Midwest Evaluation of the Adult Functioning of Former Foster Youth: Outcomes at Age 21 Mark E. Courtney Amy Dworsky Gretchen Ruth Cusick Judy Havlicek Alfred Perez Tom Keller December 2007 ©2007 Chapin Hall Center for Children at the University of Chicago Chapin Hall Center for Children at the University of Chicago 1313 East 60th Street Chicago, IL 60637 773-753-5900 (phone) 773-753-5940 (fax) www.chapinhall.org CS-146 TABLE OF CONTENTS Acknowledgments Introduction Background and Overview of Study Demographic Characteristics Time Since Discharge from Foster Care Living Arrangements Relationships with Family of Origin Social Support Foster Care Experiences Independent Living Services Education Employment and Earnings Income Economic Hardships Receipt of Government Benefits Physical Health and Access to Health Care Services Mental Health and Utilization of Mental Health Services Sexual Behaviors Pregnancy Marriage, Cohabitation, and Relationships Children and Parenting Criminal Behavior and Criminal Justice System Involvement Victimization Civic Participation Religion Feelings about the Transition to Adulthood Life Satisfaction and Future Orientation Mentoring Trends over Time Connectedness Summary and Next Steps References 3 4 5 7 12 14 15 17 20 22 23 27 32 36 38 39 42 45 48 51 55 56 65 69 71 73 74 75 77 78 85 87 91 ACKNOWLEDGMENTS The authors wish to thank our three state partners, the Illinois Department of Children and Family Services, the Wisconsin Department of Health and Family Services, and the Iowa Department of Human Services. The study would not have been possible without their cooperation and financial support. We also want to recognize our other funders, the William T. Grant Foundation, the National Institute of Justice, and the National Institute of Mental Health, as well as the University of Wisconsin Survey Center in Madison, Wisconsin, for all of their hard work. 4 INTRODUCTION For most young people, the transition to adulthood is a gradual process (Goldschieder & Goldscheider, 1999; Settersten, Furstenberg, & Rumbaut, 2005), and many continue to receive financial and emotional support from their parents well past age 18. Approximately 55 percent of young men and 46 percent of young women between 18 and 24 years old were living at home with one or both of their parents in 2003 (Fields, 2003). Recent estimates also suggest that parents provide their young adult children with material assistance totaling approximately $38,000 between the ages of 18 and 34 (Schoeni & Ross, 2004). A very different situation is faced by young people for whom the state has been their parent. Too old for the child welfare system, but often not yet ready to live as independent young adults, the approximately 24,000 foster youth who “age out” of care each year (U.S. Department of Health and Human Services, 2006) are expected to make it on their own long before the vast majority of their peers. The federal government has been providing states with money specifically to help prepare their foster youth for this transition to adulthood since Title IV-E of the Social Security Act was amended in 1986 to create the Independent Living Program. The Foster Care Independence Act of 1999, which created the John Chafee Foster Care Independence Program, doubled available funding to $140 million per year, expanded eligibility for services, broadened the purposes for which the funds can be used to include room and board, and granted states the option of extending Medicaid coverage for youth who age out of foster care until age 21. It was subsequently amended to include vouchers for post-secondary education and training. 5 The Midwest Evaluation of the Adult Functioning of Former Foster Youth (hereafter referred to as the “Midwest Study”) is a prospective study that was designed, in part, to provide a comprehensive picture of how foster youth as they transition to adulthood since the John Chafee Foster Care Independence Act of 1999 became law. Two earlier reports from the Midwest Study (Courtney, Terao, & Bost, 2004; Courtney, Dworsky, Ruth, Keller, Havlicek, & Bost, 2005) described what was learned from survey data collected from young people in Wisconsin, Iowa, and Illinois, first at the age of 17 or 18 and then again at age 19. This third report is based on interviews conducted with the young people when they were 21 years old. BACKGROUND AND OVERVIEW OF STUDY The Midwest Study is a collaborative effort among the public child welfare agencies in Illinois, Iowa, and Wisconsin, Chapin Hall Center for Children at the University of Chicago, and the University of Wisconsin Survey Center. Planning for this project began in early 2001 when the public child welfare agencies agreed to use some of their federal Chafee funds to study the outcomes for youth who age out of care. Chapin Hall Center for Children at the University of Chicago assumed primary responsibility for overseeing the project, constructing the survey instruments, analyzing the data, and preparing reports for the participating states. The University of Wisconsin Survey Center was contracted to conduct the in-person interviews. Each state provided Chapin Hall with a list of 17-year-olds currently in care who had entered care prior to their sixteenth birthday and whose primary reason for placement was abuse and/or neglect. The sample included all of the Iowa and Wisconsin youth who fit these 6 criteria as well as two-thirds of the youth in Illinois, which has a larger out-of-home care population. Youth with developmental disabilities or severe mental illness that made it impossible for them to participate in the initial interviews and youth who were incarcerated or in a psychiatric hospital were excluded from participation. Youth were also ineligible to participate if they were on run or otherwise missing from their out-of-home care placement over the course of the field period for the initial interviews or if they were in a placement out of state. This resulted in a sample of 758 eligible youth.1 Baseline interviews were conducted with 732 of these youth, including 63 from Iowa, 474 from Illinois, and 195 from Wisconsin, between May 2002 and March 2003. That translates into a response rate of almost 97 percent. Among the reasons eligible youth were not interviewed were the care provider’s refusal to participate, the youth’s refusal to participate, or inability to make contact with the youth. All of the youth were 17 or 18 years old when they were interviewed. They were asked about their education, employment, physical and mental health, social support, relationships with family, delinquency and contact with the criminal justice system, victimization, substance abuse, sexual behavior, foster care experiences, and receipt of independent living services. Eighty-two percent (n = 603) of these 732 youth were re-interviewed between March and December 2004. This included 386 young adults from Illinois, 54 from Iowa, and 163 from Wisconsin. Most of these young adults (n = 575) were now 19 years old. Forty-seven percent (n = 282) (all but 2 from Illinois) were still in foster care when they completed their second interview; the other 53 percent (n = 321) had already exited the child welfare system. This 1 Appendix A provides state-specific information about the reasons youth were not interviewed. 