Homicide Investigation Tracking System (800) 345-2793 CASE MANAGEMENT for Missing Children Homicide Investigation May 2006 Rob McKenna Attorney General of Washington US. Department of Justice Of?ce of Juvenile Justice and Delinquency Prevention Katherine M. Brown Robert D. Keppel Joseph G. Weis Marvin E. Skeen INVESTIGATIVE CASE MANAGEMENT FOR MISSING CHILDREN HOMICIDES: REPORT II ROB MCKENNA ATTORNEY GENERAL OF WASHINGTON US. DEPARTMENT OF JUSTICE OFFICE OF JUVENILE JUSTICE AND DELINQUENCY PREVENTION 2006 Cooperative Agreement 98-MC-CX-0001 Katherine M. Brown,1 Robert D. Keppel,2 Joseph G. Weis,3 Marvin E. Skeen4 1 Sam Houston State University, Huntsville, Texas, USA 2 Seattle University, Seattle, Washington, USA 3 University of Washington, Seattle, Washington, USA 4 Washington State Atton?iejyr General?s Of?ce, Seattle. Washington, USA The opinions, ?ndings, and conclusions or recommendations expressed in this publicatlon are those of the authors and do not necessarily re?ect the Views of the Department of Justice. This was supported by Grant No. awarded by the grantor agency? 5 Of?ce of Juvenile Justice and Delinquency Prevention, Of?ce of Justice Programs . ACKNOWLEDGMENTS The authors wish to recognize and thank John Scott Blonien, Division Chief of the Criminal Justice Division, Washington State Attorney General?s Of?ce, and the members of the Homicide Investigation Tracking System of the State Attorney General?s Office for their assistance in this research: Chief Criminal Investigators John Turner, Lockheed Reader, and Darryl Roosendaal; Investigators George Fox, Robert Gebo, Kenneth Han?and, James Hansen, Robert LaMoria, Darrell Noble, Richard Steiner, Frank Tennison, and Gary Trent; and staff Nicole Siver and Victoria Woods. In particular, the help of hivestigator/Analyst Tamara Matheny was invaluable. We also want to thank the individuals who provided data input: Carolyn Brooks, Inge A. Fordham, Janusz Springer and Allan Womac. We are grate?Jl to Kc L. Wendler Deaver for her assistance in data result veri?cation. The followmg organizatlons were instrumental in the primary? data collection for this project: It The National Center for Missing and Exploited Children (NCMEC). The O?ice of Juvenile Justice and Delinquency Prevention (OJJDP). Project Alert and Project Alert Volunteer John Hamilton. We want to acknowledge Governor Christine Gre go ire, who was the Attorney General of the state of Washington at the inception of this grant project. We also wish to thank the detectives who investigated the murders of abducted or missing children and the agencies that provided data. TABLE OF CONTENTS LIST OF TABLES vi CHAPTERS I INTRODUCTION Missing and Abducted Children De?nitions 3 II METHODOLOGY 5 Child Abduction Murder Data 5 Case Criteria ..6 Data Collection Data. Integrity .. 3 Resulting De?nitions 8 Purpose of the 10 Data Analyse ..11 VICTIMS 12 Introduction 12 Initial Police Involvement When an Abducted Child is Killed 13 Victims Overview 14 Victim?s Race and Gender 15 IV TABLE OF CONTENTS (Continued) Victim?s Age __16 Lifestyle of Victims __17 KILLERS "13 Introduction .18 Killer Attributes .13 Killer?s Age ""13 Killer?s Race ._19 Killer?s Gender __20 Killer?s Marital Status .20 Killer?s Residential Status .21 Killer?s Employment and Occupation Status __22 Killer?s Lifestyles "22 Killer?s Past Behavror .23 Killer?s Prior Crimes .24 Custody Status 25 VICTIM-KILLER RELATIONSHIP .26 Introduction 26 Victim- Killer Relationship 26 Victim- Killer Relationship by Age and Gender __27 ii TABLE OF CONTENTS (Continued) VI KILLER MOTIVATION AND VICTIM SELECTION __30 Introduction __30 Sexual Motivation 3 1 Pornography .32 Crises and Stressors "32 Choosing and Controlling the Victim Binding of Victim __36 Cause of Death __37 Unusual Acts __38 Body DisposalfRecover}r __38 Post Offense Behavior __40 VII INVESTIGATION __43 Introduction __43 Police Re?tsal to Accept a Missing or Runaway Child Report __43 Multiple Police Agencies 43 Unknowing Witnesses __44 Neighborhood Canvass __44 Area Search 46 Who Discovered the Body of the Victim __47 Investigative Steps in the First 48 Hours __48 Police Contact with Killer 49 Physical Evidence __50 TABLE OF CONTENTS (Continued) Discarded Evrdence 51 Polygraph __52 News Media __53 Outside Forces That Created Problems for the Investigators "54 Red Herring-?3 .55 MURDER INCIDENT SITES __56 Introduction __56 Components of the Murder Incident __56 Separation of Components by Time and Distance "57 Investigative Implications __58 Separation of Components in Child Abduction Murders __59 When Time is Known __59 When Place is Known __60 When Time or Place is Known __61 When Both Time and Place are Known __61 Time Interval Between Incidents 62 Distance Between Sites __67 Analysis of Distance by Murder Incident Component __72 Victim Last Seen Site 72 Initial Contact Site __72 Murder Site __74 Body Recovery Site 74 iv TABLE OF CONTENTS (Continued) IX SERIES CASES -76 Introduction .76 Comparison of Series and Non-Series Cases __76 Victims? Gender 77 VictiIn-Killer Relationship .77 Killers? Persona] Problems __73 Prior Crimes Against Linking Cases __79 SUMMARY AND CONCLUSIONS ,,81 Introduction ..3 1 How Can We Protect Our Children? .33 REFERENCES .35 PROJECT STAFF 37 TABLE LIST OF TABLES PAGE CAM Data Set __21 Percentage of Missing Children Reports within Time Periods __23 When an Abducted Child 1s Killed 24 Victim Race __25 Victim Gender __25 Distribution of Victims by Age Groups 26 Distribution of Killers by Age Groups __29 Killer Race __30 Marital Status of Killer 31 Living Arrangements of Killer "31 Killer Occupations __32 Perceived Lifestyle of Killer __33 Personal Problems of the Killer ?34 Prior Crimes Against Children by Killer __34 Victim-Killer __36 Female Victim-Killer Relationship, By Age Group __37 Male Victim?Killer Relationship, By Age Group __33 Victim-Killer Relationship by Victim Gender 38 Modus Operandi (MO) Similarities Between Other Crimes and Child Abduction Murder __41 How Killer Chose Victim 43 vi LIST OF TABLES (Continued) How Killer Initiated Contact with Victim Area of Initial Victim- Killer Contact Area of Initial Victim-Killer Contact by Type of Murder Why Killer Was at the Victim Contact Site How Killer Gained Control of Victim Cause of Death Cause of Death by Type of Murder Area of Body Recovery Area of Body Recovery by Type of Murder 45 45 47 48 49 49 How Killer Selected Body Disposal Site .. How Killer Disposed of Body Post O??ense Behavior of Who Discovered Victlm?s Body Major Investigative Steps in the First 48 Hours Physical Evidence Related to Relationship of Persons Polygraphed to Victim .. Effect of Media Coverage on the Investigation Problems Caused by Outside Forces . Murder Incident Component Time Known Murder Incident Compon?nt Place Known .. Murder Incident Component Time or Place Known Murder Incident Component Time and Place Known vii 72 LIST OF TABLES (Continued) Time Span Between Victim Last Seen and Initial Contact .73 Time Span Between Victim Last Seen and Murder __74 Time Span Between Victim Last Seen and Body Recovery 75 Time Span Between Initial Contact and Murder __75 Time Span Between Initial Contact and Body Recovery __76 Time Span Between Murder and Body Recovery 77 Distance Between Victim Last Seen Site and Initial Contact Site __78 Distance Between Victim Last Seen Site and Murder Site __78 Distance Between Victim Last Seen Site and Body Recovery Site __79 Distance Between Initial Contact Site and Murder Site .79 Distance Between Init 1a] Contact Site and Body Recovery Site "80 Distance Between Murder Site and Body Recovery Distance Between Murder Incident Component Sites __31 Victim Gender by Series and Non-Series Cases __87 Victim-Killer Relationship by Series and Non-Series Personal Problems of Killers by Series and Non- Series Cases __83 EXECUTIVE SUMMARY Cases of abducted and missing children found murdered continue to traumatize victims? families and friends and the public. While these cases are still quite rare, with the amount of national and international media attention they are given, we too quickly recognize the victims" names. Our ?rst research into this subject was ?undertaken in an effort to better understand these types of murders and to identify investigative techniques and strategies that will improve the efficiency and effectiveness of the criminal invest1gations and the apprehension rate of murderers who abduct children.? This follow-up research continues that effort. We collected 175 solved cases and added them to the original data set. We obtained the new cases from agencies of all sizes in the United States and even went to some states not included in the original survey. The additlonal cases are re?ective and supportive of the original report with several signi?cant and de?nite differences between the ?rst study and this one. With more killers identi?ed, we ?nd that the relationships between them and their victims change from strangers to an almost equal likelihood of being friends/acquaintances. The probability of the killer?s name coming up during the ?rst week of the investigation decreases Another Signi?cant change is the increase of the use of poniography by killers as a trigger. This should not be surprising, given the overwhelming sexual motivation of killers in these cases. ix contmue to be females over ll years of age, leading normal lives and living with normal families typical low?risk victims. The initial contact between the victim and killer is still within 1/6: mile of the victim? residence. The killers remain around 27 years old and are predominately unmarried, with half of them living either alone or with their parents. Half of them are unemployed, and those who are employed work 1n unskilled or semi?skilled occupations. More than half of these cases are initially reported to a law enforcement agency as a ?mlssing child.? Fast action is necessary smce there is typically over a two-hour delay in making the initial missing child report, and the vast majority of the abducted children who are murdered are dead within three hours of the abduction. One question answered by this research is this. What can we tell parents to help them protect their children? Even though child abduction murders are rare, parents must eliminate, or minimize, the opportunity for their children to become victims. We need to tell parents that if their child is unaccounted for, call the police immediately. CHAPTER I INTRODUCTION Child abduction murder is every parent?s worst nightmare. To compound the problem, child abduction murders are incredibly difficult to solve and deeply impact lavtr enforcement of?cials involved in the investigation. According to the National Center for Missing and Exploited Children (NCMEC), abductions resulting in a child?s death present many investigative and emotional obstacles for law enforcement officers (National Center for Missing Exploited Children, 2000). Han?and, Keppel, and Weis (1997) state that the rarity of child abduction murders, even among criminal homicides, and their complex, emotion-laden, high pro?les, make them extremely difficult to investigate. The 1979 abduction of seven-year?old Etan Pat: and the 1981 abduction and murder of six-year-old Adam Walsh terri?ed parents throughout the nation. As a result of the notoriety of these cases, inflated and unsubstantiated numbers of missing children were widely reported to be over 2 mil lion per year. The number of children who were abducted and then murdered was erroneously reported to be as high as 5,000 per year (National Center for Missing Exploited Children, 2000). The public outcry over concem for the safety of America?s children contributed to the establishment of the National Center for Missing and Exploited Children by the United States Congress. To the contrary, care?ll research has shown that between 40 and 150 incidents of child abduction murder occur each year, which is less than one half of one percent of the murders committed nationally. On average, there is one child abduction murder for every 10,000 reports ofa nussmg child (Finklehor, Hotaling, Sedlak,1992). Missing and Abducted Chiidren Statistics MISS ing children cases are typically placed Into ?ve categories: family abduction, nonfamily abduction, runaways, throwaway or abandoned children, and children who become lost or injured (Baker, Burgess, Rabun, Nahirny, 2002). In 1988, an estimated 354,100 family abductions occurred. Forty-sot percent of those abductions mvolved concealment of a child, out of state transportation, or the intent to keep the child indefinitely or to permanently alter custody. During this period, an estimated 3,200 to 4,600 nonfamily abductions were known to police (Finklehor et al., 1992). Unfortunately, less than 5% of those types of cases are reported to police (Han?and, et al., 1997), and they are more likely to result in harm or death. Appro ximate [y 200 to 300 of them were stereotypical kidnappings in which the child was transported a distance of more than 50 feet or kept over night. An estimated 446,700 children ran away during this period, and approximate 1y 127,100 children were told to leave their home or were abandoned (Finklehor et al., 1992). Unfortunately, it is extremely difficult to gather accurate information on child abduction for a variety of reasons. Existing studies on child abduction are limited in scope, and there are discrepancies in the statistical information produced. Boudreaux, Lord, and Etter (2000) propose an explanation for the de?ciencies in data on child abduction: First, a number of highly publicized stranger abduction cases in the early 19805 resulted in a social climate of heightened concern and emotion regarding the safety of children. This yielded overestimated initial incidence rates of child abduction (Best Thibodeau, 1997', Finklehor et a1, 1990). Second, early statistics combined many different types of child abduction family and nonfamily abductions) and age groups of children preteen and teenaged children), impeding researchers? abilities to identify speci?c critical issues and draw clear conclusions regarding the dynamics of child abduction. Third, many child abductions are not reported to law enforcement agencies such as the Federal Bureau of Investigation (FBI) or child assistance agencies such as the National Center for Missing Exploited Children (NCMEC). Finally, data collection has been hindered by variations in state laws de?nitions) and the use of different data collection methods. Locating child abduction case ?les within law enforcement agenc1es can be particularly cumbersome because abductions may be ?led under other crime categories homicide and sexual assault). 64) De?nitions There is no consistent, standard social scienti?c de?nition of child abduction. According to Pass (1999, p. 9), ?our de?nition of the crime (even, indeed, the courts? de?nition) has been historically derived.? According to Boudreaux et al. (2000), a single de?nition is problematic for two reasons: there is no single accepted de?nition of child abduction the legal de?nitions of the terms child and abduction can vary ?om jurisdiction to jurisdiction), and there are different forms of child abduction, each with inconsistent terms, which if not specifically de?ned, can cause con?lsion g. parental abduction, family abduction, stereotypical abduction, missing children, and kidnapped children). 