including information concerning transfers of juveniles from one jurisdiction to another. TCJPD welcomed the value of capturing race and ethnicity as separate concepts and, as part of this project, chose to use two user-defined fields in Caseworker to more fully capture both race and ethnicity information for use at the county level. While the county must still report its mixed race and ethnicity data to TJPC, the user-defined fields allow the county to capture more detail for its own use. Now, workers ask the current method is to ask each juvenile for a primary and a secondary racial or ethnic identification, each of which is recorded in the primary and secondary fields. Youth may self-identify as two of the following: African American or black; American Indian; Anglo/white; Arab; Asian; Hispanic; Latino; Middle Eastern; Native Hawaiian or other Pacific Islander; Semiticspeaking of Near East, North Africa, or Arab; other; and unknown. As part of this new approach, TCJPD also identifies the juvenile’s primary language. Youth may report DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-12 primary language as American Sign Language, Braille, Chinese/Mandarin/Cantonese, English, French, Japanese, Korean, Spanish, Vietnamese, or other. Although it does not report aggregated data using these fields, TCJPD uses this information extensively for its culturally sensitive work with individual juveniles and their families. County officials have expressed that, ideally, the TJPC system would separate race and ethnicity for reporting purposes, allowing youth to self-identify with as many race and ethnicity categories as they wished. This would require a decision from TJPC and a transition period. Data in the current system would have to be converted, and informational gaps would occur. For example, for those who did not identify themselves as ―Hispanic‖ under the current system, the county would not have ethnicity information, only race information. Furthermore, unless the Texas Department of Public Safety and other law enforcement entities in Texas also transitioned their data systems at the same time, both TJPC and TCJPD would experience data compatibility issues when sharing data across jurisdictions and among other law enforcement entities. These challenges highlight the potential benefits of working at the State level to modify race, ethnicity, and language information collection across systems. Research, Analysis, and Findings on Violation of Probation To learn more about youth who violated probation, and why youth of color violated probation and were detained at a higher rate than whites, project participants engaged in both qualitative and quantitative analysis. For qualitative analysis, project staff interviewed TCJPD line workers and managers, judges, public defenders, district attorneys, community-based organizations, and service providers. Staff conducted focus groups with youth in the Intermediate Sanctions Center and with their parents. Researchers conducted separate sessions in English and in Spanish and met with both boys and girls. Structured interviews with stakeholders from the various systems help paint a picture of people’s experiences and insights and allow the opportunity for suggestions. In addition, two quantitative studies described youth who were detained for probation violations in Travis County. TCJPD conducted an analysis of 901 youth who completed probation in 2005 to understand the characteristics, risk levels, and needs of probation violators. As part of this study, researchers examined 100 case files to determine the types of probation violations that youth committed. In addition, project staff, with the assistance of local graduate students, reviewed files of 87 youth detained in 2005 (representing 150 detentions) to learn about their offending behavior, language capability, length of stay in detention, and other factors. Following are some of the key findings from the case file reviews:5 The top three probation violations were truancy (27 percent), curfew violations (21 percent), and substance use (15 percent). On average, youth committed three different types of violations. Three other categories figured prominently: missed meetings with probation officers, failure to participate in a program, and school suspensions. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-13 Juveniles under supervision for substance abuse had the highest detention rate for probation violations. Compared with supervised youth not detained for probation violations, those detained for probation violations had notably higher rates of substance abuse. Youth scored on the risk assessment instrument mostly in the low and medium ranges (45 percent in low range; 53 percent in medium range), with only 2 percent deemed at high risk of recidivating. These factors suggested that any program seeking to reduce detentions of youth who commit probation violations would need to address substance use, truancy, curfew violations, and other individual risk factors. Youth detained for probation violations were almost entirely in the low- to medium-risk categories (as measured by the Adoms III risk assessment tool that TCJPD used; see appendix B) suggested that a large proportion of them might be good candidates for a different form of intervention that did not involve secure custody. Researchers6 examined the underlying offenses for which youth were on probation. As would be expected for youth being supervised in the community, these youth had not committed dangerous violations that threatened public safety. Figure 1 shows the breakdown for underlying offenses leading to probation. Figure 1: Underlying Offenses Committed by Youth Who Violated Probation Criminal Mischief 5.1% Other 18.8% Marijuana Possession 24.1% Criminal Trespass 5.1% Evading/Resisting Arrest 8.9% Burglary 8.9% Theft < $500 15.2% Assault 13.9% The following two charts indicate that Hispanic youth spent more days, on average, in detention for probation violations than white youth, but they were also detained more times. Figure 2 illustrates that Hispanic youth detained for probation violations served an average of 34 days total in detention during the sample period, whereas white youth served an average of 20 days. This may be explained by Figure 3, which illustrates that Hispanic youth detained for probation violations were detained an average of 2.0 times, whereas the average for white youth was 1.4 times. These findings suggested that any DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-14 project that provided an alternative to detention for probation violators would significantly reduce the number of youth of color detained for probation violations. Figure 2: Mean Number of Days in Detention: Youth Who Had Violated Probation (by Race/Ethnicity) 40 30 Number of Days in Detention 34 20 20 10 0 Hispanic Caucasian Figure 3: Average Number of Times in Detention: Youth Who Had Violated Probation (by Race/Ethnicity) 3 Number of Times in Detention 2 2 1.4 1 0 Hispanic White Researchers also examined the language capability of youth detained for probation violations and their families. Nine percent of the sample of youth did not speak English as their primary language, and for at least 17 percent of the youth, the family language was not English. These data illustrate the importance of gathering information beyond the DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-15 youth’s primary language, as communication with family members may require specialized resources, such as bilingual staff, even if youth communicate well in English. Another significant finding from the qualitative research was the common theme raised by a number of stakeholders: Parents and youth who are new to the juvenile justice system, and especially those new to the United States in general, need help understanding the system and its expectations. Families from countries with corrupt or abusive criminal justice systems approach the experience with distrust, fear, and lack of knowledge about how to help their children succeed during their probation. Several stakeholders thought it would be helpful to offer a program of information, outreach, and support to families (possibly staffed with family members of formerly adjudicated youth), which could help bridge the cultural and information gaps experienced, especially by new immigrant families. Because researchers were required to conduct individual case file reviews to better understand the bases for probation violations, the county recognized the value of electronically recording the probation violation types and their frequencies. For youth involved in the new Sanction Supervision Program described below, this information is being tracked. In addition, for some probation violators, the county conducts administrative hearings in which a juvenile probation supervisor meets with the probation officer and family to discuss violation problems and to arrange new probation terms. This is another method to prevent