7 reflects the fact that court supervision of foster youth in Iowa and Wisconsin generally ends once youth turn 18 or in some cases 19 years old, whereas Illinois foster youth can remain under the supervision of the court until their twenty-first birthday. The second interview covered many of the same domains as the first but focused on the period since the baseline interview. A third wave of survey data was collected between March 2006 and January 2007. Eightyone percent (n = 591) of the 732 study participants were re-interviewed over the course of those 11 months.2 Nearly all were 21 years old at the time. Eighty-seven percent (n = 513) of these young adults had been interviewed at age 19. The other 13 percent (n = 78) were last interviewed when the baseline data were collected. Because some of the questions dealt with sensitive topics that study participants might not have felt comfortable talking with the interviewer about, a portion of the survey was administered using Audio Computer Aided Self Interviewing (ACASI).3 Study participants listened to a recording of these questions through headphones and entered their responses into a computer.4 The use of this technology has been found to increase reporting of highly personal behaviors (Gribble et al., 1999; Turner et al., 1998). 2 Data for one of the young adults were not discovered until after all of the analyses for this report had been completed. Thus, we only report outcomes for 590. 3 Fifty-two study participants did not complete the ACASI portion of the interview, including thirty-six who were interviewed by telephone, four who were incarcerated, and twelve who refused. These study participants are missing data for all of the ACASI questions. 4 ACASI was also used during the first and second waves of survey interviews. 8 The Midwest Evaluation of the Adult Functioning of Former Foster Youth: Outcomes at Age 21 describes what these young adults told us about themselves and their experiences at age 21 across a variety of domains, including living arrangements, relationships with family of origin, social support, receipt of independent living services, education, employment, economic well-being, receipt of government benefits, physical and mental well-being, health and mental health service utilization, sexual behaviors, pregnancy, marriage and cohabitation, parenting, and criminal justice system involvement. Like the two previous reports, this report is meant to be descriptive. It does not examine causal relationships between the outcomes they experienced and either individual characteristics or out-of-home care histories. Nor does it attempt to explain differences among study participants in the outcomes we observed. Our analysis of those causal relationships and the predictors of various outcomes is ongoing and will be the focus of future reports. As in the earlier reports, we make comparisons between our sample of young adults who “aged out” of foster care and a nationally representative sample of 21-year-olds who participated in the National Longitudinal Study of Adolescent Health (henceforth referred to as “Add Health”). This federally funded study was designed to examine how social contexts (families, friends, peers, schools, neighborhoods, and communities) influence the health-related behaviors of adolescents (Harris et al., 2003). In-home interviews were completed with a nationally representative sample of students in grades 7 through 12 in 1994 and then again, with these same adolescents, in 1996. Study participants were interviewed a third time in 2001 9 and 2002, when they were 18 to 26 years old. The purpose of these interviews was to explore the relationship between adolescent health behaviors and young adult outcomes. Comparisons between the two samples were made whenever our wave 3 survey instrument contained a question that had been taken directly from Add Health. The Add Health data used in the comparisons were collected during the third wave of interviews. Our comparison group includes the 744 young adults in the core sample who were 21 years old.5 We tested whether any differences we observed between the two samples were statistically significant. For categorical variables we used chi-square as our test statistic, and for continuous variables we used a t-statistic. All of the statistical tests were done using a significance level of p < .05. Unless otherwise noted, statistically significant differences are indicated by a single asterisk. Although these Add Health comparisons provide a sense of how the former foster youth in the Midwest Study were faring during the transition to adulthood relative to a nationally representative sample of their peers, they do have several limitations. First, the Add Health sample includes young adults from many different states—not just Wisconsin, Iowa, and Illinois. Second, the third wave of Add Health data were collected 4 to 5 years before the third wave of Midwest Study data; so policy or economic factors that affect the transition to adulthood may have changed. Third, the two samples were quite different demographically. For example, approximately three-quarters of the Add Health 21-year-olds identified 5 Several groups were oversampled (e.g., African American youth from highly educated families or with a parent with a college degree), but only youth in the core sample were included in our analyses. 10 themselves as White compared with only one-third of the Midwest Study young adults. Similarly, given that approximately half the children in foster care are Title IV-E eligible (U.S. House of Representatives, 2004), it is probably safe to assume that the young adults in the Midwest Study were removed from families that were disproportionately poor, and thus they had a much lower socioeconomic background than the young adults in Add Health. We also made comparisons between the young men and the young women in the Midwest Study for most of the outcomes we examined. In general, those comparisons are only shown where statistically significant gender differences were found. DEMOGRAPHIC CHARACTERISTICS Table 1 shows the demographic characteristics of the 590 young adults who completed an interview at wave 3.6 Nearly all of these young adults were 21 years old, and the young women outnumbered the young men. Approximately two-thirds of these young adults identified themselves as belonging to a racial or ethnic minority group, primarily African American. Table 1. Demographic Characteristics of Midwest Study Participants Interviewed at Wave 3 # % Age 21 524 88.8 22 66 11.2 Gender Male 276 46.8 Female 314 53.2 Race 6 Unless otherwise noted, any discrepancies between the sample sizes reported in the tables and the overall sample size are due to missing data on particular survey items. 