64) In addition, legal de?nitions of child abduction vary wrdely across Jurisdictions. Finklehor et al. (1992) propose that the de?nition of child abduction should be ?the coerced unauthorized movement of a child, the detention of a child, or the luring of a child for the purposes of committing another crime" 228). The National Incidence Studies of Missing, Abducted, Runaway, and Thrownaway Children (NISMART) (2002) de?nes stereotypical kidnapping as: nonfamily abduction perpetrated by a slight acquaintance or stranger in which a child is detained overnight, transported at least 50 miles, held for ransom or abducted with intent to keep the child permanently, or kille 2) and nonfamily abduction as: (1) An episode in which a nonfamily perpetrator takes a child by the use of physical force or threat of bodily harm or detains the child for a substantial period of time (at least 1 hour) in an isolated place by the use of physical force or threat of bodily harm without lawful authority or parental permission, or (2) an episode in which a child younger than 15 or mentally incompetent, and without law?il authority or parental permission, is taken, detained or voluntarily accompanies a nonfami 1y perpetrator who conceals the child?s whereabouts, demands ransom, or expresses the intention to keep the child permanently. (P- 2) There has been disagreement over the amount of distance required to transport a child in order to establish an abduction (Finklehor et al., 1992). According to Forst and Blomquist (1991), some jurisdictions follow the ?any movement? rule, while some jurisdictions abide by the ?incidental rule? in which movement in the context of another criIne is incidental. To add to the con?lsion, some de?nitions depend on the motivation of the killer, and there is even disagreement about the de?nition of child (Lanning, 1995). The confusion and ambiguity regarding defm itlons contribute signi?cantly to dif?culties In collecting rigorous data on child abduction murder. However, those obstacles are not insurmountable. Summary This report discusses the ?ndings from a research project which examined the investigations of more than 800 child abduction murders. The report is particularly salient to homicide detectives who are confronted with an unsolved murder case involving child abduction. Results from this study will help police investigators identify strategies and implement tactics to focus invest igatlons that will improve their ability to solve child abduct1on murder cases. CHAPTER II METHODOLOGY Introduction The data used in this study were collected through a cooperative agreement between the Washington State Attorney General?s Of?ce and the Office of Juvenile Justice and Delinquency Prevention, United States Department of Justice. The objective of the collaborative research project was to examine murders of abducted children. The data set will be referred to as the Child Abduction Murder (CAM) data set. This report, the second of two, updates the data and ?ndings of a prior three-year research project that examined the investigations of murders of more than 600 abducted children. The ?ndings of the previous study are referred to in this study as the ?rst-stage find ings and were originally published as Investigative Case Managemenrfor Missing Children Homicides (Hanfland at 3.1., 1997). The ?rst-stage study, as well as the second-stage data collection and were conducted by criminal justice professionals with extensive murder investigation and research backgrounds. Child Abduction Murder Data The CAM data were collected from 227 municipal police department and county sheriff" of?ces in the United States with a service population of 100,000 or more, or that had ?fteen or more murders reported to the Federal Bureau of Investigation?s (FBI) Uniform Crime Report (UCR) in 1987 (Han?and et al., 1997). The department head, detective division commander, or a detective of each of the agencies meeting the above criteria was contacted by telephone and asked to participate in the data collection project. To follow up, an introductory letter and a formal request for case information was mailed to identified agencies, with a resulting initial response rate of 75% (Han?and et al., 1997). An additional teletype was sent to each police agency in the United States at three different time intervals on different days of the week, with a request to contact the coordinator of the project. Basic screening information was then recorded about the cases. Additional methods were employed to identify relevant cases for the study. Appropriate state and federal agencies were contacted and asked for case Information. Homicide detectives across the country were also contacted additional information (Han?and et al., 1997). Initially, data were collected by interviews with the detectives and the review of investigative case ?les ?om 1,025 cases. Responses were received from both large and small agencies and departments. The agencies were representative of all regions of the country, covering 44 states. Of the original 1,025 cases, 621 were found to meet the criteria established for inclusion in the original data collection collaboration (Hanf'land et a1., 1997). The second stage of the research, described in this report, was designed to expand the original CAM data set. Over a number of months, more than 200 additional cases have been collected, resultmg in 333 child abduction murder cases from 1968 to 2002. The additional cases came from municipal police departments and county sheri?" of?ces of all sizes. Case Criteria The cases of murder in the original CAM data set were chosen for 1nclusron based on the following criteria. 1 The victim was younger than 18 years old (except as described in #3 below), whose body had been recovered; or if the body had not been recovered, the killer was identi?ed, tried, and convicted. 2. The police agency receiving the initial contact about the case, whether as a missing, abducted, runaway, or dead body case, acted on the premise that abduction was a possibility. 3. The case was part of a series in which at least one victim in the series met the above stated criteria (Han?and et al., 1997, p. 14). Additionally, cases were included that were not considered ?closed? in the traditional sense. If the reporting agency believed that abduction was a possibility, and began investigating the case as a child abduction, it was included in the data set. The murders of abducted children to be examined in the second-stage research were selected from the CAM data set based on the following criteria: the victim was 17 years old or wunger, whose body had been recovered; or if the body had not been recovered, the killer was identi?ed, tried, and convicted. A total of 735 cases met all of the criteria and were the focus of analysis. Data Collocation Instrument The CAM data collection instrument was designed by homicide investigators Kenneth Han?and and Dr. Robert D. Keppel, and criminologist Dr. Joseph G. Weis. It was used to collect information on 412 items representing the essential characteristics of the murder of an abducted child and of its investigation (See Han?and et al., 1997). The instrument was designed to evaluate the criminal investigation process, including the initial response of the police agency, basic investigation, extended investigation, physical evidence, geographical considerations, and victim and killer information. It was ?eld tested on ten cases of child abduction resulting in murder, after which a few changes were made. Data Integrity After the data collection instrument was received from participating agencies, it was reviewed for validity and internal consistency. The data were then entered into a database designed speci?cally for the project. To ensure data integrity, a printout was generated from each new computer record, and was Visually compared with the original data collection instrument. Any necessary corrections were made by the data entry cleric and the resulting printout and the original data collection instruments were given to the project coordinator for an additional review for errors. Any additional errors were corrected, and the form was placed in a ?le (Han?and et al., 1997). Resulting Dam The data set used for this research was collected with the purpose of determining proper and effective avenues of investigation in missing and abducted children cases. In order to determine which factors were e?ective investigation tools, information on solved cases was collected for this report. Information on unsolved cases was also collected to examine the differences between solved and unsolved cases. Thirty-?ve percent of the cases of child murder from the original CAM data set were unsolved at the time the data were collected. The original CAM data set contained 577 case investigations with a total of 621 victims (some cases had multiple victims) and 419 killers. Of the 735 cases analyzed for this report, 27. 4% were unsolved at the time of data collection. De?nitions Defining the terms used in this research project was critical. In order to select appropriate variables for analysis in our study, certain terminology was de?ned: abduction, components of the murder incident, time and distance intervals, and solved cases. Once the terminology was de?ned and the proper parameters set, appropriate variables were selected for statistical analysis. Abr?rc?on. For purposes ofthis research, abduction was de?ned as. 1. 2. The victim was kidnapped. The was detained and hisf her freedom of movement was restricted. A victim ofdomestic violence was reported by the family (or someone else) as a missing child. The police were initially of the opinion that the victim was taken or held against his or her will, whether or not that turned out to be the case in the end (Han?and et al., 1997). Components of?re Murder Incident. Each of these components occurs in a murder, and therefore each is important to a murder investigation. The components are de?ned as: 1. The Victim Last Seen Site (VLS) was defmed as the location where and time when the victim was last seen. The VLS was determined from eyewitness information and records indicating when and where the victim was last seen alive. The Initial Contact Site (10) was de?ned as the place where and time when the killer initially contacted the victim. The IC was established ?'om evidence indicating that the killer ?rst met the at a certain time and at a speci?c location during the course of the murder incident. The Murder Site (MS) was de?ned as the place where and time when the victim sustained the death-producing injuries. The Body Recovery Site (BR) was defined as the location where and time when police, medics, or witnesses found the victim, dead or alive, prior to transportation to a medical facility or morgue (Keppel, 1992; Koppel 3c Weis, 1993a). Solvabilily. In the CAM data set, solvability was de?ned and measured two ways. ?Has the killer been arrested, or does probable cause exist for an arrest?? and ?Has the investigation resulted in a conviction?? For the analysis here, solvability was based on investigations resulting in an arrest. Cases with a ?Yes? to the question ?Has the killer been arrested, or does probable cause ex1st for an arrest?? at the time of coding were consrdered ?solved?; cases with a ?No? to that question were considered ?Unsolved.? Cases with ?Unknown? as the answer to the question were considered to have missing data and were not included in the analysis. Purpose ofthe Study According to Han?and et al. (1997), the rarity of these types of cases has allowed a body of ?commonly held beliefs? to develop that has little to no basis in fact. Hence detectives, case managers, police executives, and the media sometimes operate with false assumptions and misperceptlons. Homic1de Investigators, through no fault of their own, sometimes fail to realize that the investigations of the murders of abducted children are very different from the other murders they usually investigate. Consequently, they sometimes make decisions about the direction of the investigation that are not ?high percentage? choices. For example, some detectives believe that in any murder of a child the logical suspect is a parent and, therefore, they devote a considerable amount of resources to prove that belief. But this research shows that the parents are the least likely suspects in an abduction murder of a child. This kind of false assumption is made, in part, from lack of experience with these types of cases and because there is very little empirical research on these types of child murders and their investigations from which detectives can draw guidance. This research will help investigators make those decisions, identify the strategies, and implement the tactics that will lead to the more certain and timely capture of the killers of abducted children. 10 Data Analysis Throughout the report data are presented as Simple percentages. Often comparisons are made with ?all murder cases? and, at times, with ?all child murders,? which include domestic child abuse murders, mutual combat murders, and others that do not involve abduction. Percentages are used to prov1de a sense of how often or how rarely a characteristic or circumstance can be expected in these types of cases. The logic is that knowing the ?spread? will help the detective make better decisions and prioritize courses of action. In most cases, simple descriptive frequencies were run on appropriate variables. Where the impact of a variable on solvability was examined, a cross-tabulation was performed to test statistical signi?cance. The CAM data set utilized for the second stage research consists of 833 child abduction murders. Of those cases, 735 victims are 17 years old or younger, with 516 of those victims being single-victim murder cases rather than part of a series. Of the 735 cases analyzed, 27.4% remained unsolved at the time of data collection. Table 1 summarizes the characteristics of the CAM data set that was used as the basis for the data analysis discussed in this report. Table 1 CAM Data Set All Cases in CAM Data set Victims <12-Years- ld Solved 589 74.1 527 72.6 Unsolved 206 25.9 199 27.4 DiSposition 38 4.6 9 1.2 Unknown Total 833 735 11 CHAPTER VICTIMS Introduction The existing research on the rarity of child abduction murders, particularly those by non-family killers, shows clearly that most law enforcement jurrsdict1ons 1n the United States will not be called on to investigate a child abduction murder. In fact, most homicide investigators will never investigate a child abduction murder over their entire career. However, it is prudent for investigators to be prepared. These investigations can put enormous strain on even the best prepared detectives, investigations, and jurisdictions. Their typically emotion-laden, volatile, high-pro?le characteristics present unique challenges to law enforcement. Initial Police Involvement Police involvement in the child abduction murder investigations in this study began with the report and identi?cation of the victim as a ?missing child? in 60.2% of the cases. The case investigation was initiated by the recovery of a dead body in 20.3% of the cases. Only 8.5% of the investigations began as reports of a runaway child, and 9.7% as abduction investigations. As illustrated in Table 2, most reports to the police of a missing, runaway, or abducted child were made relatively soon after the child was missed. 0f the children reported missing, 19.6% of the victims were reported missing ?immediately,? 