secure detention, but it had not been tracked electronically, nor had written records been consistently kept in youths’ files. The county is now able to track the use of administrative hearings electronically. Additionally, even though the county had translated almost all of its juvenile court and probation-related documents into Spanish, project staff identified a few additional documents that needed translation, which the county completed. Although many jurisdictions have been unable to complete accurate translations for all documents available to youth and their families, Travis County staff have accomplished this important aspect of language accessibility. Strategic Intervention On the basis of many interviews and the quantitative research, researchers recommended potential interventions to address the disproportionate representation of youth of color in probation violation detentions. TCJPD chose to focus its resources on the creation of additional intensive supervision positions to be used specifically as a strategy to reduce detention of probation violators. Because the department’s research indicated that an average of 40 to 50 percent of the county’s detention beds were assigned to juveniles committing technical probation violations, creating alternatives to detention was a high priority. The Sanction Supervision Program (SSP; see Figure 4) provides for additional case management to ensure that youth and their families receive help with identifying and accessing services they need. The choice to have a community-based agency offer more intensive case management substantiates research findings. When youth who had previously violated probation and DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-16 their parents were asked what would have made a difference, both the youth and their parents expressed the need for more attention from their probation officers. The prevalence of youth violations involving substance abuse, truancy, and school suspension in the studies reflected a need for more intensive service provision. The organization chosen to provide case management for the program has a long history of culturally competent service provision to youth involved in the juvenile justice system in Travis County. Due to these responses, TCJPD received funding for four additional intensive supervision staff positions, which resulted in a reduced caseload of 12–15 youth for each intensive supervision staff member, compared with regular probation caseloads of 26–28 youth when the agency is fully staffed. The program can serve 60 youth and is expected to serve approximately 150 youth annually; youth are supervised at this level for approximately 4 months. At its inception, this program was expected to reduce the average daily population at Travis County’s Gardner-Betts Juvenile Justice Center by nine youth per day.7 Along with other initiatives the county has implemented, the SSP has contributed to a drop in the average daily population at the detention center from 93 during the first half of fiscal year 2007 to 83 during the first half of fiscal year 2008. These accomplishments have significantly eased population concerns for the detention center.8 The county has been careful to ensure the program is used only for those youth who would otherwise be referred to detention for their violations so as not to ―widen the net‖ of youth receiving intensive services beyond those for whom the program is intended. Youth with felony charges represent 64 percent of the program participants, and youth with higher level misdemeanors (A and B) represent 34 percent of program participants. As a group, the youth have had an average of 6.8 referrals to the juvenile justice system before their participation in the program.9 Reflecting efforts to serve populations overrepresented in detentions for probation violations, 61 percent of youth in the program are Hispanic, 8.5 percent are white, and 29 percent are African American; Asian and American Indian youth each constitutes 0.7 percent. Program participants range in age from 12 to 17 years, and males represent 73 percent of the group. Forty-seven percent of participants have completed the program successfully.10 Creating this alternative to detention that mainly serves youth of color is a concrete and measurable achievement for Travis County to address both DMC and overall rates of detention (see Figure 4). DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-17 Figure 4 SANCTION SUPERVISION PROGRAM Adjudicated Chronic probation violations (curfew, truancy, and positive drug screens) JPO addresses chronic probation violations through interventions (home detention, office visits, and discussions with the Sr. JPO and/or Casework Manager Sanction Supervision Referral Form is completed Administrative Review Hearing conducted Casework Manager Utilizes Alternative Plan Recommendation for Sanction Supervision is submitted to the Division Manager Case is denied into SSP and remains in the Regular Field Probation Unit and alternative plan included in the Administrative Review Hearing is utilized. Judge signs amended probation rules DTA or MTM Juvenile is transferred to Sanction Supervision Juvenile completes probation and is terminated from probation Juvenile is transferred back to the Field Probation Unit Note: JPO = juvenile probation officer. DTA = directive to apprehend; a warrant for law enforcement to pick up the youth. MTM = motion to modify; asking the court to modify the original disposition (such as extending the length of probation). DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-18 Recommendations for Additional Activity As Travis County’s Sanction Supervision Program unfolds, staff will track the participants to learn whether they re-offend. It may be helpful to follow whether the rate of successful completion of the program and recidivism following the program are different for any particular youth profile. Are recent immigrant youth succeeding at the same rate as others? Does home language make a difference in families’ experience of the program? Are services that clients need actually available in the community, or are there resource gaps that need to be filled? Are all of the services to which youth are referred effective and culturally competent? Some information could only be found through time-intensive case file review, which illustrated the potential usefulness of tracking additional data elements about the county’s clients. The county has already begun to track administrative hearings. While the SSP tracks probation violations of its participants, it might be helpful to collect data on the frequency and type of probation violations for all youth. In addition, the county may find it helpful to explore new ways to increase families’ understanding of the juvenile justice system and support parents as they navigate the system and help their children succeed during their probation. The disproportionality of minority youth referrals to the juvenile justice system remains a challenging problem that requires many more active participants to address successfully. Concerns related to disproportionate minority contact with the criminal justice system and negative encounters with the education system in the Travis County/Austin area include the following: African American and Hispanic individuals are more likely to be searched by police than whites.11 African American and Hispanic students are more likely to undergo discretionary removal from school than whites.12 African American and Hispanic youth are more likely to appear in juvenile court for nontraffic offenses than whites, and are particularly overrepresented in appearances related to school-related offenses and disorderly conduct/abusive language.13 African American and Hispanic youth fail to appear in municipal court and have warrants issued for their arrest at much higher rates than whites.14 Stakeholders could examine referrals from school system police to discover whether school-based discipline issues can be resolved without involving the juvenile court. To address disproportionality in referrals, any initiative would need to take into account the patterns of policing, arrest, diversion, and referral to begin to address the disparities. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-19 Recommendations for Race and Ethnicity Data Collection Addressing data collection at the State level will be important to ensure the most accurate picture of the juvenile justice system. Although collection of race and ethnicity information does not provide a complete picture of the needs of a juvenile justice population, it is an important starting point to accurately determine the population served. Both counties in this site-based project faced challenges while attempting to improve electronic data collection on race and ethnicity. The constraints of the data system in the Travis County site precluded disaggregated collection of race and ethnicity information at the local level because of the prior existence of a statewide database that the probation department was required to use for data collection. Other DMC reduction projects, such as the one in Pennsylvania, have made changes at the state level that enabled the state to collect more accurate race and ethnicity data. The Center for Children’s Law and Policy (CCLP; their staff were responsible for the Travis County, Texas, part of this project), along with the National Center for Juvenile Justice (NCJJ), developed guidelines on race and ethnicity data collection for the Commonwealth of Pennsylvania that could easily be adapted for other States’ purposes (Guidelines for Collection and Recording the Race and Ethnicity of Juveniles in Conjunction with Juvenile Delinquency Disposition Reporting to the Juvenile Court Judges’ Commission [JCJC], 2006). The JCJC guidelines incorporate the requirements for data collection by all Federal agencies set forth by the U.S. Office of Management and Budget (OMB). The OMB standards are used not only in data collection for the U.S. Census but also to count populations involved in medical research, mortgage lending applications, and other government-related activities. The focus of the JCJC guidelines is as follows: [To] enhance the accuracy of the demographic information collected by the Federal Government by having categories for data on race and ethnicity that will enable the capture of information about the increasing diversity of our Nation’s population while at the same time respecting each individual’s dignity. In the approach required by OMB and adopted by the Pennsylvania JCJC, race and ethnicity are two separate questions, and a third optional question allows respondents to describe any other country of origin, ancestry, or tribe with which they wish to report an affiliation. Interviewers are instructed to use self-identification as the primary method for answering the questions; observer identification or reference to written documents is used if the youth, parent, or guardian does not answer the questions. The data system allows the recorder to indicate whether or not the information is based on self-identification. To prompt self-identification, interviewers are instructed to begin the series of questions by explaining, ―I am now going to ask you some questions about how you prefer to describe yourself.‖ The first question asked is, ―Are you Hispanic or Latino?‖ By not asking, ―What is your ethnicity?‖ the question avoids introducing additional categories or any misunderstanding about the meaning of ethnicity. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-20 The second question, ―What is your race?‖ allows for answers in five categories: American Indian or Alaska Native, Asian, black or African American, Native Hawaiian or other Pacific Islander, and white. For both race and ethnicity questions, coders are encouraged to avoid the unknown category unless the youth specifically requests that unknown be checked in addition to another race, or the youth is not present and the information is missing from other sources. The optional third question, ―Do you identify primarily with a particular country of origin, ancestry or, if you are Native American, a particular tribe?‖ allows counties to collect information about populations or subgroups not listed in the first two questions and provides an opportunity for youth and families to self-identify with groups meaningful to them if the county chooses to incorporate this question. Pennsylvania has been adapting its statewide database to accommodate this approach since its inception in October 2006 and has been conducting implementation training across the State. Although the Pennsylvania work is outside the scope of the project summarized in this report, the Commonwealth’s State-level data collection method serves as a model for jurisdictions seeking a standardized, clearly explained approach to more accurate race and ethnicity data collection. Lessons Learned This project’s two main goals were to develop new and accurate data collection methods for Hispanic youth in the juvenile justice system and to reduce DMC for Hispanic youth at critical decision points in the system. At both sites, despite the existence of a Federal model for race and ethnicity data collection, existing data protocols that the probation departments were required to use precluded the disaggregation of race and ethnicity data. However, this project enhanced data collection in other ways. At both sites, researchers conducted file-based research to better understand the factors that contribute to overrepresentation at the chosen system contact points. The case reviews revealed deficiencies in the systems’ ability to track the youth’s interventions. The case review instruments provide an interim DMC data assessment tool that may be used until automated data systems are modified to track the elements necessary to complete post intervention assessments. At the Washoe County site, a process for collecting information on the language proficiency of youth and their families was developed. Instead of just asking youth whether they and their parents spoke English, a simple limited English proficiency questionnaire was developed. This project provided lessons to participants in three main areas. Data collection, an essential component to any effort to reduce racial and ethnic disparities remains an ongoing challenge for many communities across the country. Fashioning interventions that follow from the data requires involvement of traditional and non-traditional community stakeholders to ensure that programs meet the needs of the youth and the DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-21 juvenile justice system. Communities committed to reducing racial and ethnic disparities must commit the time and resources to careful analysis and program development. Some lessons from the project are described below. Data Collection Several additional lessons were learned, as described below. Although reduction in disproportionate minority contact with the juvenile justice system needs to occur at the local level, reforms in data collection may need to occur at the State level before county data systems can be changed. Although counties may appreciate the value of disaggregating race and ethnicity information to better understand the juvenile justice population, they may still be hampered by the limitations of statewide databases they are required to use. In States where information sharing between law enforcement agencies is significant, not only must data solutions occur at the State level, but also stakeholders must share information with other agencies so that systems are updated simultaneously and in similar ways. Targeting the collection of particularly relevant data elements can yield considerable benefits when developing interventions to reduce disproportionality of minority contact with the juvenile justice system. Systems will contain data that better address disproportionality if the tracking is conducted systematically, for example: the language proficiency of parents and youth, the countries where parents and youth were born and reside, the alternatives to incarceration/interventions used in each case, the successful completion of or reason(s) for termination from programs or interventions, the types and frequency of probation violations committed by youth, and their health and education status. Case file review is time intensive but can be a helpful tool for learning about municipal, county, and State systems when data from their automated databases cannot answer all of community leaders’ questions needed to develop strategies for improvement. A case file review may also provide clues as to how to modify the data fields in an automated data system to more accurately capture the desired data. Staff who engage in file reviews should be trained and supervised to ensure consistent methods of data gathering and accurate interpretations of the data. The individual case notes, treatment plans, court pleadings, social studies, and other documents typically found in case files of youth in the juvenile justice system are full of helpful information but are subject to different interpretations. In the initial phases of case file data collection, it is helpful to have more than one researcher review the same file and compare notes for accuracy. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-22 A combination of quantitative and qualitative information can be especially helpful when trying to understand how different groups experience the juvenile justice system. Interventions to Reduce Racial and Ethnic Disparities Resources from local public agencies may be leveraged to achieve needed reforms. When designing a new program, staff, youth, families, and other stakeholders can provide important insights into programming, supervision, and cultural competence needs. Community-based agencies are valuable resources for providing culturally competent services to clients close to their homes. Juvenile justice agencies should look beyond their traditional community partnerships to identify and fill gaps in services. New programs developed by juvenile justice systems to reduce incarceration must establish admission criteria that ensure a reduction in incarcerations rather than widening the net and involving youth who would not otherwise be incarcerated. Although jurisdictions may feel they are more successful if they target a decision point with the highest RRIs and greatest number and/or magnitude of contacts for further assessment and intervention, sometimes other factors, such as difficulty getting the cooperation of the necessary stakeholders or overcoming funding barriers, will lead a jurisdiction to choose another decision point to tackle first. After experiencing some success, the jurisdiction may then be ready to tackle the decision points with more significant disparities. Time and Resource Commitments DMC reduction efforts are time and resource intensive and will require a prolonged commitment that transcends one community leader, champion, or administration to sustain. Sufficient time and resources must be devoted to both data collection issues and substantive interventions. To support a local DMC effort by achieving a statewide policy change first is time consuming, requires different resources than those needed at the local level, and may require a separate effort to achieve the statewide objectives. Reducing the incarceration of youth of color may have other collateral benefits, such as reducing overcrowding in facilities and improving and expanding relationships between youth and community service providers. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-23 Endnotes 1. This chapter was adapted from the final report, dated November 29, 2007, for OJJDP award 2004-JL-FX-0083, to the Youth Law Center (www.ylc.org), San Francisco, California. 2. Although the RRI for African American youth was 20.09, the low volume of African American youth in this population (7 of 89) meant that the project’s DMC focus was on Hispanic youth. Findings related to African American DMC have been omitted. 3. Limited English proficiency (LEP) is an official designation under the federal No Child Left Behind Act (see LEP fact sheet at http://www.ed.gov/nclb/accountability/schools/factsheet-english.html). This designation carries certain legal obligations for schools receiving federal education funds and states receive specific federal education funding to serve students with the LEP designation. 4. DMC data for African American youth are not included in this report. 5. These findings are from Watson, J., Rogers, E., and Miller, J., Unlocking the Keys to Success, a report by the Travis County Juvenile Probation Department, July 2006, p. 1. 6. Research was supervised and compiled by Francisco A. Villarruel, Ph.D., University Outreach and Engagement Senior Fellow, and Professor of Family and Child Ecology, Michigan State University, with assistance from the Center for Children’s Law and Policy, July–August 2006. 7. This information is based on a conversation with Chief Estela Medina and staff of the Travis County Juvenile Probation Department, on November 9, 2006, and with Britt Canary and other staff of the TCJPD, on April 4, 2008. 8. Conversation with Britt Canary and other TCJPD staff, on April 4, 2008. 9. Data provided by TCJPD, April 11–29, 2008. 10. See note 8. 11. African American individuals are more than three times as likely to be searched by the Austin Police Department as whites, and Hispanic individuals are 2.3 times more likely to be searched than whites. African American and Hispanic individuals were less likely than whites to be in possession of contraband when searched (0.8 and 0.9 times, respectively). Search data from the Travis County Sheriff’s Department indicate that their officers are 1.5 times more likely to search African Americans or Hispanics than to search whites. (From a report on racial profiling, Don’t Mind If I Take a Look, Do Ya? An Examination of Consent Searches and Contraband Hit Rates at Texas Traffic Stops, prepared by Dwight Steward, Ph.D., Steward Research Group, and Molly Totman, Texas Criminal Justice Coalition (TCJC), on behalf of the American Civil Liberties Union of DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-24 Texas, National Association for the Advancement of Colored People Texas, League of United Latin American Citizens Texas, and TCJC, February 2005.) 12. Austin Independent School District data reveal that, for discretionary removals from school (those not required by law), African American youth represented 31.3 percent of the removals, despite being only 13.3 percent of the school population. Hispanic youth represent 52.3 percent of the discretionary removals although they comprised 54.7 percent of the school population. White youth represent 15.4 percent of the discretionary removals but constituted 29 percent of the school population. (From the Austin Independent School District Report, with all of the data from August 17, 2004, through May 25, 2005.) 13. In Austin Municipal Court, of youth ages 10 to 16 charged with nontraffic cases from August 2004 through July 2005, African American youth represented 20.3 percent, Hispanic youth represented 58.2 percent, and white youth represented 20.4 percent. For offenses that occurred solely on school campuses (disruption of classes, trespass on school grounds, or loitering on school grounds), African American youth represented 31.7 percent of the cases, Hispanic youth represented 61.8 percent of the cases, and white youth represented 6.2 percent of the cases. One of the most starkly disproportionate charges was for disorderly conduct/abusive language: African American youth represented 30.6 percent of the cases, Hispanic youth represented 65.7 percent, and white youth represented 3.7 percent (statistics provided by Judge John Vasquez, Demographic Profile of Class “C” Misdemeanor Charges Filed in the Austin Municipal Court Between August 2004 and July 2005). 14. For 2005, the Austin Police Department’s Juvenile Accountability and Community Service Office (JACS) processed 224 warrants out of 470 citations of African American youth. For Hispanic youth, officers processed 501 warrants out of 1,428 citations. For white youth, officers processed 80 warrants out of 559 citations (from the Austin Police Department, Juvenile Unit, JACS Office, February 3, 2006). DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-25 References Holman, B. 2001. Masking the Divide: How Officially Reported Prison Statistics Distort the Racial and Ethnic Realities of Prison Growth. Alexandria, VA: National Center on Institutions and Alternatives. Hsia, H.M., Bridges, G.S., and McHale, R. 2003. Disproportionate Minority Confinement: Year 2002 Update. Summary. Washington, DC: U.S Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. NCJ 201240. http://www.ncjrs.gov/App/Publications/abstract.aspx?ID=201240. Leonard, K.K., Pope, C.E., and Feyerherm, W.H. (Eds.). 1995. Minorities in Juvenile Justice. Thousand Oaks, CA: Sage Publications. Office of Management and Budget Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity, Part II, Notices. Federal Register, October 30, 1997, Vol. 62, p. 58781. Poe-Yamagata, E., and Jones, M.A. 2000. And Justice for Some. Washington, DC: Building Blocks for Youth. Retrieved March 3, 2008, from Building Blocks for Youth Web site: http://www.buildingblocksforyouth.org/justiceforsome/jfs.html. Torbet, P., Hurst, H., Jr., and Soler, M. 2006. Guidelines for Collecting and Recording the Race and Ethnicity of Juveniles in Conjunction With Juvenile Delinquency Disposition Reporting to the Juvenile Court Judges’ Commission. 2006. Center for Children’s Law and Policy and National Center for Juvenile Justice. Retrieved March 3, 2008, from Pennsylvania Juvenile Court Judges’ Commission Web site: http://www.jcjc.state.pa.us/jcjc/cwp/view.asp?Q=395139&A=7&jcjcNav= 6178 . Villarruel, F.A., Walker, N.E., with Minifee, P., Rivera-Vasquez, O., Peterson, S., and Perry, K. July 2002. ¿Donde Esta La Justicia? A Call to Action on Behalf of Latino and Latina Youth in the U.S. Justice System. East Lansing, MI: Institute for Children, Youth, and Families, Michigan State University. NCJ 196500. http://www.ncjrs.gov/App/Publications/abstract.aspx?ID=196500. DMC Technical Assistance Manual, 4th Edition • Chapter 7: Strategies for Serving Hispanic Youth 7-26 Appendix A: Washoe County Juvenile Services Limited English Proficiency Questionnaire Youth’s Name: ____________________ Age: _________ 1. Do you feel comfortable answering these questions in English? ___Yes ___ No If no, you do not have to answer the rest of the questions. 