11 White African American Asian or Pacific Islander Native American Multiracial Don’t know/refused Hispanic Identity Yes No Don’t know/refused State Illinois Wisconsin Iowa 192 328 4 7 56 3 32.5 55.6 0.7 1.2 9.5 0.5 46 541 3 7.8 91.7 0.5 364 176 50 61.7 29.8 8.5 These 590 young adults represent 81 percent of the 732 foster youth who completed a baseline interview. Table 2 compares their demographic characteristics with the demographic characteristics of the full baseline sample of 732. None of the differences between the young adults who were interviewed at wave 3 and the full sample was statistically significant. Table 2. Midwest Study Young Adults Interviewed and Not Interviewed at Wave 3 Full Baseline Sample Wave 3 Sample (N = 732) (N = 590) # % # % Gender Female Male Race White African American Multiracial Other Don’t know/refused 378 354 51.6 48.4 314 276 53.2 46.8 226 417 71 14 4 30.9 57.0 9.7 1.9 0.5 192 328 56 11 3 32.5 55.6 9.5 1.9 0.5 12 Hispanic Origin Non-Hispanic Hispanic Don’t know/refused State Illinois Iowa Wisconsin 666 63 3 91.0 8.6 0.4 541 46 3 91.7 7.8 0.5 474 63 195 64.8 8.6 26.6 364 50 176 61.7 8.5 29.8 TIME SINCE DISCHARGE FROM CARE We used administrative data from the public child welfare agencies in each of the three states to determine when these young adults had exited foster care.7 On average, these young adults had been “out of care” for a mean of 26 months and a median of 30 months when they completed the wave 3 interview. However, this varied considerably by state. In particular, young adults from Illinois had been out of care for significantly fewer months than young adults from either Iowa or Wisconsin. It is also worth noting that although one-third of the total sample had been out of care for 12 months or less, all of the young adults who had exited recently were from Illinois. Conversely, 82 percent of the young adults from Wisconsin and 72 percent of the young adults from Iowa had been out care for 3 years or more compared with just 16 percent of the young adults from Illinois. These differences reflect the fact that Illinois is the only one of the three states that allows foster youth to remain under the supervision of the courts until age 21. 7 Discharge dates were not yet available for fifty-seven of the young adults from Illinois. The vast majority of these young adults were recorded as being in an independent living placement. There were also 191 young adults from Illinois whose twenty-first birthday preceded their “official” discharge date. For the purpose of this analysis, we assigned all of these young adults a discharge date corresponding to their twenty-first birthday. 13 Table 3. Number of Months Since Exiting Foster Care at Time of Wave 3 Interview Total Wisconsin Illinois Iowa 12 months or less 12 to 24 months 24 to 36 months 36 to 48 months More than 48 months Mean Median N = 590 # % 205 34.7 56 9.5 91 15.4 196 33.2 42 7.1 26.2 29.6 n = 176 # % 0 0.0 0 0.0 31 17.6 117 66.5 28 15.9 41.5 41.7 n = 364 # % 205 56.3 54 14.8 48 13.2 45 12.4 12 3.3 17.2 10.0 n = 50 # % 0 0.0 2 4.0 12 24.0 34 68.0 2 4.0 38.4 39.5 LIVING ARRANGEMENTS We asked the young adults in the Midwest Study about their current living arrangements and compared their living arrangements with the living arrangements reported by their Add Health counterparts. The largest percentage were living in their “own place.” In fact, these young adults were as likely to be living in their own place as young adults in Add Health. Where the two samples differ is in the percentage who reported living with their biological parents or other relatives. Young adults in the Midwest Study were much less likely to be living with their biological parents but much more likely to be living with other relatives than young adults in Add Health. Altogether, the percentage living with their biological parents or other relatives was still significantly higher among the Add Health young adults (44 percent) than among the young adults in the Midwest Study (24 percent). Even if the definition of parent is broadened to include former foster parents, the percentage living with their parents or other relatives was still significantly higher among the Add Health young adults (44 percent) than among the young adults in the Midwest Study (30 percent). 14 Importantly, 7 percent of the total sample, including 14 percent of the males but just 1 percent of the females, were incarcerated when they completed their wave three interview. Table 4. Current Living Arrangements: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study Add Health (N = 590) (N = 744) # % Own place 261 44.3 349 46.9 With biological parent(s) 45 7.6 305 41.0 With other relative 99 16.8 22 3.0 With nonrelative foster parent(s) 33 5.6 0 0.0 With spouse/partner 39 6.6 3 0.4 With a friend 38 6.5 9 1.2 Group quarters (e.g., dormitories; barracks) 18 3.1 50 6.7 Jail or prison 42 7.1 Other 14 2.4 6 0.8 Missing 1 0 Most of the young adults in the Midwest Study had been in fairly stable living arrangements since their discharge from care. Nevertheless, one-third had lived in at least three different places, including 20 percent who had lived in four or more. Table 5. Number of Living Situations Since Exiting Foster Care (N = 590) # 273 Onea Two 116 Three 74 Four 56 Five or more 63 Missing 8 a % 46.9 19.9 12.7 9.6 10.8 Includes young adults who continued to live where they were living on their discharge date. Although less than 1 percent of these young adults were currently homeless at the time of their interview, 18 percent had been homeless at least once since exiting care. Unfortunately, 15 homelessness was often not a one-time event. Over half of the ever homeless young adults had been homeless more than once. Table 6. Homelessness Since Exiting Foster Care (N = 577) # Ever homeless since exiting 102 Number of times homeless 1 45 2 20 3 10 4 or more 22 Missing 5 Length of longest homeless spell 1 night 18 2 to 7 nights 31 8 to 30 nights 23 31 to 90 nights 18 More than 90 nights 12 % 17.7 46.4 20.6 10.3 22.7 17.6 30.4 22.5 17.6 11.8 RELATIONSHIPS WITH FAMILY OF ORIGIN Despite the fact that the young adults in the Midwest Study had been removed from home after being maltreated by their parents or other caregivers, almost all had maintained at least some family ties, and in many cases those ties were quite strong. Altogether, 94 percent reported feeling somewhat or very close to at least one biological family member, and 77 percent reported feeling very close. They were most likely to report feeling close to siblings and least likely to report feeling close to their fathers. 16 Table 7. Closeness to Biological Family Members (N = 590) # Biological mother Very close 172 Somewhat close 152 Not very close 58 Not at all close 105 Not living 81 Don’t know if alive 22 Biological father Very close 74 Somewhat close 107 Not very close 44 Not at all close 170 Not living 87 Don’t know if alive 108 Grandparents Very close 210 Somewhat close 91 Not very close 39 Not at all close 80 Not living 137 Don’t know if alive 33 Siblings Very close 334 Somewhat close 139 Not very close 41 Not at all close 60 Not living 15 Don’t know if alive 1 Close to any other relative 265 Aunt/uncle 160 Cousin 79 Other 26 % 29.2 25.8 9.8 17.8 13.7 3.7 12.5 18.1 7.5 28.8 14.7 18.3 35.6 15.4 6.6 13.6 23.2 5.6 56.6 23.6 6.9 10.2 2.5 0.2 44.9 27.1 13.4 4.4 Another measure of family ties is frequency of contact. Eighty-three percent of these young adults reported having contact with one or more biological family members at least once a week. Contact was most frequent with siblings and least frequent with fathers, the same family members to whom they reported feeling the most and least close. 