42.9% within two hours, and 86.6% within 24 hours of being missed. Two hours lapsed before reports were made in 56.2% of the cases, and at least a whole day passed in 12.5% of the cases. Unfortunately, 1.0% of cases never had a missing person report ?led. 12 Table 2 Percent of Missing Children Reports within Time Periods Case Reported to Police Immediater 19.6 Within 1 Hour 27.5 Within 2 Hours 42.9 Within 4.5 Hours 68.4 Within 24 Hours 86.6 24 Hours 99.0 These reporting delays are important to the course oi?the investigation. The data show that delays are much more critical in child abduction murders than in other types of investigations, because missing children who are murdered are killed quickly after their abduction. When an Abducted Child is Killed Missing children were killed within a very short period of time after their abduction. Table 3 illustrates that 46.8% of the victims were dead within one hour after the abduction, 76.2% within three hours, and 88.5 within 24 hours after being abducted. There is a misconception that killers keep children alive for long periods of time after an abduction. Clearly, this misconception is not supported by the data. Hours and even minutes are critical in an abducted child investigation. While the murder of an abducted child is rare, immediate reporting of the abduction and swift investigative measures may ensure that the child is recovered alive. 13 Table 3 When an Abducted hi Id is Kiiled Time Period <1 1 Hour 46.8 Withhl 3 Hours 76.2 Withm 24 Hours 88. 5 Within 7 Days 97.9 Within 30 Days 100.0 The dictum that the ?rst 24 to 48 hours of an investigation are the most critical must be modi?ed in child abduction cases. An absent child should be reported to authorities immediately. Police should concentrate their investigative resources as quickly as possible once a child is reported missing. Swift action by law enforcement will increase the odds that a child is found alive; and if the child is murdered, it will increase the probability that the killer is caught. In addition to information about the time lapse between when a child is missing and when the absence is reported to police, this research identi?ed unique characteristics of child abduction murder victims. This information will assist investigators in identifying and differentiating potential murder victims from the vast majority of missing children who eventually turn up alive and well. This information will help investigators prioritize investigative resources. Victims Overview The typical child abduction murder victim was a white female approxrmately 11 years old They were predominantly from a middle class or ?blue collar? family, living in an urban or suburban neighborhood, in a single-family residence The victim?s relationship with family was good (49. and the family situation was not considered high risk In summary, the typical child 14 abduction murder victim was an 1 1-year-old, white female from a middle class or ?blue collar? family with good family relationships. Victim Race and Gender As found 1n the previous ?rst-stage analyse, the race of child abduction murder VlCtu?l?lS is not signi?cantly different from all murder victims both are predominantly white. Table 4 illustrates this similarity. Table 4 Victim Race Child Abduction All Murders White 74.5% 66. 0% Black 14.3% 17. 0% All Other 11.2% 17.0% The gender of child abduction murder victims differs significantly from the gender of victims in all child murders and murders in general. The first-stage ?ndings indicated that child murder victims were more likely to be female while only 38.0% of murder victims in general are female. But the second-stage analysis indicated that child abduction murder victims were predominantly female Table 5 shows the differences at gender by type of murder. Table 5 Victim Gender Child Abduction Murders All Child Murders All Murders Female 74.0% 55.0% 38.0% Male 26.0% 45.0% 62.0% 15 Wolf?: ?3 Age According to Han?and et a1. (1997), approximately one-half of all cluldren who are murdered are between the ages of 15 and 17 years old. Among child abduction murder victims, only 22.2% of victims were older teenagers between the ages of 16 and 17 years old. Table 6 illustrates the age distribution of child abduction murder victims. The mean age of VlCtil?l?lS was 11.52 years. About 10% of the victims were children ?ve years old and younger, a group smaller than one might expect, based on the common perception that it is ?little kids? or ?young children? who are most vulnerable and selected most often as victims. Older children (6-9), preteens (10-12), young teens (13-15), and older teens (16-17) each constituted about 20% of the victims of child abduction murders. Only a slight majority of the victims in this sample were not yet teenagers at the time of their death. Table 6 Distribution of Victims by Age Groups Victim Age Group to 5 years old 10.1 6to9years old 21.5 10to 12 years old 20.7 13 to 15 years old 25.5 16 to 17 years old 22.2 16 Lifestyle of Victims There is also a public misconception that children who are killed durmg an abduction are particularly vulnerable or high-risk victims, but most of the victims were described by those who know them as ?normal kids.? On the other hand, appro ximately one-third of the victims were not considered average kids. Only 16.9% of the victims were described as street kids and 13% as runaways. As one might expect, those types of victims are typically teenagers rather than younger children. Community resources focused on runaway children and at-risk children who are prone to live on the street would be prudent preventive measures. Summary It is vitally important to identify which children are most at risk for child abduction murder. It is also imperative that the killer is examined and, hopefully, better understood. The next chapter discusses the characteristics of the typical child abduction murderer. l7 CHAPTER IV KILLERS Introduction Just as child abduction murder victims are unique, their killers are unique among murderers. Those who kill abducted children share many characteristics with other types of murderers, but they differ in important ways that suggest a different etiology to their predatory behavior. Their unique characteristics call for different investigative strategies. This chapter will focus on the personal and social attributes and behavior of child abduction killers. Killer Attributes In addition to their choice of victims?the substantial overselection of female victims child abduction killers are unique among murderers in general. Hanfland et a1. (1997) characterized child abductlon murderers as ?social margmals.? Child abduction killers are not: active, successful participants in mainstream, conventional social life, but, rather, they occupy a position in society that is, indeed, on the ?edge, brink, border, precipice, or margin.? They are not integrated, personally or socially, into the kinds of relationships or activities that produce and sustain effective self or social controls. Their personal and social attributes establish and de?ne their social marginality. (Han?and et a1., 1997) Killer?s Age The mean age of child abduction murderers is 27.8 years old. Table 7 shows the distribution of child abduction killers by age group. 13 Table 7 Distribution ofKiHers by Age Groups Killer Age Group 18 years old 12.2 18to 20 years old 21.2 21 to 25 years old 22.3 26 to 30 years old 22.3 31 to 40 years old 23.4 40 ars old 10.] Only 12.2% of the killers were Juveniles (under 18), and only 10.1% were over 40 years old. The great majority of killers were young adult men between 18 and 30 years old. The oldest killer in this sample was 61 years old, with the youngest killer being 9 years old. Contrary to the belief that these killers are ?dirty old men," the evidence shows that they are more similar in age to killers in general young adults. Killer?s Race The race of the killers in this sample was predominantly white (69. Only 19.1% were black and the remaining 11.0% were other moral/ethnic groups, when adj usted for the missing race identi?cation of 158 killers. The results on the race of killers supported the earlier ?rst-stage fmdings, that there were no important diiferences in race between child abduction killers and other killers. Table 8 shows the racial breakdown of child abduction murder killers compared to murderers in general. 19 Table 8 Killer-Race Child Abduction All Murders White 69. 3% 75 . 0% Black 19.1% 20. 0% All Other 11.0% 5.0% Killer?s Gender Murder is a predominantly male phenomenon, with the great majority of victims and killers being male. The ?rst-stage research observed that child abduction murderers were even more like 1y to be male 98.5% of child abduction murder killers were males, compared to 87% of murderers in general. By including the additional cases collected in stage two of the study, 96.1% of the killers were male, still a substantial overrepresentation of male killers. These murders were carried out almost exclusively by males, and were perpetrated typically against young females. There were also a number of indicators of social mmginalig) among these killers, including marital status, residential status, employment and occupation, lifestyle, problem behavior, and criminal behavior. Killer?s Marital Status Only 16.9% of the killers were married at the time they committed the murder, and another 13.2% were divorced. Accordingly, 83.0% of the killers had no intimate attachments or bonds with another person at the time of the abduction and subsequent murder. The marital status of the killers is illustrated in Table 9 20 Table 9 Marital Status ofKr'lIer Marital Status Single 69. 8 Married 16. 9 Divorced 13 . 2 Killer?s Residential Status Contrary to perceptions that killers who kill children are ?loners,? only 17.1% of the killers lived alone, while 74.8% lived with someone. It is perhaps a little unusual that 33.2% of them lived with their parents. Table 10 describes the killers? living arrangements. The killers in this sample also changed their residence frequently. The majority of them moved at least once within the ?ve years preceding the murder Of this group, 44. 5% of killers changed residences three or more times, and 19.7% moved ?ve or more times. Table 10 Living Arrangements ofKiHer Living Ajrangement Living with Parents 33.2 Living Alone 17.1 Living with Spouse and/or Children 17.1 Living with 14.2 Living with Other Roommates 10.3 Other 8.2 21 Killer?s Emptoyment and Occupation Shims This transient nature may be related to their uneven employment history. Approximately half of child abduction murderers were unemployed at the time of the murder; and if they were employed, they worked in unskilled or semi-skilled labor occupations. Table 11 describes the most common occupations of killers at the time they committed the murder. Table 11 Killer Occupations Occupation Construction Worker 26.4 Food Industry Worker 7.5 Service Industry 6.9 Student 6.6 Truck Driver 6.6 Auto Maintenance Kitter?s Lifest?es Only 10.1% of the killers were described as ?model? citizens, and a mere 3.9% were active in church or civic groups. In fact, 20.1% of the killers were on probation or parole for another offense at the time they committed the child abduction murder. Many of them were described as ?strange? by others who knew them, 14.7% were described as ?friendly to and 16.0% were cons1dered reclusive. Table 12 reveals other behaviors of the killers that are considered outside the norm by our society. A number of them abused alcohol (24. used and abused drugs or were sexually promiscuous according to others who 22 knew them and characterized their behaviors and identity before they were identi?ed as a child abduction killers. Table 12 Perceived Lifestyle of Killers Perceived Lifestyle Described as ?Strange? 30.2 Alcohol Abuser 24. 6 Drug User! Abuser 22. 1 Reclusive 16.0 Sexually Promiscuous 15.5 Friendly To Children 14.? Transient/Semi?transient 1.9 Killer ?3 Past Behavior Child abduction murderers experience more serious personal behavioral problems than the typical killer. Over three-fourths of them (7 had a recognized and identi?ed history of at least one serious behavioral problem. The child abduction murderers in this study differed signi?cantly from general murderers in the extent of their problems (Table 13). Almost one- third of the killers of abducted children suffered from ?sexual problems,? in contrast to an extremely low rate among killers in general murder The former were ten times more likely than the latter to have a history of identi?ed sexual problems. 23 Table 13 Personal Problems of the Killers Child Abduction All Murders Sexual Problems 30.4% 3.0% Alcohol Problems 23. 8% 27. 0% Drug Problems 22.3% 14.0% Mental Problems 13.4% 13.0% Killer?s Prior Crimes In addition to victim and killer demographic and behavioral characteristics, information was collected regarding the killers? crimmal background and motivation to commit murder. A large percent of the murderers had a substantial history of prior crimes against children Table 14 shows the types of crimes committed against children by these killers. Table 14 Prior Crimes Against Children By Killers Prior Crimes Sexual Assault (Non-Rape) 21.0 Rape (or attempt) 13.3 Murder (or attempt) 10.7 Kidnap 6.5 Assault 6.1 These killers clearly demonstrated violent predispositions toward children prior to the child abduction murder. Sexual assaults and rapes were the prior crimes most frequently committed against children. Incredibly, the murder of another child was committed previously by 10.7% of the killers in this study. An additional 6.5% had also preViously 24 kidnapped children. Sexual assault of the victim occurred in 62.3% of these child abduction murder cases those easily characterized as acts of violent, sexual predation on children. Custody Status In spite of the substantial evidence of prior crimes of violence against children, appro ximately half of child abduction killers, like murderers in general were not in any ?of?cial custody status? at the time of the extant murder. A higher percent of child abduction killers were either on parole or probation when they killed as compared to 17 . 0% of all murderers. While most child abduction killers may not have been immediately accessible in the custodial system, they were more likely than other murderers to be found in the active ?les of the correctional or judicial systems. Summary Child abduction killers can be characterized as social marginals who have a propensity to crimes of violence against children. Most of them exhibit weak social bonds to conventional contexts, relationships, and activities--strong predictors of involvement in crime. Individuals with weak social controls are more likely to commit child abduction murders given the appropriate motivation and opportunity. The victim-killer relationship discussed in the next chapter illustrates in greater detail the opportunities available to the killers. 