2. What is the language that you first learned to speak? ___ English ___ Spanish ___Other 3. Even though you can speak English, what language do you feel most comfortable speaking? ___ English ___ Spanish ___Other 4. What language do you most use at home when you are speaking to your brothers/sisters or other children at home? ___ English ___ Spanish ___ Other 5. What language do you most use at home when you are speaking to your parents? ___ English ___ Spanish ___Other 6. Do your parents speak English? ___ Yes ___ No 7. Are they fluent in English or do they need help? ___ Yes ___ No 8. What language do you speak when speaking with friends outside the home? ___ English ___ Spanish ___Other 9. When you were attending school, were your classes in English? ___ Yes ___ No 10. Did you ever attend any classes to help you learn English? (English as a Second Language or ESL classes) ___ Yes ___ No 11. Do you think that you need more help to learn English? ___ Yes ___ No 12. Even though you can speak English, can you read in English? ___ Yes ___ No 13. Do you think you need more help to learn how to read in English? ___ Yes __ No 14. Can you read in your home language? ___ Yes ___ No. Appendix B: ADOMS ll Risk Assessment Tool I?l' .. ans-ms II Risk Assess I?ll?: :4 Juvenils Nr-Irn re: TEJPD CUE HT Status: ACTIVE D?'lccr: JGE Earn plsim Number: Author: Nancy Snyder Date Number of Prior current] {3 1 3 r. 5 More . 2. Number of Prior Counts Au?judicatEMore 3- Age at ?rst rslerrs! C3 13 ur Tcungur 14 cr 15 C) 13 cr cider 4. Number cl Referrals Em Assault {If 1 (J 2 :rr Marc 5. Numh?f sf Hs1srrals fur TthCar Theft 0 In 2 3 or More 5. Number 0? Hefcirsls fur 31.5 3 cr Mare T. Number c1 {or C) L1 Burglary?nbbay?mspass?am agar [Include current] 1 tn 3' 4 or More I 8. Deiinrzusm Fear Gruunr?anu Al?lls?c-n H: .1 Ho I Deliriquenl Gr?up I?d-kl .3 Gang Af?liation a. Mending School without '3 No mettle-ms IE: ?35 Hr 1? 4? Keel-m Alechel or Dru-g Problems Hens Drug Preblems TDTAL ECDFIE else Law 1. What is 1he Mnsi Serious CURRENT .I'idjudicatec IEJffense?i Runaway Dita-nee Code Offense E53 55 Sub Class Offense Category Eiffense Severity BUSH 1:3 Offense Desoripfion Ftu naway 51 .03} 2. What is the Most Serious Adjudicated foensei? foense {lode Offense Class Set: Ella ss Offense EadegorgiI Offense Offense Abbreviation 3. Based on D?ense Severity quesii one 1 and 2, the meet serlcue current or prior adjodlcated effecse category Is; 1 -eINs - Misdemea nor A or with no use or passessien of ?rearm 1:3- 3 - Misdemeanor with use or possession of firearm or 3rd or 4th degree teen},- . 4 - 2nd degree fetcny 5 - 1 degree nan-aggravated terrain- - degree aggravated felon:..r - Capital Felony 4. Recommended Dispesition: tJ nkeevm Discretionary ?uerride: 5. Overnde Disposition Close: :3 is More Restrictive F-acement 3 merride to Less Reelric?ue Placement at No Override I ?nnh-inn-Hnl l'f-l 4- I Oelirquent F'eer IEre-up Gene Af?liatien El. Attending SehedI willieut elseplinery Me pre-blems .3. .1, es f' Hnewn Aleehel er Drug Problems '31 Nerte Marijueneteleehelt?mer Elmg F'I?dbleme TOTAL El- mer: LEVEL Law . 1. What is the Meet Seridus OLJRRENT ediudteetee Offense? Runaway Offense Cede Offense Oless Sub Oless Offense Category Offense Severity 113 Offense Runaway Fetsmsi - 2. What is the I'd-est Eeriuus PRIOR Adjudieeted O?ense?r?j Offense Oede Offense Class Sub Ole ss Offense IOe'IegeryI Offense Severity Offense Abbremetlen 3. Based en Offense Seeder}r questidns 1 end 2. the meet seridus current dr pridr adjudieeted e?er-se is: 1 -eIr~is 2 - Misdem eener A er with ne use er peseessien df ?rearm C3 :3 - MisdEmt-EI-Lrtur with use er pessessien df firearm Of 3rd or degree t?e'drqiI 4 - 2nd degree fefen'gir he 5 - 1 degree Item-aggravated fefdny' '3 - tet degree aggravated felengr if} - Oepitel Feldny 4. Reedmmended OIspe-sitlen: United-um Override: 5- Override Dispesitien Oede: :5 Override td Mere Restrletl're Override te Lees Restrietwe Placement Ne Override I E. Override :aetnrs: [13} Family stperl is present and stable- LI [14} Treatment needs best in a less'sesure setting- [16}Mitlgatlng e-reernsta rises related ten the effenee. [wt-Shim is net 3 serieus publie safety.I risk. [13} ether Cnher Esplanatien; a Flea se press button ts he left is add an explanaiisrl. Discre?nnar; Userrids Esplanatien'. 2] Please press bulten ta the left te add an explana?en. 3-. JL diei all Demise Dispssilisn Cede-1:3 1 - F?retlatien s-s 2 - Freesten at Hemsn'?ats '3 - Fmbehen at Hem eJ'Ca1s [JrH'u'ilZIIL-lj It: {Jmuid-E LIIlEIi-si'e's in I: {21] Pattern Elf wielent hehatulm' re?ected in the record. {22} Use ef mapen in eurrenl sr past sffenses- I: {23:6 es: abuse effense or multiple sex affense in reenrd- I: {24}?f?uth is high public safety {26} lnstiequeie suspert er supmisien. l?I {snether rum-a.? INTAKEIOETEMTIOH FORM CHILD Je _nTrr?ton see . ON PROBATION ON CONDITIONS I PSI: t??tOE? OFFENSE CAUSE: ?y QUUENILEF INFORMED OF [efroleji _i intake orooess i Detention nean'ng fJ .i Court prooeec?inos .II DPLJ (meditation. 33R. No r?inos .i Referred TO: NOTIFIED OF OET. Wig RNET. SOHOOL. OPRS, Other: Wifl osrentiguardian be 'ent For detention I?iesring'?I Cl ~'r'es 5 ND Why'- not? Oo need interpreters {Iangusgei?sig?'f El?f?es El NU Requested Interpreter? ?res MD Reason fin Detention {Fraii'nilyr Code 53.02} DLiker to absoond or he removed from the jurisdiction DI seurt. DSoiteole supervision, oars, oroteotion nor being provided by parent. euardian, oustodian or other person. leo osrentt guardian. eto, able to return ohiid to court. be dangerous to self or t'nsz,r threaten the safety of the points if released. feund to he delinquent child or previouslyr sonviotod of a oensl offense ounishehie byjsiifprison and LiIteiy to commit an offense if released. alleged to have engaged in delinquent oonduot and to have used, possessed. or exhibited a firearm, ?ts defined by Section 45.01! Penal oode. in the oomrnissien of the offense. Assessments [tone {sirele}: Oomments: mess: .. POE ELIE B?f EASE WORK MANAGER: I .4 .. H- i. DECIDED TO RELEASE TH [53 BECAUSE OTHER ISSUES: 11" ?it-'1 THE hietTt'Ei-l Chi"- 3 THE i327? ESTHER SGUEF GF 5 fi?lli?ii'?. EE-iT'i'li'iiE ME THE UFJ lit-l 5MB Jj??i'ulT?i'. liar-1i; ilL J?l?i? L'l1 The court tunes that it is in the laeet interest the child for tl Ie child te he piecet? ei hisl'her heme in! ih? teltewtne Check the aperep'ttete ctilerie te e'etatn; tlieu indicete which epectlic tar each. A blank heat is incitIdE-Zi it?! Eli't}f specific reaeur lh' net iiar'ae'. the child is lilteti; te altecencl treru the til the ceurL The child has at hiettu'glr cut running away. The eereel has threatened le t'emeire the chittiirem the iurtaciictien til the ceuri- 3. tither: Suitable care. preteetim ta net being etevidect let; parent, guardtet'r. cilai?dierl. Di ether eereen. I .l the eerertlieuetdteh taclte suf?cient parenting eittlie ta ereil'ttla afiequete eupervieien fer the child Due in .hitre-tity. er advanced age at the earet'ttiguetciian the child tic-ea net edeetie eupen-teien. .- rite-equate eletitirig, icing-ti. er is not being 'i'er the child. The eatent'elgearciteh'e certtirrnet'i euuetaece ehcllae :iLJte the Child at riali. the eerent'eiguardteri'e eurttitmerl intreliremal?tt in criminal activities the ehitcl et tielt. . '2 Chile accepi eai'eiltel Either; The lizii'ii itee he Daren: cttetetiiet a. erethci Inereert ee'e :e i-etli'1 i'iiFiliilE't" le 1 .etltIi-?ett The cert net he lecetetl the paleetrgeertliart 'il?E mcane rut it: return the chilc Le the ceurt he eerentiguetdiai'i te eetee tn; the tleiertetl The te ceeuerete with the el'tlere ei the ceur?t. 1 filthei - emu,- may [3.51. {term-him 15; it: i'tit'rieetIl'lLei'EEii u; met,r threaten the Plate-lb,- ei the iltiijiit'. ti I the cl'ti'lti has :Lutattleii itieatienc cl has; ptelrltittieitr alterhetEL: the cltilti hae hiettn'y ei aeti-rtiiltilatien. 't'hc has. hie?tet'y et tetrt-aiti athere- the chilrl ueetl a i ire-ere: ril' tithe: weapeh .ill cemmieateh et alleged t'i?itenee. t.t Hitter 515: "rule hee preview-ll]: Een te lee i'les'lrteLreti?. hel; heeu I'ii litrt'IF'. ?ery-"I 1- er in teen .?tnrl IE: i_ Il ii felt-tastitl. 'iti:- chi-ti in". "trehnttehl'ea ?it-I. l'iia -:iH rl ll tin-seas? eerie Htliltelie't - tiling .- . r'iII'Ii Ell-'il: I-iT'rli'l'IIFI-flii'll'l ?fth l. ?ti-5' Ii'iF'idili-i .i .c See- Ei'l. IDE. RIGHT TD BE Cit: PROCEEDIHE. let The earent eta ehitd reterree te a ta entitled as seen ae practicable after the teterrai te ee intermed_by__e_tatt deeIehated he the beard. teaeed en the intermatien the referral te the tLItIenite eeLIrt, efr .- . lug-ru- .. 5? 'I'i the date end time ei the ettanee: Li {Ell the date and time the ehild tahen inte eIIetedII; ll the name e'i the e'ttenee and tie eenel I I {It} tL_?gIee et' if thatwae ehild waa 'tahen iIIte with 3dL?t?