17 Table 8. Frequency of Contact with Biological Family Members (N = 590) # Biological mother Every day 162 At least once a week but not every day 106 At least once a month but not once a week 91 At least once a year but not once a month 60 Less than once a year 18 Never 49 Not living 81 Don’t know if alive 22 Biological father Every day 54 At least once a week but not every day 65 At least once a month but not once a week 66 At least once a year but not once a month 57 Less than once a year 27 Never 126 Not living 87 Don’t know if alive 108 Grandparents Every day 107 At least once a week but not every day 87 At least once a month but not once a week 78 At least once a year but not once a month 73 Less than once a year 22 Never 52 Not living 137 Don’t know if alive 33 Siblings Every day 198 At least once a week but not every day 168 At least once a month but not once a week 104 At least once a year but not once a month 49 Less than once a year 8 Never 46 Not living 15 Don’t know if alive 1 Other relativea Every day 105 At least once a week but not every day 100 At least once a month but not once a week 43 At least once a year but not once a month 15 Less than once a year 1 Never 1 a Among young adults who identified another relative to whom they felt close 18 % 27.5 18.0 15.4 10.2 3.1 8.3 13.7 3.7 9.2 11.0 11.2 9.7 4.6 21.4 14.7 18.3 18.1 14.7 13.2 12.4 3.7 8.8 23.2 5.6 33.6 28.5 17.6 8.3 1.4 7.8 2.5 0.2 17.8 16.9 7.3 2.5 0.2 0.2 SOCIAL SUPPORT Social support can play an important role during the transition to adulthood. However, relatively little is known about social support among young adults who have exited foster care. We measured perceptions of social support among young adults in the Midwest Study using the Medical Outcomes Study (MOS) Social Support Survey (Sherbourne & Stewart, 1991). This 19-item measure contains subscales for four types of social support: emotional/informational, tangible, positive social interaction, and affectionate. For each item, respondents rate how often a specific type of support is available to them using a 5-point scale that ranges from 1 = none of the time to 5 = all of the time. Table 9 shows the mean scores for each of the four subscales as well as mean scores for each of the individual items.8 The mean scores for affectionate support and positive social interaction were higher than the mean scores for emotional/informational support or tangible support. The mean score across all items was 3.8, indicating that the young adults in the Midwest Study perceived themselves as having social support some or most of the time. Table 9. Perceived Social Support Emotional/Informational Support Someone to listen to you when you need to talk Someone to give you information to help you understand a situation Someone to give you good advice about a crisis Someone to confide in or talk to about yourself or your problems Someone to give you advice you really want Someone to share your most private worries and fears with Someone to turn to for suggestions about how to deal with a personal problem Someone who understands your problems Emotional/Informational Scale Score 8 N Mean S.D. 589 589 588 589 588 588 588 589 589 3.85 3.90 3.85 3.88 3.59 3.54 3.77 3.57 3.75 1.19 1.11 1.19 1.26 1.23 1.46 1.24 1.32 1.05 The mean subscale score was imputed for missing subscale items to compute the total score. 19 Tangible Support Items Someone to help you if you were confined to a bed Someone to take you to the doctor Someone to prepare your meals if you were unable to do it yourself Someone to help you with daily chores if you were sick Tangible Support Scale Score 586 589 589 587 589 3.43 3.83 3.71 3.60 3.64 1.34 1.30 1.35 1.38 1.10 Positive Social Interaction Support Items Someone to have a good time with Someone to get together with for relaxation Someone to do something enjoyable with Positive Social Interaction Scale Score 589 588 589 589 4.14 3.75 3.99 3.96 1.12 1.31 1.16 1.08 Affectionate Support Items Someone to show you love and affection Someone to love and make you feel wanted Someone who hugs you Affectionate Support Scale Score 589 589 589 589 4.15 4.07 3.82 4.01 1.19 1.21 1.42 1.34 Total MOS Scale Score 589 3.80 .982 We also asked these young adults about the adequacy of their social support network. In other words, did they have enough people to whom they could turn for different types of needs? Depending on the specific need, between one-half and two-thirds of the young adults in the Midwest Study reported that they had enough people to whom they could turn. Table 10: Adequacy of Social Support Network (N = 590) Enough Too few N # % # % People to listen to you 590 390 66.1 156 26.4 People to help with favors 590 349 59.2 183 31.0 People to loan money 586 295 50.3 192 32.8 People to encourage goals 590 375 53.6 170 28.8 20 No one # % 44 7.5 58 9.8 99 16.9 45 7.6 FOSTER CARE EXPERIENCES We asked the young adults in the Midwest Study to look back on their experiences while in foster care. Almost two-thirds agreed that they were lucky to have been placed, and nearly as many reported feeling satisfied with their foster care experience. Table 11. Feelings about Foster Care N 588 Feel lucky to have been placed in foster care Agree or agree strongly Neither agree nor disagree Disagree or disagree strongly Satisfied with experience in foster care Agree or agree strongly Neither agree nor disagree Disagree or disagree strongly # % 378 57 153 64.3 9.7 26.0 367 47 176 62.2 8.0 29.8 590 Adoption is generally regarded as the most desirable permanency outcome for foster youth who cannot be reunified with their family. However, it was a relatively rare outcome among the young adults in the Midwest Study. Only 9 percent reported that they had been adopted. Another 21 percent wished that they had been. The Adoption and Safe Families Act (ASFA) of 1997 requires state child welfare agencies to seek the termination of parental rights if a child has been in foster care for 15 of the most recent 22 months. There are also some exceptions to this requirement, including if the child has been placed with kin, if there is a compelling reason to believe that termination would not be in the child’s best interest, or if the parent has not been provided with the services outlined in the reunification plan. All of the young adults in the Midwest Study had been in foster care for at 21 least 1 year, and all but one had been in care for at least 15 months. This may explain why nearly half of the young adults reported that their parents’ rights had been terminated. Table 12. Adoption and Termination of Parental Rights N 589 524 590 Adopted Wanted to be adopted (if not adopted) Biological parents’ rights terminated Yes No Don’t know # 54 113 % 9.2 21.2 280 256 54 47.5 43.4 9.2 INDEPENDENT LIVING SERVICES The John H. Chafee Foster Care Independence Program provides federal funds to help states prepare their current and former foster youth for independent living. Youth may receive services in six domains, including education, vocational training or employment, budgeting and financial management, health education, housing, and youth development. Independent living services can be provided by case managers, out-of-home care providers, or social service agencies. Table 13 shows the percentage of young adults in the Midwest Study who reported that they had received at least one service in a particular domain since their last interview. There was no domain in which even one third of these young adults had received any services. It is also worth noting that although former foster youth are eligible for Chafee-funded services until the age of 21, most of the young adults who