25 CHAPTER RELATIONSHIP Introduction It is vitally important to understand the demographics of both the victim and killer in the study of the murder of abducted children. However, it may be more important to understand the victim-killer relationship in order to protect our children. The following chapter will present an overview of the relationship between the child abduction murder victims and their killers. Viaim-Ki?er Relationship Child abduction murderers were strangers to the victim in 44.4% of the cases, and were family friends or acquaintances in 41.9% of the cases. The killer was an intimate or family member of the victim in 13.7% of the cases (Table 15). Table 15 Victim-Killer Relationship Child Abduction Relationship Murders All Child Murders All Murders Stranger 4441.9% 28.0% 42.0% Familyz?lntimate 13.7% 67.0% 3 8.0% The killers were strangers to the victim in 44.4% of the cases, in contrast to 5.0% of all child murders and 20.0% of murders in general. Victims are much more likely to be killed by family members in child murders than in child abduction murders 26 Mains-Ki?er Relationship by Age and Gender The victim-killer relationship was also examined by age and gender. The victim-killer relationship varied by the age and gender of the victim. Table 16 shows the victim-killer relationship for females by age group. The youngest females (1 to 5 years old) were most likely killed by a friend or an acquaintance In contrast, older female victims (10 to 17 years old) were more likely to be killed by strangers. Table 16 ens-ale Victim-Killer Relationship, By Age Group Victim Age Group Relationship 1-5 6-9 10- 12 13-15 16-17 Stranger 15.7% 33.0% 41.6% 37.2% 40.2% Friendz? Acquaintance 4922.7% 17.6% 8.3% 11.9% 7.4% 14.4% Unknown 17.6% 19.3% 13.9% 27.0% 22.7% Table 17 shows the differences in the victim-killer relationships for males. The youngest victims (1 to 5 years old) were almost as likely to be killed by family members (34. as they were by strangers The youngest victims were least likely to be killed by friends or acquamtances. The older male victims (16 to 17 years old) were almost as likely to be killed by strangers as friends or acquaintances In this sample, family or intimates did not kill male teenagers, but only younger boys. 27 Table 17 Male Victim-KillerRelationship, By Age Group Victim Age Group Relationship 1-5 6-9 10-12 13-15 16-17 Stranger 30.4% 36.7% 35.3% 28.6% 41.9% FriendiAcquaintanee 26.1% 34.7% 39.2% 40. 0% 48.4% Familyz?lntimate 34. 8% 14.3% 7.8% 0% 0% Unknown 8.7% 14.3% 17.6% 31.4% 9.7% The overall victim-killer relationship by gender is shown in Table 18. Females were most often killed by strangers while males were most liker killed by friends or acquaintances (3 Both genders were least like 1y to be killed by a family member or intimate. Table 18 Victim-Killer Relationship by Wcrim Gender Victim Gender Relationship Female Male Stranger 35.9% 34.7% Friend! Acquaintanee 32. 0% 37. 9% 1.1% 10. 5% Unknown 21.1% 16.8% 28 Sulnmary The differences in the victim-killer relationship by age and gender of the victims can be more clearly seen 1when the different types of murders and modus operandi are examined in light of the research ?ndings in the following chapter. The next chapter will detail the killer?s motivation for abducting the child. In addit 1011, how the killer selected the VlCtiIIl will be examined. 29 CHAPTER VI KILLER MOTIVATION AND VICTIM SELECTION Introduction Surprisingly, over half (5 of the prior crimes committed by a killer against children had a modus operandi (MO) that was Similar to the extant murder. For example, a child abduction killer who lures the child away by asking for help locating a lost pet is very likely to lure away the next victim in a similar way. The similarities in M05 produced other surprises' they were most alike, by a large margin, in the ?commission of the offense,? or the way the criines were committed (Table 19). In the cases that had similar MOs, the priors were committed in similar ways for example, the choice of weapon was the same across different crimes committed by a killer. There is a misconception that child abduction killers are looking for a child with a certain appearance. Contrary to murderers in general, child abduction killers were much less likely to select certain types of victims based on personal characteristics. A mere 7.2% of prior crimes were committed against similar victims. The killers were even less likely to approach their victim in similar ways. Finally, 16.5% of the prior crimes against children were similar to the extant murder in the kinds of spec i?c acts that were performed during the commission of the crimes. For example, the killer used rope to bind and control a kidnap Victim, as well as the subsequent murder victim. These findings regarding the similarity of MOs across the majority of crimes committed by child abduction killers show that there is more consistency in the M03 of child abduction killers than expected, especially compared to other types of murderers. The data also suggest that there may be a greater predisposition to serial offending among child abduction killers. 30 Table 19 Modus Operandi M0) Similarities Between Other Crimes Murder MO Commission of the Crime 16.5 Victim Characteristics 7.2 Speci?c Acts Committed 5.7 Approach to Victim 5.2 Sexual Motivation Child abduction murderers share a characteristic with classw serial killers both show a sexual component to their motivation to kill. More than two-thirds of child abduction murders involved a sexual motive, compared to only 5% of all murders and 14% of child murders. Almost one?half of the child abduction murders were classi?ed as rapes and 21.9% as other sexual assaults. As one would expect in these types of murders, a large group were kidnappings, which is approximately 15 times greater than in all child murders and 30 times greater than in murders in general. Some of the kidnappings also included a secondary sexual component, because there was physical evidence that almost two-thirds of the child abduction murder victims had been sexually assaulted, compared to only 7% of all murder victims and 15.0% of all child murder victims. The dramatic differences in the role of sexual motivation and conduct across the different types of murder clearly demonstrate the unique role that sex plays in child abduction murders. The great majority of these killers would qualify as sexual predators in most states. 31 Parnagrqahy There is a common belief in our society that pornography plays an important role 11?] the process of motivating sex killers and lust killers. Interestingly, approximately one-quarter of the killers in this study used pornography and porno graphy was a ?trigger? for the murder in 20.7% of the cases. This was an increase the use of pornography as a trigger over the 4% reported for the ?rst-stage data. It may be that the role of pornography in the sexual motivation of these murders was more prevalent but was still not common. It could also re?ect differences in the selection of the ?rst- and second-stage cases. One explanation for the low reliance on pornography is that child abduction killers may have a prediSposition to engage in these types of violent and sexual acts with children as a deep-seated element of their ?awed characters, making the exposure to porno graphic materials unnecessary in the process of ?getting motivated? to commit the murder. Crises and Stressors Some observers of murder have proposed that certain kinds of personal problems usually revolving around employment or marriage may serve as ?precipitating crises? that contribute to the motivation of the killer. In the case of child abduction murders, there was evidence of at least one precipitating crisis (or stressor) in the life of the killer in 25.4% of the cases. What is striking is that of those cases, the usual crises or stressors emphasized in the literature did not seem to be as important as others that seemed to resonate with the character of the killers and with their choice of predominantly female victims. For example, only 4.9% involved marital problems, 3.3% involved employment problems, and 6.0% involved ?nancial problems. 32 This research shows that 1.3% of the child abduction murderers had a ?con?ict with a female,? and 8.4% with ?cruninalflegal problems.? The majority of these killers had extensive criminal histories, so it is not surprising that their related criminal/legal problems might be implicated somehow in the motivation to commit murder. We also know that there was a dramatic, preponderance of young, vulnerable, female victims. Choosing and Committing the Vidim The child abduction murderers overwhelmingly chose their victims because the opportunity presented Killers rarely chose their victims for their physical characteristics (9 or because of a prior relationship with the victim Killers had a speci?c motivation to murder a particular victim in only 12.7% of the cases. Table 20 ow Killer Chose Victim Choice Victim of Opportunity 40.3 Prior Relationship with Victim 14.4 Speci?c Motivation 12.7 Physical Characteristics 9.7 Table 21 indicates that some killers initiated contact with their victims through deception and others through normal activities Only 12.9% of the killers contacted their victirns through other illegal or criminal activity. In 34.2% of the cases the manner of initial contact was unknown. 33 Table 21 How Killer Initiared antacr with Victim Method Through Deception 27. 3 Through Normal Activity 22.9 Through IllegaUCrirninal Contact 12. 9 Other 2.7 Unknown 34. 2 The initial victim-killer contact most often took place in a suburban area and least often in a rural area (16.7). Table 22 Area Contact Area Suburban Urban Rural 32.9 50.5 16.7 34 Table 23 illustrates the differences in area of initial contact by type of murder. Table 23 Area of Initial Victim-Killer Contact by Type of Murder Child Abduction Area Murders All Child Murders All Murders Urban 50.5% 17.0% 31.0% Suburban 32.9% 74.0% 58.0% Rural 16.7% 9.0% 11.0% Typically, killers came into contact with the victim because the killer lived in close proxnnity to the victim An additional 16.4% came into contact with their victim because of other criminal activity. Table 24 shows the reasons the killer was at the victim contact site. Table 24 Why Killer Was at the Victim Contact Site Reason Lived in Area 26.3 Criminal Activity 16. 4 Normal Social Activity 13.9 Normal Non-social Activity 10.2 Meeting Victim Intentionally 4.7 Other 2.2 Unknown 27.9 35 The majority of killers gained control over the victim by using direct physical assault An additional 2.1% used verbal threats, with deception and victim incapacity used equally as often to gain control of their victims. Table 25 illustrates the methods of control used by these child abduction murderers. Table 25 How Killer Gained Control of Victim Method Direct Physical Assault 47.2 Deception 11.6 Victim Incapacity 11.1 Verbal Assault 2.1 Unknown 30.6 Binding of Victim There is evidence that child abduction killers are 6 to 12 times more likely than other murderers to ?bind? their victims. Child abduction killers bound their victims in one-fourth of the cases, compared to only 2.0% in child murders and 4.0% in all murders. The much more frequent binding of child abduction murder victims re?ects both control and sexual elements. Binding a victim makes control easier, and for uncooperative, strong victims it may be absolutely necessary. For child victims, this control ?lnction of binding is less critical. But in these types of murders, with their strong sexual component, the binding (or ?bondage?) is more likely to serve more primary and secondary sexual functions. These victims are being bound less to physically control them than to ?ll?ll the sexual fantasies and needs of the killers. 36 The source of the binding material was known in 96.8% of the cases involving binding. In those known cases, 16.2% of the killers brought the binding matenal with them to the crime scene. This has evidentiar}.I implications for matching binding material found in the possession of the killer after he is identi?ed. Cause afDeath Death by strangulation was the leading cause of death of victims followed by blunt force trauma and stabbingfcutting Firearms, a frequent cause of death in general murders, only accounted for 11.8% of child abduction murder victim deaths. Table 26 Cause of Death Cause 0,4, Strangulation Ligature and Manual 33.2 Blunt Force Trauma 23.9 Stabbinga?Cutting 23 . 3 Unknown Means 12.2 Firearm 11.8 Drowning 3.7 Table 26 lists common causes of death for victims, and Table compares the common causes of death in general murders and child murders ?ndings by Han?and et a1. (1997) to the causes of death in this analysis. 37 Table Cause ofDearh by Tme of Murder Child Abduction Cause of Death Murders All Child Murders All Murders Strangulation 33.2% 13.0% 9.0% Blunt Force Trauma 23.3% 37.0% 18.0% Stabbinngutting 23.9% 9.0% 23.0% Firearms 11.8% 16.0% 43.0% Unusual Acts There is a common belief that killers who commit murders that are out of the ordinary are involved in a variety of unusual acts during the murder incident, ranging from cult rituals, to ?posmg? victims, to grotesque mutilation. In general, the data suggest that child abduction murders are not characterized by unusual, bizarre, or weird acts and rituals. There is almost no evidence that unusual ceremonies or acts had been performed at the crime scene evidence of ritual (1 masochistic acts burnt candles, dead animals, satanic symbols). The extreme rarity of these kinds of acts in child abduction murders is consistent with what is found for all murder case s. Boa? Disposal/Recovery The geographic location of the body disposal site chosen by killers is shown in Table 28. These child abduction killers chose rural areas for the body disposal location in 52.6% of the cases. 33 Table 28 Area ofBoajr Recovery Area Rural 52.6 Suburban 20. 5 Urban 27. 0 The choice of a rural location for body disposal contrasts with ?ndings by Han?and et al. (1997) for all murder cases (Table 29). Killers in general murders choose suburban body disposal locations in about halfof all cases but rural locations in only in 22% of the murders. Table 29 Area ofBoaj; Recovery Sire by Type ofMurder Child Abduction Area Murders All Child Murders All Murders Urban 27.0% 14.0% 25.0% Suburban 20. 5% 61.0% 54.0% Rural 52. 6% 26.0% 22. 0% The killers deliberately chose the body disposal location in 48.0% of cases. It was a random selection in 32.6% of cases. As shown in Table 30, killers were forced by circumstances to choose the body diSposal location in only 10.4% of the murders. 39 Table 30 How Killer Selected Body Disgrami Site Selection Deliberater Selected 48.0 Randomr' Arbitrary Choice 32. 6 Forced by Circumstances 10.4 Killers concealed the victim?s body in 55.4% of cases and were unconcerned with the location of the bodyr in 36.4% of cases. As illustrated in Table 31, the victim?s body was simply left in the open in only 8.1% of cases and was left in an unusual position even less often Table 31 How Killer Disposed of Body Selection Concealed 5 5 . 4 Unconcemed 36. 4 Open 8. 