_ er ether names ef i ?Hit the aspect-a rIf the itIy?nite eeLII't that tn the etIi_l_I_t;_ tee it I the ehild'e Iieht te be er en _atterneItand the leeal etandaide and ter determining whethel the parent euali?ee tea et te the ehild; and tji?i the methee?e theeam .1 ean the ehiiIt with the ee. tel 'It the ehild wee released en tield reteaee I'I'err: the entereeineht etatien be the eetiee. he intette, eI' ettijte iIIdee er iLIdtIe at the Itetentten hearine thete .ferrnatien rectum-edgy {raj-113.? he eernreLIHI'I?etEd tet eehene er in written. te} It the ehiid' Ia net heteIe eI at the initial detentien he -aIinI_:I the inteima'tien he ta} ehalt tIe eentrnenieated te the matte??mt event it he eernrnunieated teleehene er in _td?I =.fernIe.tien te a earent under Ie net admie eeiele in a endej tI_"Iie_ti_tIe ea eehetantiee egidene?e er_ ee te the teetirnenII _e_t fer the state. --I'eIIe-ieI tee?; e- _Ie._t Appendix C: NCJJ Guidelines for Collecting and Recording the Race and Ethnicity of Juveniles in Conjunction With Juvenile Delinquency Disposition Reporting to the Juvenile Court Judges’ Commission GUIDELINES FOR COLLECTING AND RECORDING THE RACE AND ETHNICITY OF JUVENILES IN CONJUNCTION WITH JUVENILE DELINOUENCY DISPOSITION REPORTING TO THE JUVENILE COURT COMMISSION IchJ #333 :33 E63 3:33:63 4 @q V33 3333:@3< @3 033333333 :3 @m 333333 :3 @@3333@3@3 333 33 3 333 3333333 @333@333@< 33M 3333333333 333333333 33 @3333 3@ 333 3333333 @@333 36 3 @@333 3333335 @@3;3@m3@3 3; @333m 333333333 33 3 333333333 3?3 33 V3 333v 33??33333 33 3333 333 33?3 333v 3333:@3< @m V33 3333:@3< @3 33333333 :3 @m 333333 :3 @@3333@3@3 333 03 3 333 3333333 @333@333@< 33M 3333?333?3 333633333 33 @3333 3@ 333 3333333 @@333 36 3 @@333 E33333q 3p @333q 333333333 33 3 333333333 3?3 33 V3 333v 33??33333 33 3333 333 33?3 333v 3333:@3< @m V33 3333:@3< @3 33333333 :3 @3 333333 :3 @@3333@3@3 333 m3 3 333 3333333 @333@333@< 33? 3333.3 3333933393 33393333 33 PREPARED FOR PENNSYLVANIA’S JUVENILE COURT JUDGES’ COMMISSION by Patricia Torbet, Hunter Hurst, Jr., National Center for Juvenile Justice Mark Soler, Center for Children’s Law and Policy October 2006 © National Center for Juvenile Justice The preparation of this booklet was supported by John D. and Catherine T. MacArthur Foundation grants awarded to NCJJ and to CCLP Anyone may use the content of this publication as is for educational purposes as often and for as many people as wished. All we ask is that you identify the material as being the property of NCJJ. If you want to use this publication for commercial purposes in print, electronic, or any other medium, you need our permission. If you want to alter the content or form for any purposes, educational or not, you will also need to request our permission. PAGE 01 INSTRUCTIONS AND GUIDELINES FOR COLLECTING AND RECORDING RACE AND ETHNICITY This booklet provides instruction and guidance to local juvenile courts and probation departments on racial coding of juveniles involved in Pennsylvania’s juvenile justice system in conjunction with reporting juvenile delinquency dispositions to the Juvenile Court Judges’ Commission.1 There are compelling reasons for accurate coding, not the least of which is to ensure the fundamental fairness principle outlined in the mission of Pennsylvania’s juvenile justice system. The mission states “...all of the services designed and implemented to achieve this mission and all hearings and decisions under the Juvenile Act—indeed all aspects of the juvenile justice system—must be provided in a fair and unbiased manner.”2 Both the U.S. and Pennsylvania constitutions guarantee rights and privileges to all citizens, regardless of race, color, creed, gender or national origin. BACKGROUND The Federal Juvenile Justice and Delinquency Prevention Act requires states to assess the extent of Disproportionate Minority Contact (DMC) of youth of color at all stages of the juvenile justice system and to take steps to address any disproportionality (or overrepresentation).3 Pennsylvania is at the forefront nationally for its ability to track these indicators for youth at various stages of the juvenile justice system. The Juvenile Court Judges’ Commission and its Center for Juvenile Justice Training and Research are to be commended for steps taken to improve the reporting of race and ethnicity in the juvenile court disposition data. However, several obstacles existed in the spring of 2006 when JCJC and CJJT&R sought specific guidance from the National Center for Juvenile Justice and the Center for Children’s Law and Policy to address them. It was decided that the goal—to enhance the accuracy of the racial data collected by juvenile courts in Pennsylvania consistent with Federal policy—could be achieved by resolving outstanding issues related to compliance with Federal standards. This could be accomplished by making changes to variables and codes for juvenile court disposition reporting, providing instructions to local juvenile court and probation department staff who are responsible for collecting and reporting the data, and providing guidance on analyzing and using racial data to monitor practice and sharing the information with other stakeholders. National Center for Juvenile Justice OCTOBER 2006 Why it isn’t easy For many people, their identification with a particular race or ethnic group is a deeply personal and sensitive issue. For government officials, statisticians, and others concerned about it, race classification is a substantively complex issue. It is also an imprecise cultural construct that changes over time. For example, the Census 2000 questionnaire offered 15 choices for coding a respondent’s race even though the 1997 Federal standards, promulgated by the White House Office of Budget and Management (OMB), set the minimum race categories at five. CENSUS 2000 RACE CATEGORIES FEDERAL MINIMUM RACE CATEGORIES White American Indian or Alaska Native Black, African Am., or Negro Asian American Indian or Alaska Native, print tribe Black or African American Asian Indian Native Hawaiian or Other Pacific Islander Japanese White Native Hawaiian Chinese Korean Guamanian or Chamorro Filipino Vietnamese Samoan Other Pacific Islander Other Asian, print Some other race, print To add to the confusion, OMB placed special emphasis on identifying the Hispanic or Latino population group. Unlike “African American,” which is a race, “Hispanic/Latino” is an ethnicity, not a race. Accordingly, Federal guidelines recommend asking two separate and distinct questions and the order in which they should be asked, the first asking respondents to indicate their Hispanic or Latino ethnicity and the second asking for respondents’ race. (See the sidebar for more information on the Federal standards.) Currently, the Federal government is only interested in distinguishing one “ethnic” group (or “origin,” as it is sometimes labeled). This has created some confusion in states that added a separate variable, typically labeled “ethnicity,” because just one ethnicity is of interest.4 Particularly problematic for local data collectors is what questions to ask of youth in order to accomplish accurate racial coding. PAGE 03 Why it’s important to describe a juvenile’s race and ethnicity accurately Even if the mechanics of asking questions and coding are resolved, why is accurate recording of a juvenile’s race important to local juvenile courts and probation departments? What’s the point beyond statewide reporting or some distant compliance report to the Federal government? In the aggregate, there are several. Juvenile court and probation administrators need to: • • • • Know “who” the system is serving Know “what” services or resources are needed to respond to the youth and families the system is serving, such as the need for cultural competency training for juvenile court staff, culturally appropriate programs and services for youth and their families, translators and interpreters, Spanish-language documents and materials, and bi-lingual and bi-cultural staff Monitor and examine “how” the system responds to youth of color Share this information with stakeholders and in annual reports to the community. The payoff for administrators following this guide is the ability to report information consistent with Federal policy while preserving the flexibility to describe local ethnic diversity of juveniles referred to the juvenile justice system. Monitoring DMC in Pennsylvania Since 1989 the DMC Subcommittee of Pennsylvania’s Juvenile Justice and Delinquency Prevention Committee has used arrest, juvenile court, and detention admissions data compiled by NCJJ to monitor statewide trends in the handling of youth involved at various stages of the juvenile justice system, identify emerging problems at certain stages for some groups, and target finite resources for system reform. The data have also been used to track the extent to which members of minority groups are beneficiaries of alternative processing options such as diversion from court or home detention. Local juvenile justice stakeholders with access to these indicators at the county level can begin to “look for the story behind the numbers” and develop strategies to assure nondiscriminatory decision-making across population groups and identify areas that may need more in-depth examination. With guidance from the DMC Subcommittee, the Center for Children’s Law and Policy, under Pennsylvania’s partnership with the MacArthur Foundation’s Models