received services did so before exiting foster care. 22 Table 13. Receipt of Independent Living Services Since Last Interview by Domain Service Domains Education Employment and vocational Health education Budgeting and financial management Housing Youth development Received Any Service in Domain Since Last Interview N # % 590 186 31.5 590 171 29.0 590 159 26.9 590 145 24.6 143 24.2 590 590 54 9.2 Recipients Who Received Services after Discharge N # % 185 54 29.2 171 63 36.8 157 52 33.1 145 37 25.5 143 43 30.1 51 16 31.4 Table 14 lists the specific independent living services the young adults were asked about as well as the percentage who reported receipt of each. In most cases, less than one-quarter of the young adults reported receiving a specific service. Table 14. Receipt of Specific Independent Living Services Since Last Interview Education Services Financial aid/loan application assistance College application assistance Career counseling School to work support Study skills training GED preparation Attend university/college fair SAT preparation Employment/Vocational Services Help developing interviewing skills Help with completing job applications Help with job referral/placement Given an explanation of workplace values Assistance identifying employers Resume writing workshop Vocational counseling Help securing work permits/Social Security card Explanation of benefits coverage Help with use of career resources library Summer employment programs Received an internship 23 N # % 589 590 587 587 587 590 590 588 109 97 91 68 65 47 47 29 18.5 16.4 15.5 11.6 11.1 8.0 8.0 4.9 589 590 589 589 588 589 590 586 589 590 590 588 123 109 78 76 75 75 64 62 59 56 42 23 20.9 18.5 13.2 12.9 12.8 12.7 10.8 10.6 10 9.5 7.1 3.9 Health Education Services Information on birth control and family planning Education on substance abuse Training on nutritional needs Training on health/fitness Training on personal care needs (basic hygiene) Training on preventive and routine health care Courses on first aid Accessing health/dental insurance information Maintaining personal health records Budgeting and Financial Management Services Training on use of a budget Training on opening a checking/savings account Training on balancing a checkbook Money management courses Assistance with tax returns Accessing information on credit Developing consumer awareness Housing Services Assistance with finding an apartment Learning about security deposits and utilities Tenants’ rights and responsibilities training Help with completing apartment application Training on health and safety standards Meal planning and preparation training Handling landlord complaints Cleaning classes Courses on home maintenance and repairs Youth Development Services Youth conferences Mentoring Youth leadership activities 590 589 589 589 117 96 79 71 19.8 16.3 13.4 12.1 590 69 11.7 590 590 588 588 63 60 58 58 10.7 10.2 9.9 9.9 589 590 590 589 590 589 582 104 100 94 88 56 55 49 17.7 16.9 15.9 14.9 9.5 9.3 8.4 590 589 589 590 590 590 590 590 590 107 91 84 84 79 75 72 49 41 18.1 15.4 14.3 14.2 13.4 12.7 12.2 8.3 6.9 586 590 590 34 32 29 5.8 5.4 4.9 Because the goal of independent living services is to prepare current and former foster youth for the transition to adulthood, young adults in the Midwest Study were asked to rate the helpfulness of the services they received in each domain on a 4-point scale, where 1 = not at all helpful, and 4 = very helpful. In general, the young adults who received independent living services perceived these services as being somewhat to very helpful. 24 Table 15. Perceived Helpfulness of Independent Living Services by Domain Not at all Service Domains Education Employment and vocational Health education Budgeting and financial management Housing Youth development N 185 171 159 145 143 52 8.6 8.8 6.3 5.5 7 13.5 Not very Somewhat % 8.1 45.9 12.9 45 4.4 41.5 6.2 45.5 7.7 37.1 5.8 28.8 Very 37.3 33.3 47.8 42.8 48.3 51.9 Mean 3.12 3.03 3.31 3.26 3.27 3.19 S.D. 0.89 0.90 0.83 0.81 0.88 1.05 We don’t know why so many of these young adults did not receive independent living services. One possibility is that services were available but the young adults did not perceive a need. Another is that they needed services but access was a problem. Consistent with the latter, 39 percent of these young adults reported that there was “training or assistance that would have helped [them] learn to live on [their] own that [they] did not receive.” Regardless of their service receipt, young adults in the Midwest Study were asked how prepared for self-sufficiency they perceived themselves to be in each of the service domains. Overall, they reported feeling not very to somewhat prepared for self-sufficiency when they exited foster care. However, their sense of preparedness seemed to increase over time. By the time they were interviewed, they generally reported feeling somewhat to very prepared. There was also some variation in their sense of preparedness across domains. They reported feeling most prepared to meet their health needs and least prepared to achieve their educational goals. 25 Table 16. Perceived Preparedness for Self-Sufficiency by Domain Not at all Prepared to achieve educational goals Prepared for employment Prepared to manage health needs Prepared to manage finances Prepared to secure housing Prepared for self-sufficiency at exit Prepared for self-sufficiency at interview N 589 587 588 589 589 586 588 8.5 3.2 3.2 7.5 4.1 17.2 5.1 Not very Somewhat % 9.7 47.4 6.5 45.1 4.8 30.1 11.0 44.0 6.1 34.5 11.4 42.3 4.6 34.4 Very 34.5 45.1 61.9 37.5 55.3 29.0 56.0 Mean 3.08 3.32 3.51 3.12 3.41 2.83 3.41 S.D. .88 .74 .74 .88 .78 1.03 .80 EDUCATION Previous research suggests foster youth approach the transition to adulthood with significant educational deficits (Blome, 1997; Courtney et al., 2001; McMillan & Tucker, 1999). Our data suggest that these deficits continue into the early adult years. Nearly one-quarter of the young adults in the Midwest Study had not obtained a high school diploma or a GED by age 21. In fact, these young adults were more than twice as likely not to have a high school diploma or GED as their Add Health peers. Conversely, only 30 percent of the young adults in the Midwest Study had completed any college compared with 53 percent of the young adults in Add Health. Table 17. Highest Completed Grade: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study Add Health (N = 590) (N = 744) # % # % No high school diploma or GED 135 23.0 80 10.8 High school diploma only 221 37.6 221 29.7 GED only 57 9.7 49 6.6 One or more years of college, but no degree 164 27.9 320 43.0 Two-year college degree 11 1.9 60 8.1 Four-year college degree 13 1.7 Graduate school 1 0.1 Missing 2 - 26 Although very few of the young adults in the Midwest Study had even a 2-year degree, only one-quarter were currently enrolled in an educational program compared with 44 percent of their Add Health counterparts. Among those who were enrolled, the young adults in the Midwest Study were more likely to be enrolled in a 2-year college but less likely to be enrolled in a 4-year college than the young adults in Add Health. Table 18. Current School Enrollment: Young Adults in the Midwest Study Compared with Add Health Young Adults Add Health (N = 744) # % 328 44.1 269 36.2 59 7.9 415 55.8 1 - P Currently enrolled in school Full-time Part-time Not enrolled Missing Midwest Study (N = 590) # % 141 24.0 95 16.2 46 7.8 447 76.0 2 - Type of school or program High school GED program Two-year college Four-year college Graduate school Total Missing 4 19 78 39 141 1 5 82 232 7 326 2 * 2.9 13.6 55.7 27.9 - 1.5 25.2 71.2 2.1 - * * * * There were a number of gender differences in educational attainment and school enrollment among young adults in the Midwest Study. Young women were significantly more likely than young men to have completed at least some college. 