1 Pest O?ense Behavior After the murder was committed and the body disposed of, the killer apparently engaged in a variety of behav1ors that are related to the murder, wh1ch for many of them constituted a prelude to apprehension and arrest. The killers did a number of things after the murder (Table 32). A surprising 23.7% of child abduction killers returned to the body disposal site. Some killers left town right after the murder or con?ded in someone about their involvement in the murder Some followed the case in the media Ten percent actually 40 interjected themselves into the murder investigation in some way. Each of the behaviors may provide valuable leads for investigators to pursue. Table 32 .Posar Offense Behavior of Killers Behavior Returned to Body Disposal Site 23.? Left Town 16.1 Con?ded in Someone 16.1 Followed Case in Media 14.3 Interjected Self Into Investigation 10.0 Contacted Victim?s Family 9.7 Contacted Police or Media 9.7 Changed Residence 8.3 In 15% of the cases, the killer kept the body longer than necessary to dispose of it and kept it in convenient and accessible places where it could be concealed, moved quickly, andfor ?played? with. The killer kept the body in his residence in his car or in a variety of other places within easy reach. However, contrary to beliefs about murderers, especially serial killers who prolong their relationship with the victim, these child abduction killers only held onto the bodies for very short periods of time. Of the bodies that were kept by the killer, 32.1% were in his possession for three hours or less and 56.6% for 24 hours or less. So it is likely that mo st of the bodies were being kept only until they could be disposed of safely. Only 15.1% of the bodies that were kept by the killer were in his possession for more than a week. It is likely that in these cases there was a reason for keeping the body other than delayed 41 disposal for example, to play out sexual fantasies with the corpse or to treat it like a trophy. And this is a very small number of cases. More striking is the number of child abduction killers who returned to the body disposal site. Almost one-fourth of the killers returned to the body disposal site after a Significant period of time. Of those killers who did return, an incredible 83.3% did so prior to the discovery of the body, and 33.1% did so within three days after the murder. Clearly, a signi?cant proportion of child abduction killers retumed to the body disposal site, particularly soon after the murder occurred. As one would expect, very few returned after the body had been discovered and reported in the media. But an opportunity exists albeit for a short period of time for investigators to observe potential suspects between the time the body is reported to the police and, then, made public by the media. 42 CHAPTER VII INVESTIGATION Introduction The previous chapters detailed valuable information about the victims, killers, their relationship, the killers? mot1vation, and how the killers chose their v1ct1ms. In addition to this information, variables relating to the actual investigation process may prove valuable to detectives. A descriptive analysis of the variables affecting abducted child murder investigations will be explored in the following sections. Police Refusal to Accept Missing or Runway Child Report The data for this study showed that a law enforcement agency re?ised to accept an initial report for a missing or runaway child in less than two percent of child abduction murder cases. Because this number was su?ic1ently small, it cannot be determined whether the refusal to accept the initial missing or runaway child report had any effect on the investigations. Fortunately, the re?Jsal to accept an initial report about a missing or runaway child was rare. Multiple Police Agencies Some murder victims are reported to be missing from within one law enforcement agency?s jurisdictional boundary, and the body is subsequently discovered within another jurisdiction. While this is not unusual, what has not been well understood is to what extent that actually happens. Of the murders studied in this project, 64% of the victim bodies were discovered within the jurisdictional boundaries of the law enforcement agency that received the initial missing/abduction report. In short, almost two-thirds of the victims apparently did not cross law enforcement jurisdictional boundaries in the course of events that led to the discovery 43 of the body. On the other hand, 36% of the killers transported victims and dumped their bodies in jurisdict ions different from where they were ?rst contacted, assaulted, or killed. It 1s a common belief that such crossing of jurisdictional lines would have a negative impact on the outcome of the investigation. What was found was that, while the investigation was at times made more complicated for the detect1ves, there no statistical difference in the clearance rates between those cases that crossed jurisdictional lines and those that did not. Unknow?ng Unknowing witnesses are witnesses who saw some aspect of the crime but at the time did not realize that they were witnessing part of a crime or potential abduction. It was discovered that this is quite common there were unknowing witnesses in 32.9% of the cases. This would indicate that a neighborhood or area canvass would be of great importance in generating investigatlve leads. Neighborhood Canvass ?Neighborhood canvass? means the checking, by police, around the area of the victim?s last known location andior the location the victim was known to be going, or around any site determmed to be important to the investigat 1011, in an effort to locate witness es or to obtain facts about the circumstances. This typically involves going from door to door, person to person. An area search differs from a neighborhood canvass in that the latter typically involves going from door to door to contact potential witnesses, while the former is more likely part of the actual search for the victim andi?or physical evidence. These two activities may occur simultaneously or separately, depending on circumstances. They are separated here for greater specificity. 44 A neighborhood canvass was conducted within less than 2.5 hours of a report of an abducted or a child 1n 62.1% of the cases. A canvass was conducted in less than 12 hours in 70.4% of the cases. But in over 21% of reported abducted or missing children investigations, no neighborhood canvass was conducted. What was not collected in this research was information on who actually conducted the neighborhood canvass. However, it is a common practice in law enforcement to have uniformed patrol officers conduct the initial neighborhood canvass. Whether or not this is done is usually a simple issue of manpower availability. Unfortunately, most police agencies do not have Standard Operating Procedures for these types of cases, and patrol of?cers frequently are not given speci?c guidance as to what to ask the people they interview. It is also unfortunate that in many cases there is no follow-up canvass by detectives. On the surface, the lack of a neighborhood canvass may not appear to be a critical factor in these types of mvestigatlons. However, it may be one of the most critical factors uncovered in this research project, especially when combined with what we know from the data concerning murder incident sites. We know that the victim?s last known location was usually very close to the site of the initial contact between the killer and the victim. When the police did not know the initial contact site, the solvability rate dropped to 23.9%. When the initial contact site was known, the solvability rate increased to 79.7%. The initial contact site is the location at which there is potentially the greatest chance for a witness to observe the killer and the victim together. This site is where the killer may be most likely to expose hiinself to observation by others. There are quite o?en witnesses, some unknowing, to the initial contact, as evidenced by the fact that agencies that conducted thorough neighborhood canvasses discovered those witnesses and obtained their statements. 45 The initial neighborhood canvass is also an opportunity for law enforcement officials to put everyone contacted ?on the record.? A use?il technique 1n conducting a neighborhood canvass is to provide patrol of?cers and detectives printed questionnaires that include speci?c questions, including the identi?cation of everyone who lives in each residence. The completion of these questionnaires facilitates easy reporting of the results of the 1nterviews of each houselbusiness, and the content of subsequent interviews can be compared with the initial statements. Such comparisons can identify inconsistent statements and persons who have left the area since the initial canvass. It is worth pointing out that in 33.5% of the cases examined in this research a physical description of the killer was obtained before he was identi?ed. In those cases the clearance rate increased by 15.0%. Area Search The area most commonly searched (98% ofthe cases) was the area in wh1ch the victim was last seen. It was searched within ?ve hours in 72.9% of the cases. Searching resulted in ?nding evidence or leads in 45.5% of the murders. Of course, when evidence or leads were found by searching, cases had a higher solution rate than those in which no evidence or leads were found. Like the neighborhood canvass, the initial area search is frequently conducted by uniformed patrol of?cers. Later searches are often conducted with volunteers, such as Explorer Scouts or Search and Rescue teams. However, experience has found that explicit instructions on what the searchers are looking for are missing from these searches. For every child abduction murder a police agency handles, there could be hundreds to thousands of reports of a missing child. This issue is important because it is easy for officers responding to a missing child call to think in terms of a lost child, and not the possibility of an 46 abduction murder. A lost child is usually thought to be alive, not dead. With a ?lost chil mindset, one could overlook the body of a murdered child that has been concealed or inadvertently destroy evidence when murder is not considered as a possibility. There are examples ofcases in which of?cers made contact with the killer at his resrdence and, unknown to the of?cers, the body of the victim was on the premises at that moment. While there is no way of knowing whether a body is in the residence of a potential witness while investigators are at the front door, in at least one of the cases, during a search of the killer?s residence, investigators missed discovering the body that was hidden in a box. The chances of such things happening are slight, but of?cers should always remain mind?ll of the possibility. The research con?rmed that the body of the victim is the single most important source of physical evidence that can be connected to the killer. It also told us that 55.4% of the victims? bodies were concealed when they were disposed of by the killer. These two facts make a thorough area search very important in the investigation. (See the section on the Body Recovery Site for ?lrther discussion of investigative implications related to the area search.) Who Discovered the Body ofthe Victim? It is important to examine who ?nds the body of a child abduction murder 1victim. Table 33 shows that in the majority of cases the victim?s body was found by an innocent passerby or by the police. The killer claimed to have ?discovered? the body in about three percent of the cases. 47 Table 33 Who Discovered Wctim ?3 Body Body Discovery Passerby 55.3 Police 25.4 Search Party 5.9 RelativefAcquaintance of Victim 4.4 Killer 2.8 Fire Departmenthid Crew 2.2 Witness to Death 1.7 Investigative Steps in the First 48 Home The investigative activities ofthc critical ?rst 24 to 48 hours of the murder investigations were classi?ed into seven general categories, as listed in Table 34. Collecting information was by far the most common investigative activity (87 . Circumstances in the early phases of an abducted child murder case can vary ?om simply accepting a missing child report and conducting no immediate investigation, to immediater dispatching officers to the area and conducting a neighborhood canvass and an area search to a crime scene investigation and arrest of the killer. 48 Table 34 Major Investigative Steps in the First 48 Hours Investigative Step Collected Information 87.2 Searched 5 1.4 Collected Evidence 40.3 Disseminated Information 24. 8 Direct Contact With Killer 14.1 Organizational Changes 3.9 ?ber 3.8 Police Contact with Killer Law enforcement contact with or knowledge of anyone who may physically be around some aspect of the murder, crime scene, or body disposal site is critical to investigations. Alarmingly, the evidence shows that the police had contact with the killer about some aspect of the murder before he became the prime suspect in half of the cases. The police often do not realize or know that they have come that close to the killer probably early in the investigation when many names are being recorded, interviews are being done, canvassing is taking place, records are bemg searched, tips are being received, and so on. Police need to know this when conducting the investigation of a child abduction murder the killer?s name may be in their possession in a substantial proportion of cases, and probably early in the investigation. Equally surprising, the data show that the killer?s name became known, not necessarily as the suspect, very early in the course of most investigations of child abduction murder. In 21.9% of the cases, the killer?s name came up immediately, in 39.8% within 24 hours, and 57.0% 49 within a week. This might be cause for alarm, but there is other evidence that the police are on the trail of the killer relatively early in the investigation as well. Police focused on the killer as a suspect or person of interest almost immediately at the beginning of the investigation in 17.7% of the cases, and within one and one-half days in an additional 39.9%. But, unfortunately, there was a dramatic drop-off after a month from the beginning of the murder investigation, there was still no primary suspect in more than 34.7% of the murders. Fortunately, it also means that by a month?s end, investigators had focused on the killer in a large majority of cases. Physical Evidence Physical evidence linked to the killer was collected in 67.2% of the cases. While there is a correlation between the gathering of physical evidence and so lvability, physical evidence by itself does not ensure solution. The most common ev1dence collected that is related to the killer was hair (Table 35). Strands of the killer?s hair were collected in 26.1% of the cases of child abduction murder. When we look at hair evidence in all murder case s, it is present in only 18.0% of the cases, and that includes killer, victim, animal, and ?imknown? hair evidence. The most common evidence collected in all murder investigations is a weapon In child abduction murders, weapons were collected in only 20.0% of the cases. This is consistent with cause of death data that show children are killed less o?en with a weapon and more o?en by human physical agency. 50 Table 35 Physical Evidence Related to Killer Evidence Hair 26.1 Weapons 20.0 Prints (Finger and Shoe) 18.0 Semen 17.2 Fibers 15.9 Blood 14.3 Discarded Evidence The police found physical evidence deliberately discarded by the killers in 24.2% of the cases. Of that discarded evidence, 36.0% of it was found along the roads on which the killer traveled in the course of the murder, body disposal, or escape. It is important to note that 56.5% of the evidence found along a road was within one mile of the body recovery site. Increased clearance rates are observed when discarded evidence is found by the police in general murders. This has important investigative implications for child abduction murder investigations. In a child abduction murder, there IS a good chance that there 1s discarded physical evidence along the road withm one mile of the body disposal site, and that improves the odds of case solution. Is it worth the e??ort to do that kind of search? The answer depends on the circumstances surrounding the case. The detectives at the scene have to make that decision. However, there is now strong evidence that it might pay off in catching the killer to have a set of objective criteria to apply during conmderatlon of investigative options. 