for Change initiative, is working in three Pennsylvania counties to help facilitate an examination of racial data at the county and neighborhood levels and to plan system improvements National Center for Juvenile Justice OCTOBER 2006 intended to reduce disparities.5 JCJC’s ability to report DMC data statewide was one of several reasons the MacArthur Foundation selected Pennsylvania as the first Models for Change state. OBSTACLES RESOLVED At a meeting convened by JCJC with representatives from NCJJ and CCLP, consensus was reached that resolved outstanding issues related to compliance with Federal standards and subsequently resulted in changes to the variables and codes for juvenile court disposition reporting.6 The collection and recording of racial data will be handled with three questions or variables, with the first two limited to fixed responses: 1. Hispanic/Latino? (Yes, No) 2. Race (5 categories) American Indian or Alaska Native Asian Black or African-American Native Hawaiian or Other Pacific Islander White 3. National Origin, Ancestry or Tribal Affiliation (any population group or subgroups not included in the first two questions) First and foremost, it is hoped that these changes will make it easier for local staff to code racial data thereby reducing the instances of unknown or missing data and improving accuracy and consistency of the racial data collected by juvenile courts statewide. The question format, fixed order of the questions, and fixed coding structure for the first two questions comply with Federal standards. The optional third question provides flexibility to counties that wish to accommodate local preferences for capturing a world of different affiliations with other population groups not included in the first two variables, while ensuring that the Federal government’s standards for minimum race categories are met first. JCJC asked NCJJ and CCLP to develop coding instructions and guidelines reflective of these changes. It is hoped that this advice reflects the reality of situations coders face when they are sitting across the desk from a youth or making decisions based on a paper review rather than an in-person interview. PAGE 05 RACIAL CODING INSTRUCTIONS Information Sources: There are three possible information sources for capturing racial data: 1. self-identification on the basis of an interview with the youth/parent/guardian, 2. observer-identification when the youth/parent/guardian fails to answer the question and the observer infers the answer, and 3. some other source on the basis of a report, face sheet or complaint filed with the court. Self-identification is the preferred source of information for collecting racial data. The guidelines in the next section provide advice for coding racial data depending on the source of the information. What changed? Coders are asked to indicate whether answers to the Hispanic/ Latino and race questions were self-reported by the juvenile/parent/guardian, recording “yes” (Y) for self identification or “no” (N) for identification by the observer or some other source. Question Order, Format and Acceptable Answers: Order of Questions: Ask the questions in the order specified: 1. Hispanic/Latino question 2. Race question 3. Optional, open-ended, question about identification with other population groups not listed in the first two questions. FYI: The first two questions force a rubric to accommodate current Federal policy on racial coding. The third question is open-ended and can accommodate any self-identity. The ordering helps to reduce confusion introduced by the multiquestion format. Prompt to Self-identify: Begin the series of questions with a prompt: “I am now going to ask you some questions about how you prefer to describe yourself.” This prompt links the questions and encourages the juvenile to self-identify. First question: Are you Hispanic or Latino?” Acceptable answers: ❏ Yes, Hispanic or Latino ❏ No, Not Hispanic or Latino ❏ Unknown (limited use) National Center for Juvenile Justice OCTOBER 2006 What changed? The ordering of the questions now puts the “ethnicity” question before the race question. The variable label of “Ethnicity” has been eliminated in favor of the label: Hispanic/Latino? The question, “What is your ethnicity?” has been replaced with the new question, “Are you Hispanic/Latino?” The biggest change, however, is that the new question/variable will have fixed “yes” or “no” answers. Previously, counties were permitted to add other ethnicities in this field, which were then recoded into Hispanic or Non Hispanic. The coding of other ethnicities will be accommodated by an optional third question. Identification with any of the other subgroups Federal policy characterizes as Hispanic or Latino (e.g., Cuban), can also be accommodated in the third question. “Unknown” should be limited to situations in which the youth is not seen and the information is not provided by the referral source. FYI: The label “Hispanic or Latino” takes into consideration regional differences in the usage of the terms, supposedly between the eastern and western United States. “Spanish” was added to the label by Census 2000, but is not required by Federal policy. Second question: “What is your race?” At this point in the questioning, hand youth a printed/laminated card with the 5 race categories. This will assist them in answering the question since reading the list out loud to them would be very confusing. Prompt the youth by asking, “Please tell me which race you consider yourself to be. You may select more than one.” Acceptable answers: ❏ American Indian or Alaska Native ❏ Asian ❏ Black or African-American ❏ Native Hawaiian or Other Pacific Islander ❏ White ❏ Unknown (limited use) What changed? The previous reporting of racial data permitted the coding of “other” race. This category has been eliminated and is not an acceptable answer. Identities outside the five minimum race categories will be accommodated in the next question. The biggest change, however, is the application of a new rule. Because many youth are multiracial, youth may identify with more than one race. The prompt for them to choose more than one race will facilitate the application of this new rule. The PAGE 07 interviewer should follow the “mark any that apply” rule based upon the youth’s self-identification of multiple races or by the observer’s identification. The use of “Unknown” should be limited to situations in which the youth is not seen and the information is not provided by the referral source, or in situations in which a youth specifically requests that “unknown” be checked in addition to another race.7 Third question: (Optional) “Do you identify primarily with a particular country of origin, ancestry or, if you are Native American, a particular tribe?” What changed? A new question with the variable label “National Origin, Ancestry or Tribal Affiliation” has been added. Previously, counties were permitted to record other origins or ethnicities in the “Ethnicity” variable. Counties now have the option of asking a separate question that collects information on population subgroups not listed in the first two questions. Counties can configure their own code list. Youth may choose from a listing of county-specified selections or write-in response on a data collection form. RACIAL CODING GUIDELINES 1. Self-identification is the preferred method for collecting racial data, best accomplished by an in-person interview with the youth. In reality, however, racial data are often based upon an intake officer’s review of a report submitted by the referral source, especially in cases involving minor offenses. The point at which delinquency disposition data collection begins is at referral to intake based on a complaint received from an arresting or other justice officer, school official, or child welfare agency. Intake decision makers have a variety of options for resolving minor complaints, including warn and dismiss, referral to another agency, or hold in abeyance and data collection stops there. Although the preferred intake practice—even in minor cases—entails some investigation into the facts behind the complaint, time is limited, of course, and so are resources. The reality is that in minor cases, intake decisions are often made on the basis of a review of an arrest report or complaint rather than an interview with the youth. In situations when it is not feasible to interview the youth in person and the intake officer makes a decision based upon a review of the complaint, the officer should code Hispanic/Latino origin and race based upon what was reported by National Center for Juvenile Justice OCTOBER 2006 the referral source. The question that asks whether the answer was self-reported should be answered “no.” If the referral source did not provide racial information, the intake officer/coder may use the “unknown” category for either question. If the case is accepted for any kind of service by the court, every attempt should be made to correct errors in racial coding that may have occurred when the record was created. JCJC has quality assurance reports that return to the counties for correction any petitioned cases with the unknown codes selected. 