27 Table 19. Highest Completed Grade by Gender Males (n = 276) # % 70 25.5 109 39.8 31 11.3 59 21.5 5 1.8 2 - No high school diploma or GED High school diploma only GED only One or more years of college, but no degree Two-year college degree Missing Females (n = 314) # % 65 20.7 112 35.7 26 8.3 105 33.4 6 1.9 - p * Young women were also more likely than young men enrolled in school, and to be enrolled in a 2-year college if they were enrolled. Table 20. Current School Enrollment by Gender Male (n = 274) # % Currently enrolled 54 19.7 Part-time 25 9.1 Full-time 29 10.6 Not enrolled 220 80.3 Female (n = 314) # % 87 27.7 21 6.7 66 21.0 227 72.3 Type of school or program High school GED program Two-year college Four-year college 2 7 53 24 2 12 25 15 3.7 22.2 46.3 27.8 2.3 8.1 61.6 27.9 P * * * Almost three-quarters of the young adults who were enrolled in a 2- or 4-year college reported that they had a scholarship to help them pay for school. The next most commonly cited sources of funding for college were student loans and earnings from employment. 28 Table 21. Funding for College among Those Enrolled in a 2-or 4-Year School (N = 117) # % Scholarship 85 72.6 Partner/spouse 3 2.6 Birth parent/relative 4 3.4 Foster or adoptive parent 3 2.6 Loans 52 44.4 Employment 30 25.6 Savings 8 6.8 Independent living funds 11 9.4 Other 12 10.3 More than half of the young adults who were not currently enrolled in school reported that they had been enrolled at some point since their last interview. One-quarter of these young adults reported graduating from the program they had been in. Among the other reasons cited for no longer being enrolled were becoming employed, becoming a parent, not being able to afford school, and losing interest. Thirty-eight percent of the young adults who were not currently enrolled reported that at least one barrier was preventing them from continuing their education. By far, the most commonly cited barrier was not having money to pay for school. Table 22. Enrollment Since Last Interview and Barriers to Enrollment (N=590) # % Currently enrolled 141 24.0 Enrolled since last interview, but not currently enrolled 245 41.7 Not enrolled since last interview 201 34.2 Missing 3 Type of school or program previously enrolled in High school GED program Two-year college Four-year college 29 32 39 148 21 13.3 16.3 61.7 8.8 Reasons not enrolled Graduated Could not afford Academic problems Lost interest Became employed Became a parent No transportation Discouraged by significant others Other Total Missing Any barrier to continuing education Biggest barrier to continuing education Could not pay Need to work full-time Need to care for child(ren) No transportation Other Missing 110 47 12 47 53 53 7 5 107 441 5 100.0 - 168 37.7 82 20 25 3 38 2 48.8 11.9 14.9 1.8 22.6 - 24.9 10.7 2.7 10.7 12.0 12.0 1.6 1.1 24.3 Thirty-seven percent of these young adults had received some job training since their last interview, including 9.5 percent who were currently enrolled in a job training program. Fortythree percent of those who had previously received training had obtained a license or certificate. Table 23. Vocational/Job Training (N = 590) Currently receiving job training Not currently receiving training, but received training since last interview Certificate or license completed, if received job training 30 # 56 148 88 % 9.5 27.8 43.1 EMPLOYMENT AND EARNINGS Nearly all of the young adults in the Midwest Study reported that they had ever held a job, and 70 percent reported that they had been employed at some point since their last interview. However, only half were currently working. Excluding the incarcerated young adults increases this figure to 56 percent, which is still lower than the 64 percent of Add Health young adults who were employed. Table 24. Employment: Young Adults in the Midwest Study Compared with Add Health Young Adults Ever held a job Ever worked since last interview Currently employed Currently employed (nonincarcerated only) Midwest Study (N = 590) Add Health (N = 744) # 561 504 303 303 # 721 % 96.9 473 473 63.9 63.9 % 95.1 85.4 51.5 55.5 P * * Although a larger percentage of the young women than the young men reported being employed, this difference was not statistically significant and was almost entirely driven by the gender difference in incarceration. Once the incarcerated young adults were excluded from the calculation, the gap between young women and young men disappeared. Table 25. Employment by Gender Ever held a job Ever worked since last interview Currently employed Currently employed (nonincarcerated) Males (n = 274) # 258 226 132 132 31 % 93.5 88.3 48.0 55.9 Females (n = 314) # 303 274 171 171 % 96.5 90.4 54.6 55.2 Young adults who were currently employed reported working a mean of 35.4 and a median of 35 hours per week. Their mean and median hourly wages were $8.85 and $8.00, respectively. Although these young adults worked about the same number of hours per week as their Add Health counterparts, the latter earned about $1.00 more per hour. Table 26. Hours Worked per Week and Hourly Wages at Current Job Midwest Study Add Healthb (N = 303) (N = 472) # % # % Hours worked per week Less than 20 hours 20 6.6 58 12.3 20-35 hours 126 41.9 167 35.4 40 hours 114 37.9 150 31.7 More than 40 hours 41 13.6 97 20.6 Missing 2 Mean 35.4 35.2 Median 40.0 40.0 Hourly wages Less than $5.15 $5.15 to $5.99 $6.00 to $6.99 $7.00 to $7.99 $8.00 to $8.99 $9.00 to $9.99 $10.00 to $10.99 $11.00 to $11.99 $12.00 or more Missinga Mean Median 3 8 38 62 54 31 27 10 31 37 8.85 8.00 a 1.1 3.0 14.4 23.5 20.5 11.7 10.2 3.8 11.7 - 11 11 33 74 42 47 33 43 79 4 9.99 9.12 P 2.9 2.9 8.8 19.8 11.3 12.6 8.8 11.5 21.9 - * Data on wages were missing for thirty-four young adults who were not paid by the hour. Because the third wave of Add Health data were collected in 2001-2002, the hourly wages were adjusted for inflation using the CPI. The values shown are in real 2006 dollars. b Although there were no gender differences in the likelihood of being employed, young women who were employed worked fewer hours, on average, and were paid less for each hour that they worked than employed young men. 32 Table 27. Hours Worked per Week and Hourly Wages at Current Job by Gender Males Females P (n = 132) (n = 171) # % # % Hours worked per week Less than 20 hours 6 4.5 14 8.3 20-35 hours 48 36.4 78 46.2 40 hours 50 37.9 64 37.9 More than 40 hours 28 21.2 13 7.7 Missing 0 2 Mean 37.8 33.5 * Median 40.0 35.0 Hourly wages Less than $5.15 $5.15 to $5.99 $6.00 to $6.99 $7.00 to $7.99 $8.00 to $8.99 $9.00 to $9.99 $10.00 to $10.99 $11.00 to $11.99 $12.00 or more Missinga Mean Median a # 0 3 5 24 21 13 14 6 22 24 9.92 9.00 % 2.8 4.6 22.2 19.4 11.2 13.0 5.6 20.4 - # 3 5 33 38 33 18 13 4 9 15 8.10 7.90 % 1.9 3.2 21.2 24.4 21.2 11.5 8.3 2.6 5.7 - * Data on wages were missing for twenty males and fourteen females who were not paid by the hour. Nearly two-thirds of the young adults in the Midwest Study who were currently employed received at least one of the seven employer-provided benefits listed in Table 28. Just over half of their employers provided paid vacation days, and almost half provided health insurance. 