51 Pahigrqah The use of the polygraph in investigations of murder of abducted or missing children is fairly While it is a common practice, its utility is in question. As shown in Table 36, the polygraph was used to test acquaintances and family members of the victim. However, in 33.8% of the cases in which the polygraph was used, a person other than the accused showed ?deception? on the test. Interestingly, of those persons who showed deception, 13.3% were family or friends of the victim, the two groups most often polygraphed. Table 36 Reiarionship ofPemons Poi?raphed 10 Victim Relationship Acquaintances 26. 3 Family Members 22.7 Strangers 19. 2 Neighbors 7.2 When two additional facts are considered, the use of the polygraph becomes more clouded: l) the younger the victim, the less likely police believe the polygraph helps the investigation; but 2) the younger the victim, the more likely the family or friends of the Victim will he polygraphed. So it seems that those who are more likely to be polygraphed are more apt to show a ?false deception.? And the less likely the polygraph may help the case, the more likely it is to be used. Overall, the use of the polygraph was perceived by detectives to have helped the investigation 111 62.1% of the cases, while in 9.0% of the cases detectives believed that it was 52 inconclusive or led the investigation in a false direction. This discussion is not intended to be a condemnation of the use of the polygraph. It is intended to provide facts on which to base future decisions regarding whether or not the polygraph is appropriate in a speci?c abducted child murder investigation. News Media Anyone familiar with criminal investigations understands that media involvement sometimes disrupts orderly investigation. At times there tends to be a distrust of the news media by law enforcement officers. There are many concems about the media; for example, fear of inappropriate information being released, telling the suspect what the police know or do not know, or that media representatives will interfere with the progress of the investigation. Of the detectives interviewed, 46.3% believed that the media coverage of the investigation was ?excessive.? Those concerns have led many police agencies to establish a Media Relations Officer, either as a time position or as a position appointed on a case-by?case basis for potentially high-visibility investigations. We know from experience that a designated media relations of?cer generally makes for a smoother working relationship with the news media. However, there is absolutely no correlation between whether a law enforcement agency has a media relations of?cer and the clearance rates of the murders of abducted children. In spite of the law enforcement concerns, and in spite of the high-pro?le nature of these types of murder, the police used the news media in one way or another in 74.9% of these abducted child murder cases. The most common use of the media was in an attempt to locate witnesses. Even while using and cooperating with the news media, there is a fear in law enforcement that news media involvement may hinder the investigation in one way or another. 53 To justify their suspicions, homicide detectives point to examples of media interference hurting an investigation. In actuality, 1n child abduction murder investigations, the media effect on the criminal investigation is more positive than negative. As shown in Table 37, in 56.8% of these cases, the media involvement had no effect at all. Table 37 E?ect of Media Coverage on the Investigation Effect No Effect 56.8 Helped 32. 9 Hindered 6.2 Unknown 4. 1 However, in 32.9% of the cases, the investigating detective believed that the use of the media helped the investigation, usually by bringing a witness forward. The detective felt that the media hindered or hurt the investigation in only 6.2% of the cases. An argument can be made that any hindrance is too often, but we will learn shortly that there are other institutions that have a more negative impact on the investigations. Outside Forces The: Created Probiemsfor the Investigation As demonstrated in Table 38, the three most frequently identified outside forces that created any kind of problem during investigations were another outside law enforcement agency, news media, and the family of the victim. 54 Table 38 Problems Caused by Outside Forces Outside Forces Police Agencies 9.9 News Media 9.1 Famin of Victim 5.0 CommunitylPolitical Pressure 2.8 1 3 Red Herrings A ?red herring? is something that diverts attention from the matter at hand. An example of a red herring in a murder investigation might be a vehicle that detectives believe to be the killer?s vehicle, with a major commitment of manpower devoted to ?nding it at the expense of other avenues of inquiry only to learn later that the car was not involved. Red herrings were reported in 33.0% of the investigations. They ranged from ?good? suspects, to physical evidence, to erroneous polygraph results. While troublesome and time consuming for the investigation, the presence of a red herring had no statistical effect on the clearance rate of the cases. 55 CHAPTER MURDER INCIDENT SITES Introduction Keppel and Weis (1993) conceptualize the crime of murder as an incident. Each murder incrdent includes multiple sites or locations of contact between the killer and the victim, as well as potential witnesses. The investigation ofa murder should emphasize the search for information about the major sites of a murder incident. The presence of information that establishes the existence of each site, coupled with when and where each site is located within the incident, and the manner in which their relationships affect each other, greatly in?uences the solution of murder investigations. According to Keppel and Weis (1993), the follow-up investigation of a murder involves the gathering of information about the various components that are locations of victim-killer contact. The types of mformation cruc1al to the investigation, in order of then usual occurrence WithJ? the murder incident are: 1. Where and when the victim was last seen. Where and when the killer initially contacted the victim. Where and when the murder took place. Where and when the body was recovered. Conyoner?s of?ng Murder Incident Each of these components occurs in a murder, and therefore each is important to a murder investigation. The components are defined as: 56 1. The Victim Last Seen Site (VLS): the location where and time when the victim was last seen. The VLS was determined from eyewitness information and records indicating when and where the victim was last seen alive. 2. The Initial Contact Site (1C): the place where and time when the killer initially contacted the victim. The was established from evidence indicating that the killer ?rst met the victim at a certain time and at a specific location during the course of the murder incident. 3. The Murder Site (MS): the place where and time when the victim sustained the death-producing injuries. 4. The Body Recovery Site (BR): the location where and time when police, medics, or witnesses found the victim, dead or alive, prior to transportation to a medical facility or morgue (Keppel, 1992', Keppel Weis, 1993). Information about the lo cation and the time of each site within the sequence of the murder incident has an impact on case so lvability. In most murder cases, the events occur simultaneously, and the research suggests that all events are located in the same place and not separated by distance or spans of time (Han?and, et al., 1997; Keppel Weis, 1993). Separation of Congloan by Tone and Distance The sites within incidents of murder may become separated in time and distance in two ways. First, the killer may consciously attempt to separate time and distance. The killer may believe that the separation of murder components will delay the discovery of various sites, perhaps for a long time, contributing to the destruction of evidence. The separation may also inhibit the investigation by creating communication and cooperation problems among police 57 agencies, because the locations are not within the authority of one jurisdiction. Second, the killer may Simply unintentionally separate the location of components over ?rm and distance. Investigative Implications The information which identi?es the location and time of each site is vitally important to investigations. The con?rmation through evidence of the date, time, and location of a site prior to the identi?cation of a possible suspect enables the investigator to more accurately check the whereabouts and verify alibis of a suspect against the time and location of that site (Han?and, et al., 1997). Keppel and Weis (1993) found that when any information on the dates and locations of the four murder incident sites was known, the probability of case solution increased. The study also showed a strong positive correlation between knowing the dates of occurrences for the murder incident locations and the ability to identify a perpetrator. Another important ?nding was that case solvability increased as the time separating pairs of murder incident sites decreased. The study supported their hypothesis that the more investigators know about the distances between the pairs of the murder incident sites, the more likely a case will be solved. The research also showed that when the distance between the locations was less than 200 feet, the solvability increased substantially. Finally, the research showed that when the times and distances decreased among pairs of murder incident sites, the solvability increased (Keppel, 1992). The conclusions regarding time and distance in the Keppel and Weis (1993) research supported what experienced murder investigators have learned through their observations of murder cases. The more information detectives have about a case, especially about the murder incident sites, the more likely it will be solved. Their study showed that it was not ?just any 53 information? that will enhance solvability some information is more valuable and use?il than other information 111 murder mvestlgations. Separation of Components in Child Abduction Murders Because information known about each murder incident component is critical to murder investigations of abducted children, the murder incident components were examined by time and distance separation. The variables used for this analysis captured the date of occurrence (exact or approximate), the type of location, andfor the address of that location. The following sections outline what information about the murder incident component was known on tinie, location, or both in the investigations of child abduction murders contained in this data set. When one is Because information known about each site is critical to the basic model for child abduction murder investigations, the murder incident components were examined by whether or not the time of the component was known. The murder incident component about which ?time? was most often known was the Victim Last Seen Site followed by the Body Recovery Site Initial Contact site and the Murder Site This knowledge is detailed in Table 39. 59 Table 39 Murder Incident Component Time Known Murder Incident Component Victim Last Seen 98.9 Initial Contact 87. 5 Murder 84. 5 Body Recovery 98.6 When Place is Know The murder Incident component about which ?location? was most often known was the Body Recovery Site This knowledge was followed in order by the Initial Contact Site Victim Last Seen site and the Murder Site Table 40 illustrates these ?ndings. Table 40 Murder Incident Component Place Known Murder Incident Component Victim Last Seen 78.1 Initial Contact 78. 2 Murder 69.4 Body Recovery 98.4 60 When Time or Place is Known The murder incident component about ?time or location? was most often known was the Victim Last Seen Site This was followed by the Body Recovery Site Initial Contact Site and the Murder Site Table 41 indicates the results of this analysis. Table 41 Murder Incident omponeni? Time or Place Known Murder Incident Component Victim Last Seen 99.3 Initial Contact 93. 5 Murder 91.3 Body Recovery 99.0 This order is unique to child abduction murder investigations. In general murder, Keppel (1992) found that the location of the Body Recovery Site was the location most often known, followed by the Murder Site, Victim Last Seen Site, and the Initial Contact Site. The data indicates that in child abduction murder investigations the order switched to the Victim Last Seen Site as the location about which information was most likely known. This is due to the nature of the victim Children are missed more quickly than adult victims in some cases, especially small children who are usually in the presence of a caregiver. It is also not surprising that in these types of investigations that the percentage of knowledge about murder incident locations is higher than in general murders because of the high-pro?le nature of child abduction murder investigations. When Bath Time and Place are Know An examination of the murder incident components about which both time and place were known produced a decrease in the known percentages. The murder site about which both time and 61 location was most often ?known? was the Murder Site followed by the Victim Last Seen Site (77 .796), Body Recovery Site and the Initial Contact Site as indicated 11?] Table 42. Table 42 Murder Incident Component Time and Place Known Murder Incident Component Victim Last Seen 77.7 Initial Contact 62. 6 Murder 93.0 Body Recovery 72.2 The Initial Contact Site was the murder incident component least identi?ed by time and place in child abduction murders. This is not surprising, given that much of this information is gathered from witness statements or statements by the killer. Time Interval Bmveen In cide In addition to ?knowing? about the time and place of the murder incident components, the actualtime span between murder incident components was examined. The time spans between the Victim Last Seen Site to Initial Contact Site, Victim Last Seen Site to Murder Site, and Initial Contact Site to Murder Site, were then coded Into one of the followmg time interval categories: 1. 0 minutes to 29 minutes 59 seconds 2. 30 minutes to 59 minutes 59 seconds 3. 1 hour to 4 hours 59 minutes 59 seconds 4. 5 hours to 24 hours 5. 24 hours 62 The Victim Last Seen Site to Body Recovery Site, Initial Contact Site to Body Recovery Site, and Murder Site to Body Recovery Site were coded rate one of the following eate genes: 1. 0 minutes to 7 hours 59 minutes 2. 8 hours to 15 hours 59 minutes 3. 16 hours to 23 hours 59 rrunutes 4. 24 hours to 47 hours 59 minutes 5. 48 hours to 71 hours 59 minutes 6. 72 hours to 167 hours 7. 7 days to 14 days 8. 15 days to 30 days 9. is 30 days. The categories were based on the natural breaks in the frequency distribution of the CAM data. It is interesting to note that the tune span between when a victim was last seen and the initial contact with the killer was less than 30 minutes in over three-fourths of the child abduction murder investigations. The time span is illustrated in Table 43. Table 43 Time Span Between Victim Last Seen and Initial Contact Time span Interval 0 minutes to 29 minutes 59 seconds 76.3 30 minutes to 59 minutes 59 seconds 7.9 1 hour to 4 hours 59 minutes 59 seconds 10.8 5 hours to 24 hours 4.1 ?2 24 hours .9 63 The data also indicated a very short t1me ?'arne between when the initial contact between the Victim and the killer occurred and when the victim was murdered. The abducted child was murdered in under 5 hours in 85.1% of cases as illustrated in Table 44. The victim was murdered in less than 30 minutes after he or she was last seen In 30. 