2. Interviewers should rely on the youth’s self-identification when coding racial data. However, in situations when the youth fails to respond to either of the first two questions, the interviewer should infer Hispanic/ Latino origin and race. This decision was reached because the goal for the coding activity is to be able to document as accurately as possible the racial characteristics of youth involved in the juvenile justice system. Not answering the questions erodes the reliability of the measure when it is used to describe race and limits its utility for research into the overrepresentation of minority groups in the juvenile justice system. The risk of miscoding an individual juvenile is overshadowed by the desire to describe, monitor and report this information in the aggregate. 3. If the youth does not answer the Hispanic/Latino question, the interviewer may repeat the question and response options. If the youth still fails to respond to the question, the interviewer must infer a response (based upon observation or information provided by another source). In instances where the interviewer infers a response, the question asking the coder whether the answer was self-reported by the juvenile/parent/ guardian should be marked “no” (N). 4. If the youth has difficulty answering the race question, interviewers should encourage the youth to select a response that falls within one of the 5 race categories. Interviewers may experience difficulty with youth who identify as Hispanic or Latino in the first question, but who are unable to answer the subsequent question regarding their race. In these instances, the interviewer should simply repeat the 5 race categories. Interviewers should not ask prompting questions such as “In addition to being Hispanic, can you describe yourself as [repeat race categories]?” or “Hispanic or Latino is generally considered an ethnicity rather than a race. Hispanic or Latino persons can be of any race.” Such questions have been found to be offensive to some people and ineffective. If the question is confusing to youth or they refuse to answer the question, apply the next guideline. PAGE 09 FYI: This problem was well documented in the testing of the 2-question format in the 2000 Census where many respondents who answered “yes” to the Hispanic/ Latino question did not respond to the race question or indicated “other race.” 5. If the youth is unable or unwilling to select a race category, the interviewer must infer the youth’s race (based upon observation or information provided by another source). In instances where the interviewer infers a response, the question asking the coder whether the answer was selfreported by the juvenile/parent/guardian should be marked “no” (N). 6. If the youth does not respond to the third question, interviewers should not infer an answer. Categories and Definitions (source: Federal Register Vol. 62, No. 210, Thursday, October 30, 1997.) • Hispanic or Latino: A person of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin, regardless of race. • American Indian or Alaska Native: A person having origins in any of the original peoples of North and South America (including Central America), and who maintains tribal affiliation or community attachment. • Asian: A person having origins in any of the original peoples of the Far East, Southeast Asia, or the Indian subcontinent including, for example, Cambodia, China, India, Japan, Korea, Malaysia, Pakistan, the Philippine Islands, Thailand, and Vietnam. • Black or African American: A person having origins in any of the black racial groups of Africa. Terms such as “Haitian” or “Negro” can be used in addition to “Black or African American.” • Native Hawaiian or Other Pacific Islander: A person having origins in any of the original peoples of Hawaii, Guam, Samoa, or other Pacific Islands. • White: A person having origins in any of the original peoples of Europe, the Middle East, or North Africa. National Center for Juvenile Justice OCTOBER 2006 Federal Standards for the Classification of Racial Data The background of the 1997 revisions to the Federal OMB standards and the principles that governed the review process provide an important backdrop to the instructions and guidelines presented in this booklet. Background: For more than 20 years, the standards provided a common language to promote uniformity and comparability for data on race and ethnicity for the specified population groups. They were developed in cooperation with Federal agencies to provide consistent data on race and ethnicity throughout the Federal Government. Development of the data standards stemmed in large measure from new responsibilities to enforce civil rights laws. Data were needed to monitor equal access in housing, education, employment, and other areas, for populations that historically had experienced discrimination and differential treatment because of their race or ethnicity. The standards are used not only in the decennial census (which provides the data for the “denominator” for many measures), but also in household surveys, on administrative forms (e.g., school registration and mortgage lending applications), and in medical and other research. The categories represent a social-political construct designed for collecting data on the race and ethnicity of broad population groups in this country, and are not anthropologically or scientifically based. Some of the more relevant principles that governed the review process include: 1. The racial and ethnic categories should not be interpreted as being primarily biological or genetic in reference. Race and ethnicity may be thought of in terms of social and cultural characteristics as well as ancestry. 2. Respect for individual dignity should guide the processes and methods for collecting data; ideally respondent self-identification should be facilitated to the greatest extent possible, recognizing that in some data collection systems observer identification is more practical. 3. To the extent practicable, the concepts and terminology should reflect clear and generally understood definitions that can achieve broad public acceptance.... 4. The categories should be comprehensive in coverage and produce compatible, nonduplicative, exchangeable data across Federal agencies. 5. Foremost consideration should be given to data aggregations by race and ethnicity that are useful for statistical analysis and program administration and assessment.... 6. The standards should be developed to meet, at a minimum, Federal legislative and programmatic requirements. Consideration should also be given to needs at the State and local government levels....as well as to general societal needs for these data. PAGE 11 7. The categories should set forth a minimum standard; additional categories should be permitted provided they can be aggregated to the standard categories. The number of standard categories should be kept to a manageable size, determined by statistical concerns and data needs. The principle objective of the review was “to enhance the accuracy of the demographic information collected by the Federal Government by having categories for data on race and ethnicity that will enable the capture of information about the increasing diversity of our Nation’s population while at the same time respecting each individual’s dignity.” Source: Federal Register (Thursday October 30, 1997 (page 58781) Part II Office of Management and Budget Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity; Notices). PLEASE NOTE: “Racial coding” and “racial data” are used interchangeably throughout this booklet to refer to the set of questions aimed at distinguishing a juvenile’s Hispanic/Latino origin, race, and identification with any other population group or subgroups. 2 See “Juvenile Justice in Pennsylvania: Mission-Driven, Performance-Based, Outcome-Focused” available from JCJC at http://www.jcjc.state.pa.us/jcjc/lib/jcjc/barj/monograph.pdf . 3 The original amendment referred to Disproportionate Minority Confinement but the mandate was subsequently expanded to any Contact from arrest through confinement. 4 Beginning with the 1997 data, JCJC required probation departments to disaggregate data on race and “ethnicity” of youth. 5 For more information about the DMC Subcommittee and its collaboration with Models for Change, please see the Juvenile Justice and Delinquency Prevention Committee’s 2006 Plan Update 6 See Background Brief for April 25, 2006 Meeting to Address Issues Surrounding Application of Federal Minimum Race Categories to Juvenile Court Data, Hurst, Jr. H. and Torbet, P. NCJJ. 7 For example, a youth may prefer to indicate “unknown” in conjunction with the selection of another race when the race of the biological parent is not known. 1 National Center for Juvenile Justice OCTOBER 2006 53% gram 333w g3