33 Table 28. Benefits Provided by Current Employer (N = 305) # Health insurance 144 Dental insurance 129 Retirement fund 98 Paid vacation days 158 Paid sick days 117 Child care 33 Maternity leave 99 Provided with at least one 300 % 48.2 43.4 34.1 52.7 39.5 11.8 35.2 64.0 Missing 6 8 18 5 9 26 24 5 Most of the young adults who were not currently employed reported that they were physically able to work, and more than 90 percent of those able to work reported wanting to do so. Nearly three-quarters of the young adults who reported wanting to work had actively looked for a job during the past 4 weeks. Table 29. Employability and Job Search Activities (N = 256) # % Ability to work Able to work Not able to work because of a disability Not able to work because of another reason 206 22 28 80.5 8.6 10.9 Want to work (if able to work) Actively sought work during the past 4 weeks 187 153 90.8 74.3 Job search activities during the past 4 weeks Contacted employers Contacted employment agency Solicited help from friends Contacted school employment center Sent resume Completed job application Responded to a help-wanted sign Job interview Attended job training Other 115 73 103 35 72 143 110 67 27 11 75.2 47.7 67.3 22.9 47.1 93.5 71.9 43.8 17.6 7.2 34 INCOME Although more than three-quarters of these young adults reported having any income from employment during the past year, their earnings were very low. Median earnings among those who had been employed were just $5,450 compared with $9,120 among their employed Add Health peers. Table 30. Income from Employment During the Past Year: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study Add Health b p N # % N # % Any income from employment during the past year 575 440 76.5 740 642 86.8 * 430 616 Amount of income from employment (if any)a 175 28.5 $5,000 or less 211 49.2 147 23.9 $5,001 to $10,000 105 24.5 218 35.4 $10,001 to $25,000 89 20.7 68 11.1 $25,001 to $50,000 21 4.9 7 1.1 More than $50,000 3 0.7 Missing 10 26 Mean $8,914 $12,728 * Standard deviation $12,142 $16,511 Median $5,450 $9,120 a Midpoint of categories was used in the calculation of means, medians, and standard deviations if an income range rather than a specific value was reported b Because the third wave of Add Health data were collected in 2001 and 2002, earnings were adjusted for inflation using the CPI. The values shown are in 2006 real dollars. Many of these young adults reported income from sources other than their own employment, including family and friends. This suggests that at least some of these young adults relied on informal income sources to help them “get by.” Although nearly two-thirds of those who had a spouse had income from his or her employment, only a small percentage of the sample were married. In addition, only 10 percent of the young parents who were living with their child(ren) had received any child support. 35 Table 31. Income from Other Sources during the Past Year N 44 190 146 66 Any income from spouse’s employmenta Any income from child supportb Any income from EITCc Reason did not receive EITC Not eligible Not aware Other Received money from a family member 555 556 555 Received money from a friend Received money from a social service agency a # 29 20 80 % 65.9 10.5 54.8 23 19 23 35.4 29.2 35.4 208 143 39 37.4 25.7 7.0 Limited to young adults who were currently married Limited to young adults who were living with at least one child c Limited to young adults who had earnings from their own or their spouse/partner’s employment and were living with a child b Asset accumulation is an important part of becoming a self-sufficient adult. This may be especially true for youth aging out of foster care, who are less likely than other young adults to have families on whom they can depend for financial support in times of need. However, only half of the young adults in the Midwest Study had something as basic as a checking or savings account, compared with 81 percent of their Add Health peers. Table 32. Asset Accumulation: Young Adults in the Midwest Study Compared with Add Health Young Adults Any savings/checking account Owns a residence Owns a vehicle Midwest Study N # % 554 287 51.9 557 17 3.1 556 217 39.1 Add Health N # % 741 598 80.7 741 67 9.0 742 542 73.0 P * * * Not only did many of the Midwest Study young adults lack assets, but a significant minority also had outstanding debts. Nearly 1 in 10 (n = 54) had borrowed at least $200 from family or friends since their last interview. More than half of these young adults (n = 38) still owed at 36 least some of the money that they borrowed. Two-fifths (n = 226) reported having “other” debt, excluding student, auto, and real estate loans. ECONOMIC HARDSHIPS The precarious economic situation of these young adults was further reflected in the material hardships they reported. Half reported experiencing at least one of the five material hardships listed in Table 33 during the past year. They were also much more likely to experience one or more hardships than their Add Health peers. Table 33. Economic Hardships during the Past Year Midwest Studya N # % (1) Not enough money to pay rent 555 147 26.5 (2) Not enough money to pay utility bill 555 147 26.5 (3) Gas or electricity shut off 556 46 8.3 (4) Phone service disconnectedb 555 182 32.8 (5) Evicted 556 46 8.3 At least one hardship 556 275 49.5 Mean number of hardships 1.02 a Add Health N # % 734 63 8.6 736 80 10.9 737 45 6.1 740 141 19.1 738 10 1.4 741 204 27.5 .46 p * * * * * * Data on economic hardships were not collected from the thirty-four Midwest Study young adults who had been incarcerated for at least 3 months at the time of their wave 3 interview. b Add Health asked if without phone service for any reason. Another indicator of economic hardship is food insecurity. Table 34 shows the frequency of affirmative responses to a series of questions taken from the USDA’s measure of food insecurity (Bickel et al., 2000) as well as one additional question about household food consumption. The young adults in the Midwest Study were most likely to report getting food or borrowing money for food from family or friends. 