8% of cases. Table 44 Time Span Between Victim Last Seen mrdMurder Time span Interval 0 minutes to 29 minutes 59 seconds 30.8 30 minutes to 59 minutes 59 seconds 16.5 1 hour to 4 hours 59 minutes 59 seconds 35.6 5 hours to 24 hours 13. 5 i3? 24 hours 3.6 64 The bodies of abducted children were recovered within 24 hours of when the victim was last seen in 37.6% of cases, within 48 hours 111 49.4% of cases, and within a week in 68.5% of cases (Table 45). Table 45 Time Spam Between Victim Last Seen andBoa?: Recovery Time Span 0 hours to 7 hours 59 minutes 17.9 8 hours to 15 hours 59 minutes 11.0 16 hours to 23 hours 59 mmutes 8.7 24 hours to 47 hours 59 minutes 11.8 48 hours to 71 hours 59 minutes 6.4 72 hours to 167 hours 12.7 7 days to 14 days 8.7 15 days to 30 days 6.1 30 days 16.7 Table 46 Time Span Between Initial Canine: and Murder Time span Interval 0 minutes to 29 minutes 59 seconds 38.1 30 minutes to 59 minutes 59 seconds 15.2 1 hour to 4 hours 59 minutes 59 seconds 31.8 5 hours to 24 hours 11.5 3? 24 hours 3-4 65 The bodies of abducted children were recovered within 24 hours of when the killer and 1victim had their initial contact in 39.0% of cases, within 48 hours in 52.6% of cases, and within a week In 73.0% of cases. The time frame between when the initial contact and when the victim?s body was recovered is illustrated in Table 47. Table 47 Time Span Between Initial Contact midBoaB: Recovery Time Span 0 hours to 7 hours 59 minutes 20.1 8 hours to 15 hours 59 minutes 10.0 16 hours to 23 hours 59 minutes 8.9 24 hours to 47 hours 59 minutes 13.6 48 hours to 71 hours 59 minutes 6.1 72 hours to 167 hours 14.3 7 days to 14 days 7.2 15 days to 30 days 5.4 3* 30 days 14- 5 The bodies of abducted children were recovered within 24 hours of when the murder occurred in 46.4% ofcases, within 48 hours in 57.6% of cases, and within a week in 76.3% of cases. The time frame between when the murder occurred and when the Victim?s body was recovered is illustrated in Table 48. 66 Table 48 Time Span Between nan-der- andBodjz Recovery Time span Interval 0 hours to 7 hours 59 minutes 26.1 8 hours to 15 hours 59 minutes 11.4 16 hours to 23 hours 59 minutes 8.9 24 hows to 47 hours 59 minutes 11 2 43 hours to 71 hours 59 minutes 5.8 72 hours to 167 hours 12.9 7 days to 14 days 6.1 15 days to 30 days 4.3 3O dam 1? ?7 Distance Between Sites The distance between murder incident components was also examined. In the ?rst-stage analysis, the distance between each pair of murder components was measured in feet or miles for each pair of components. Then the actual distance was placed into one of the following categories before inclusion in the CAM data set: 1 0 feet to 199 feet 2. 200 feet to <1 1/4 mile 3. 1/11 mile to 4 mile 4. 11/2 mile to 12 miles 5. 12 miles 67 Table 49 indicates the distance between the Initial Contact Site and the Victim Last Seen Site. Table 49 Distance Between Victim Last Seen Site and hiitiai Contact Site Distance 0 feet to 199 feet 64.2 200 feet to 1,4 mile 15.9 1/3 mile to 1% miles 8.3 3? 11/2 miles to 12 miles 7.2 12 miles 4_4 This distance was 199 feet or less in 64.2% of murders. The distance between the Victim Last Seen Site and the Murder Site was 199 feet or less 26.6% of child abduction murder cases (Table 50). Table 50 Distance Between Victim Last Seen Site andMurder Site Distance 0 feet to 199 feet 26.6 200 feet to 1/2 mile 19.2 1/2 mile to miles 13.7 33? 11/2 miles to 12 miles 24.5 12 miles 16.0 The killer trampmted the victim more than one-fomth mile from where the victim was last seen in 54. 2% of cases. The victim was tranSported under one-fomth mile in 45. 8% of murders. 63 Table 51 indicates the distance between where the victim was last seen and where their body was recovered. The victim?s body was recovered at a distance of over one-fourth mile in 67.3% of the murders. The victim?s body was recovered within 199 feet in only 15.8% of cases. Table 51 Distance Between Victim Last Seen Site and 1?de Recovery Site Distance 0 feet to 199 feet 15. 8 200 feet to 1/3 mile 16.8 1/4 mile to 11/: miles 13.4 33' 11/2 miles to 12 miles 30.5 12 miles 23.4 The distance between where the victim and the killer had their ?rst contact and where the ViCllI?? was murdered was less than 199 feet in 34.0% of murders. The distance between the Initial Contact Site and the Murder Site is illustrated in Table 52. The distance between the Initial Contact Site and the Murder Site was greater than one-fourth mile in 51.8% of murders. Table 52 Distance Between Initiai Contact Site and Murder Site 0 feet to 199 feet 34.0 200 feet to 1/51 mile 14.3 >IAmile to 11/: miles 11.5 1% miles to 12 miles 24.4 12 miles 15.9 69 'Ihe victim?s body was recovered over one and one-half miles away ?'om where the mitlal contact with the killer occurred in 53.5% of the murders. The distance between the Initial Contact and the Body Recovery Site is indicated in Table 53. Table 53 Distance Between Initial Contact Site andBady Recovery Site Distance 0 feet to 199 feet 19.7 200 feet to mile 15.2 >%mile to 11/: miles 11.6 2* 1% miles to 12 miles 28.7 12 miles 24.8 The distance between the Murder Site and the Body Recovery Site was less than 199 feet in 69.1% of the murders as shown in Table 54. (The killers informed police where the Murder Site was or con?rmed its location in their statement to police in 55.9% of cases). Table 54 Distance Between Murder Site and Body Recovery Site Distance 0 feet to 199 feet 69.1 200 feet to 1/4 mile 6.1 l/fi mile to miles 4.5 11/: miles to 12 miles 10.3 12 miles 9.6 7O Finally. the distance between each of the murder incident component sites was examined in relat1on to the other murder incident component sites. Table 55 illustrates this comparison. The distance between the Initial Contact Site and the Victim Last Seen Site was less than 199 feet in 64.2% of the murders. The distance between the Murder Site and the Body Recovery Site was less than 199 feet in 69.1% of the murders. Table 55 Distance Between Murder Incident Sites Initial Contact Murder Body Recovery Victim Last Seen 199 feet ?9 1/4 mile 3* 1A mile Initial Contact ?a 1/4 mile 3" 1/11 mile Murder 199 feet The distance was greater than one-fourth mile between the Victim Last Seen Site and Murder Site and the Initial Contact Site and the Murder Site in just over one- half of the murders. A majority of the cases showed a distance of greater than one-fourth mile between the Victim Last Seen Site and the Body Recovery Site and between the Initial Contact Site and Body Recovery Site 71 Analysis of Distance by Murder Incident Component Distance clearly has an impact in murder investigations of abducted children. In order to more fully understand the investigative implications of how distance between murder incident component locations relate, the murder 1nc1dent components will be examined individually. A detailed analysis follows. Wdim Last Seen Site The Victim Last Seen Site was within one-fourth mile of the Initial Contact Site in 79.4% of child abduction murder cases. Also, in 40.3% of the murders, the Victim Last Seen Site was less than 199 feet from the mam '3 home. In 36.1%, the Victim Last Seen Site was within one- fourth mile of the killer?s home. The data show that the typical victim is near his or her home when last seen prior to the abduction, and that often the killer is not far away. Initial Contact Site Even though these types of murders most often involve stranger relationships, in 57.2% of the cases, the victim lived less than one-fourth of a mile from the Initial Contact Site and less than 199 feet from the Initial Contact Site in 35.0% ofcases. In 36.5% of the murders, the killer lived within one-fourth mile of the Initial Contact Site. The initial contact between the killer and the victim was also typically very close to the victim?s last known location. In fact, it was located within less than 199 feet in 63.8% of cases. After the initial contact is made, the victim is taken or transported some distance away and killed. The victim?s body is then disposed of very near the Murder Site. 72 The Initial Contact Site, and knowledge about its location and time of occurrence, is the Single most important site in terms of its effect on the outcome of the investigation. If the Initial Contact Site was not discovered by police, the solvability rate dropped to 23.9%. If the Initial Contact Site was discovered by police, the solvability rate was 79.7%. We visited this issue brie?y when the Importance of the neighborhood canvass was discussed. At that time we learned that the initial contact site is where there is probably the greatest chance for a witness to observe the killer and the victim together. This is where the killer is most likely to expose and connect himself to the victim, to be observed by a witness. In 79.4% of the murders, the Initial Contact Site was within one-fourth mile of the victim?s last known location. If the Initial Contact Site is not found, the witnesses who might put the killer with the victim are not identi?ed. This information is so important that it is recommended that the neighborhood canvass be comprehensive and thorough and followed up with a repeat canvass. Why was the killer at the Initial Contact Site? The killer was in the area because he lived in the area was involved in normal social activity or was there for some other purposeful activity like working The child abduction may make more sense in light of this information, because the research shows that 40.3% of the victims were ?victims of opportunity.? The fact that the killer belonged in the area of the Initial Contact Site in almost half of the cases suggests that officers conducting the neighborhood canvass should ask, ?What and who did you see?? 73 Murder Site Unfortunately, the site of the actual murder is known less frequently than any other site. For investigative effectiveness, without the Murder Site, the police have less evidence to tie to any particular killer. study has shown that the Murder Site IS the richest site 111 terms of phys1cal evidence collection It is second in importance only to the actual body of the victim for evidence that is connected to the killer. Ifthe Murder Site is so importantonly 69.4% of the child abduction murders? The data do not provide a speci?c answer to that question. It is likely that in some of the cases, the searchers didn?t know where to look We now know that for over two- thirds of the murders, the distance between the Murder Site and the Body Recovery Site was less than 200 feet (one-half of an average city block) (Table 55). It was stated above that the Murder Site 13 the most important site in tenns of physical evidence associated with the killer. We know that it is quite often within 199 feet of the body recovery site. Armed with this information, it is recommended that investigators do a ?ne-focus search over the 200 feet in every direction from the Body Recovery Site. While the distance between the Murder Site and the Body Recovery Site was very short in the majority of the cases, the distance between the Murder Site and the Initial Contact Site was more variable. However, in approximately 52.0% of the cases the distance was greater than one-fourth mile, and in 40.7% it was greater than one and one-half miles. Batty Recovery Site The terms ?Body Recovery Site? and ?Body Disposal Site? will be used interchangeably in this report, depending on the context and whether it is from the point of view of the police or the 74 killer. For the location of the Body Recovery Site, the general geographical areas were shown previously in Table 28. Of particular note, it was discovered for abducted children murders that even though the body recovery sites were found in all of the geographical locations, they were more often found in rural areas. This contrasts with all murder cases, in which rural body disposal is much less frequent (Table 29). The killers deliberately chose the Body Disposal Site for their own reasons In 32.6% of the cases, the choice of the Body Disposal Site was purely random, while in 10.4%, the killer was forced by outside circumstances to use a certain site. In general murder cases, the killer is typically unconcerned whether or not the body is found. However, murders of missing or abducted children are quite different. The most telling information about the location of the child?s body is that the killers disposed of them mostly by concealment while concealment of the body is present in only 14.0% of all murder cases (Han?and et al., 1997). The imp heatlon for investigation is that when the police are searching for the body of a murdered abducted child, they should pay close attention to ground that has been disturbed for burial purposes and should move material available to conceal victims, such as broken tree branches or large portions of discarded rugs. In 62.7% of the cases, the Body Recovery Site was ?lrther than one and one-half miles from the victu'n?s home. Interestingly, the younger the victim, the closer the body was found to the victim?s home. The body was found at the victim?s residence in only 2.9% of cases, and in 5.4% at the killer?s residence. 75 CHAPTER IX SERIES CASES Introduction There were 116 series eases examined in this study. The data were analyzed in two ways. The ?rst involved comparisons between the series and non-series cases. For example, was the ratio of male to female victims the same? The second analysis entailed an examination of the individual cases within a series. This approach was an effort to identify ??ags? that would aid in the linking of cases. Basically, it was an attempt to ?nd a ?common thread? within a series. Congomisons of Series and Non-Series Cases There are several MO and investigative issues that are similar between series and non? series murders of abducted children. Following are some of those similarities: 1. 2. 6. Concealment of victims? bodies when disposed of. Consistent rate of body parts being removed. Average distance from the initial contact site to the body recovery site. Average distance from the victims? last known location to the body site; The rate of sexual assault. The rate at ?red herrings? show up in the investigation. There are also between series and non?series cases. 76 Wdims? Gender When the victims? gender was examined, the data show that there was a difference in victim selection. Males were overselected in the series cases. In non-series cases, males represented almost one-fourth of the Victims, while In series cases they represented almost one- third of the victims. Table 56 shows that 31.9% of the series victims were males, compared to 24.8% male victims in non-series cases, a difference of7. Table 56 Victim Gender by Series and Non-Series Cases Series Non-Series Total Female 68.1% 75.2% 70.0% Male 31.9% 24. 8% 30.0% Warm-Ki?a Relationship The series killer is more likely to be a stranger to the victim and less likely to be an acquaintance than in non-series cases. Table 57 shows that the stranger relationship jumped ?om 28.3% in non-series cases to 71.3% in series cases. The important investigative implication is In prioritizing leads. If it is reasonably suspected that the case at hand 1s 11] a series, strangers (71.3% of series killers) should become the major focus as suspects. 0n the other hand, if it is suspected that it is a non-series case, strangers (28.3% ofnon?series killers) and acquaintances (36.2% of non-series killers) should be given equal consideration as suspects. 77 Table 57 Victim-Killer Relationship by Series and Non-Series Cases Relationship Series Non-Series Total Stranger 71.3% 28.3% 66. 7% Friendz?Acquaintance 20.0% 36.2% 22.2% 6.1% 1.9% 11.1% Killers Person or! Problems Table 58 shows that of those series killers with a history of personal problems, 58.0% of them had a history of sexual problems, more than twice the rate for non- series killers Table 58 Personoi Problems of Killers by Series and Non-Series ases Series Non- Serie Sexual Problems 58.0% 25.3% Drug Problems 33.9% 20.1% Alcohol Problems 30.4% 22.6% Mental Problems 25.9% 12.1% Here, too, there are investigative implications that go to the issue of evaluating potential suspects. If the child abduction murder is known to be part of a series; the presence of a history of sexual problems in the background of a suspect should statistically make him a better suspect than one without such a background. 73 Prior Crimes Against Children Dramatically, 76.9% of the senes killers had committed prior crimes against children, while one-half that many of the non-series killers had victimized children before. And as discussed earlier, a substantial portion of those priors were violent crimes against children. Again, there are investigative implications that impact the evaluation of potential suspects. Prior arrest records and police contact records, particularly evidence of prior crimes against children, should be considered carefully when evaluating and prioritizing suspects. Linking Crises Being able to determine whether two or more murders (or other types of offenses) were committed by the same killer is often help?il in an investigation. Linking cases can be an objective, scienti?c process when it is done by matching physical evidence, such as latent prints, DNA, or spent bullets. Unfortunately, not all cases have physical evidence for seientific comparison, and we then enter the realm of subjectivity and uncertainty. One of the aims of this research project was to identify elements in child abduction murders that would, absent physical evidence, make the linking process more objective. To some extent that aim has been ?ll?lled by the information that has been discovered about series killers. This information will allow us to make better judgments about whether an arrested killer might be a series killer. Knowing the facts of the case for which he was arrested may help ?nd other cases for which he is responsible. However, the more dif?cult task of linking unsolved cases continues to be more subjective. After hundreds of factors were reviewed and thousands of calculations were made, only a couple of factors emerged that may lend some objectivity to the task of linking unsolved cases: binding of the victim and gender of the victim. It was ft)qu that in series, the killer more often 79 bound his victims. Binding is used by killers in only 4% of all murder cases, and in only 2% of all child murder cases. But child abduction murder, the binding of the victim was present in 24.0% of the cases. In non-series cases, binding was used on 21.2% of the victims, but in a much larger 38.9% of the serial murders of abducted children. Clearly, the binding of child ViCtll?l?lS may be a useful indicator of a serial killer at work. 80 CHAPTER SUMMARY AND CONCLUSIONS Introduction The murder of an abducted child, particularly by a stranger, is a rare event. There are estimated to be about 100 such inc1dents in the United States each year. The Victims are typically ?average? children, leading normal lives, and growing up in normal families. The vast majority of them are girls, with the average age between 11 and 12 years old. In most cases, the initial contact between the victim and killer is within one?fourth mile of the victim?s residence. These cases are typically reported to a law enforcement agency as a ?missing child.? Often there is no initial indication of foul play, just a report that the child is unaccounted for. This is a dif?cult time for the investigator, not knowing whether the ?missing child? is simply late in returning home, is the victim of foul play, or has been abducted. Any report to the police of a missmg child should be taken seriously. As many facts as possible surrounding the m1ss1ng child occurrence should be obtained as quickly as possible. An assessment of the nature of the case made should be made expeditiously. Factors to consider in assessing the case should include the age and gender of the child, the circumstances surrounding the child?s missing status, and the history of the child. Immediate action is dictated by two facts: 1) there 1s typically over a two-hour delay in making the initial missing child report, and 2) the vast majority of the abducted children who are murdered are dead within three hours of the abduction. Because of these critical time features, there is a need to reSpond quickly in a comprehensive, labor intensive investigation. Over half of the child abduction murders are committed by a killer who is a total stranger to the victim. However, the relationship between the victim and the killer varies with the gender and age of the victim. The youngest females, 1 to 5 years old, tend to be killed by friends or acquaintances, 81 while the oldest females, 16 to 17 years old, tend to be killed by strangers. But both the youngest and oldest male victims are likely to be killed by strangers. The average age of killers of abducted children is around 27 years old. They are predominantly unmarried, and half of them either live alone or with their parents. Half of them are unemployed, and those that are employed work in unskilled or semi-skilled labor occupations. The killers can generally be characterized as ?social marginals.? Almost two-thirds of the killers have prior arrests for violent crimes, with more than half of those prior crimes committed against children. The most frequent prior crimes against children are rape and other types of sexual assault. Most of the murderers? prior criines are similar in M0 to the child abduction murder. Commonly, the killers are at the initial victim?killer contact site for a legitimate reason. They either live in the area or were engaging in some normal activity. Most of the victims of abduction murder are victims of opportunity. Seldom does the killer choose his victim because of some physical characteristic of the victim. The primary motivation for the child abduction killer is sexual assault. A unique pattem of distance relationships exists in child abduction murders. Often, the initial contact site is located very close to the victim?s last known location. Conversely, the distance between the Initial Contact Site and the Murder Site increases to distances greater than one-fourth mile. The distance ?'om the Murder Site to the Body Recovery Site again decreases to less than 199 feet in the vast majority of cases. There are investigative implications of these Spatial relationships. If the Initial Contact Site is not identified by the police, the clearance rate drops drastically, and vice versa. The close proximity between the Initial Contact Site and the Victim?s Last Seen Site suggests thorough neighborhood canvass and area searches be completed to locate the initial contact site. The 82 Murder Site is second only to the body of the victim as a source of physical evidence that can be connected with the killer. Its close prox1mity to the Body Recovery Site suggests that a thorough search be completed to locate it. It was discovered that once the murder investigation has begun, the name ofthe killer is likely to be the investigative ?le within fke?rsz? week. This provides an opportunity for investigators who may have run out ofviable leads to regroup and review everyone whose name has been uncovered during the investigation. Similarly, it is not uncommon for the police to have actual contact with the killer before he becomes a primary suspect, for example, during the initial neighborhood canvass. Also, while at times the media seem to ?get in the way,? in the end they are much more likely to have a positive effect on the investigation than a negative one. In short, the media are more likely to bring witnesses forward than to aid the killer in his escape. How Can We Protect Our Children One question answered by this research is this: What can we tell parents to help them protect their children? Even though child abduction murders are rare events, the thing for parents to do is to eliminate, or minimize, the opportunity for their children to become victims. The ?rst step is to be aware that children are not immune from abduction because they are close to home. In fact, well over half of these abductions that led to murder took place within three city blocks of the victim's home and approximately one-third of the abductions occurred within one-half block. Perhaps the most important single thing we can do as parents to protect our children is to be certain that our children are supervised, even if they are in their own ?ont yard or neighborhood street. ?Stranger Danger? has become a common warning issued by American parents to children. There has been extensive pub lic ity about educating children to avoid abduction by ?not 83 speaking to strangers? and ?not getting into cars with strangers.? We should carry those precautions one step further. Our children should be taught not to ever.' approach a car, whether the occupam of the ear is a stranger or not, no matter what they tell or ask them. Citizens need to be aware of strangers and unusual behavior in their neighborhood. They need to have the presence of to observe and to write down descriptions of people, vehicles, and license numbers. Many child abductions are witnessed by people who do not realize that a crime is being committed. For example, when a citizen observes an adult pulling a struggling child in a public place, it is easy to interpret the event as a guardian taking control of an unruly child. h] most instances, that is exactly what it is. However, nothing prevents a citizen from evaluating the circumstances, intervening, and, certainly, from noting descriptions and license numbers, because it may be a person with criminal intentions. Finally, we need to remind parents that if their child is missing, they must call the police immediately. An immediate response to a missmg or abducted child may be the difference between life and death for the child. 84 REFERENCES Baker, T., Burgess, A, Rabun, J., Nahnny, C. (2002). Abductor violence in nonfamily infant kidnapping. Journal of Interpersonal Violence, 7(11), 1218-1233. Boudreaux, M. (3., Lord, W. D., Dutra, R. L. (1999). Child abduction: Aged-based analysis of killer, victim and offense characteristics In 550 cases of alleged child disappearance. Journal of Forensic Sciences, 44(3), 539-553. Boudreaux, M. (1., Lord, W. D., Etter, S. E. (2000). Child abduction: An overview of current and historical perspectives. Child Maltreatment, 63-387. Boudreaux, M. (1., Lord, W. D., Jarvis, J. (2001). Behavioral perspectives on child homicide. Journal ofTraumo, Violence, Abuse, 56-78. Bureau of Justice Statistics. (1999). Homicide trends in the United States. Crime Data Brief (NCJ Publication No. 173956, pp. 1-4). Washington, DC: U.S. Department of Justice. Bureau of Justice Statistics. (2000). Hom1c1de trends in the United States: 1988 update. Crime DataBrief (NCJ Publication No. 179767, pp. 1-4). Washington, DC: U.S. Department of Justice. Bureau of Justice Statistics. (2003). Homicide trends in the United States: 2000 update. Crime DataBrief. (NCJ Publication No. 197471, pp. 1-4). Washington, DC: U.S. Department of Justice. Fass, PS. (1999). Kidnapped: Child abduction inAmerica. Cambridge, MA. Harvard University Press. Finklehor, D., Hotaling, G. T., Kc Sedlak, A J. (1992). The abduction of children by strangers and nonfamily members. Estimating the incidence using multiple methods. Journal of Interpersonal Violence, 226-243. Forst, M. L, Blomquist, M. E. (1991). Missing children: Rhetoric and really). Lexington, 85 MA. Lexington Books. Hanfland, K. A., Keppel, R. D., Weis, J. G. (1997). Investigative ease managementfor missing chiidren homicides. (Cooperative Agreement Olympia, WA: Attorney General of Washington. Keppel, R. D. (1992). An analysis of the effect of time and distance relationships in murder investigations (Doctoral dissertation, University of Washington, 1992). Dissertation Abstracts International, 933. Keppel, R. D., Weis, J. (1993). Time and distance as solvability factors in murder cases. Joumai of Forensic Sciences, 39(2 286-401. Lanning, K. V. (1995). Sexual homicide of children. APSAC Adviser, 40-44. National Center for Missing Exploited Children. (2000). Missing and abducted chifdmn: A iaw-enfamement guide to ease investigation and program management. Alexandria, VA: Author. National Incidence Studies of Missing, Abducted, Runaway, and Thrownaway Children in America (NIS MART). (2002). National estimates of missing children: An overview. Washington, DC: U.S. Department of Justice, Office of Juvenile Justice Programs, Of?ce of Juvenile Justice and Delinquency Prevention. 86 PROJECT STAFF Katherine M. Brown, MA. Ms. Brown received her Master of Arts Degree in Criminal Justice and Criminology from Sam Houston State University and is currently ?nishing her Doctorate in Criminal Justice at the same institution. Ms. Brown?s research interests include child abduction murder investigation, crime scene investigation, serial murder, offender profiling, forensic evidence, and other solvability factors affecting murder mvestigatlons. Her work experience includes employment in Texas State Government and the legal ?eld. Dr. Robert D. Keppel, Project Director: Dr. Keppel is a visiting professor of Criminal Justice at Seattle University. He retired after 17 years as the Chief Criminal Investigator with the Washington State Attorney General's Office and was the founder of the Homicide Investigation Tracking System Prior to that, he was a member of the King County Sheriff?s Office for 11 years. He has investigated, reviewed, and consulted on over 2000 homicide investigations for local, county, and state law enforcement agencies. He has been the project director of several federal grants. He 15 the author of several and four books. Dr. Joseph G. Weis, Research Manager: Dr. Weis is Director of the Center for Law and Justice and a Professor of Sociology at the University of Washington. He was the past Editor of Criminology, the of?cral Journal of the American Society of Criminology. His research interests include the etiology of juvenile crime, delinquency prevention, the measurement of crime, violent crime, and the measurement of violence. He has authored numerous articles in the ?eld of criminology. Marvin Skeen, B.S., Project Coordinator: Mr. Skeen is an investigator with the Homicide Investigation Tracking System of the Criminal Justice Division of the Washington State Attorney General?s Office. He was a police of?cer for 25 years in the Bellevue Police Department, where he spent nineteen years as a detective investigating violent crimes. He is also the president of the Washington Violent Crime Investigators Association. 87