37 Six of these items (shown in boldface) were used to compute a food security composite score for each young adult. This 6-item measure was developed by researchers at the National Center for Health Statistics in collaboration with Abt Associates, Inc. (Blumberg et al., 1999). Based on their number of affirmative responses to these items, more than one-quarter of these young adults would be categorized as having low or very low food security. Table 34. Food Insecuritya Sometimes or often not enough food to eat Got food or borrowed money for food from friends or family Put off paying bill to buy food Received emergency food Received a meal from a soup kitchen Cut size of meals because could not afford more Cut size of meals because could not afford more almost every month Did not eat for a whole day because there was not enough money for food Did not eat as much as should have because did not have enough money for food Hungry but didn’t eat because could not afford food Lost weight because didn’t have enough food Sometimes or often worried about running out of food Sometimes or often food didn’t last and could not afford more Sometimes or often could not afford to eat balanced meals Food security categorization based on 6-item measure (items in boldface) High food security (0 affirmative responses) Marginal food security (1 affirmative response) Low food security (2 to 4 affirmative responses) Very low food security (5 or 6 affirmative responses) Missing a N 554 556 556 556 556 556 556 556 # 65 133 104 95 22 100 25 59 % 11.7 23.9 18.7 17.1 4.0 18.0 4.2 10.6 556 106 19.1 556 554 556 556 556 90 50 47 38 56 16.2 9.0 8.5 6.8 10.1 360 49 85 62 34 64.7 8.8 15.3 11.2 Data on food insecurity were not collected from the thirty-four Midwest Study young adults who had been incarcerated for at least 3 months at the time of their wave 3 interview. RECEIPT OF GOVERNMENT BENEFITS In addition to any services they may have received from the child welfare system, many of the young adults in the Midwest Study have relied on government benefits to help support themselves. Where gender differences were found, females were more likely to report benefit receipt. Three-quarters of the young women (n = 232) and just over one-third of the young 38 men (n = 84) had received benefits from one or more of the need-based government programs (i.e., excluding unemployment insurance) since their last interview. Among the young women who were living with at least one child, that figure was 96 percent (n = 151). Table 35. Receipt of Government Benefits Since Last Interview by Gendera Females Males n # % n # Unemployment insurance 312 23 7.4 243 19 Supplemental Security Income (SSI) 312 39 12.5 240 32 Food stamps 311 197 63.3 242 54 Public housing/rental assistance 312 44 14.1 242 14 TANFb 157 31 19.6 31 0 c WIC 156 122 78.2 % 7.8 13.3 22.3 5.8 0.0 p * * * a Data on government benefit receipt were not collected from the thirty-four Midwest Study young adults (thirtythree males and one female) who had been incarcerated for at least 3 months at the time of their wave 3 interview. b Parents living with at least one child. c Female parents living with at least one child. Examining current benefit receipt reveals a similar pattern. Two-thirds of the young women (n = 200) and 22 percent of the young men (n = 53) were currently receiving benefits from one or more of the need-based government programs. Among females who were living with at least one child, this figure was 86 percent (n = 137). Table 36. Current Receipt of Government Benefits by Gendera Females n # % Unemployment insurance 312 2 0.6 Supplemental Security Income (SSI) 312 35 11.3 Food stamps 311 156 50.2 Public housing/rental assistance 312 24 7.7 a TANF 157 14 8.8 WICb 156 84 53.8 a n 243 240 242 242 31 Males # 6 31 24 8 0 % 2.5 12.9 9.9 3.3 0.0 p * * Data on government benefit receipt were not collected from the thirty-four Midwest Study young adults (thirtythree males and one female) who had been incarcerated for at least 3 months at the time of their wave 3 interview. b Parents living with at least one child. c Female parents living with at least one child. 39 Young adults in the Midwest Study were asked about benefit receipt since their last interview, whereas young adults in Add Health were asked about benefit receipt during the past year. For this reason, we limit our comparisons to the current receipt of benefits. Young adults in the Midwest Study were significantly more likely than their Add Health counterparts to be current food stamp recipients. However, the difference was only statistically significant between the females in the Midwest Study and the females in Add Health. By contrast, there was no difference in current TANF receipt between young mothers in the Midwest Study and young mothers in Add Health. This could reflect the fact that the Add Health interviews were conducted in 2001 and 2002, whereas the Midwest Study interviews were conducted in 2006. Although Wisconsin’s average monthly TANF caseload remained relatively flat during those years, both Illinois and Iowa experienced significant caseload declines.9 Table 37. Current Receipt of Government Benefits by Gender: Young Adults in the Midwest Study Compared with Add Health Young Adultsa Food stamps* TANFb a Midwest Study Females Males # % # % 156 50.2 24 9.9 14 8.8 1 0.3 Add Health Females Males # % # % 25 6.3 0 0.0 7 7.5 0 0.0 Data on the receipt of government benefits were not collected from the thirty-four Midwest Study young adults (thirty-three males and one female) who had been incarcerated for at least 3 months at the time of their wave 3 interview. b Parents living with at least one child. * Statistically significant difference between Midwest and Add Health females. 9 Iowa’s average monthly TANF caseload fell 20 percent and Illinois’s average monthly TANF caseload fell 39.5 percent between calendar year 2001 and calendar year 2006 (U.S. Department of Health and Human Services, 2007). 40 PHYSICAL HEALTH AND ACCESS TO HEALTH CARE SERVICES The young adults in the Midwest Study were asked a series of questions about their physical well-being. The vast majority described their health as good to excellent and indicated that they had no chronic conditions or disabilities. Nevertheless, they were more likely than their Add Health counterparts to describe their health as being fair or poor, and to identify themselves as having a disability. Twenty-eight percent of the young adults in our sample reported two or more emergency room visits during the past year, and 19 percent had been hospitalized at least once. Overall, the largest percentage of hospitalizations were pregnancy-related. However, if the hospitalizations of males and females are examined separately, accidents and injuries account for the largest percentage of hospitalizations among the young men (41 percent). Table 38. Health Status at Age 21: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study Add Health (N = 590) (N = 744) # % # % P * Description of general health Excellent 191 32.4 262 35.2 Very good 160 27.1 292 39.2 Good 155 26.3 158 21.2 Fair 72 12.2 30 4.0 Poor 12 2.0 2 0.3 Any chronic medical conditions Yes 76 12.9 No 514 87.1 Missing Health conditions or disability limits daily activitiesa * Yes 65 11.0 35 4.7 No 524 89.0 709 95.3 Don’t know 1 41 Number of ER visits during the past yearb 0 1 2 or 3 4 or more Missing Number of hospitalizations during the past yearb 0 1 2 or more Missing Reason for most recent hospitalization Illness Injury or accident Alcohol or other drug problem Emotional or mental health problem Pregnancy-related a Other 282 141 98 65 4 48.1 24.1 16.7 11.1 476 80 32 2 81.0 13.6 5.4 21 18 1 7 55 18.8 16.1 0.9 6.3 49.1 10 8.9 Add Health question asked whether any health conditions limited ability to engage in moderate activities. b Add Health question asked about ER visits and hospitalization during the past 5 years. We also asked the young adults in the Midwest Study about their ability to access health care services. Only half reported that they currently had medical insurance, and only 39 percent had insurance for dental care. In both cases, most of those who were insured were covered by Medicaid. Sixty percent of these young adults reported that they had a routine physical exam but only 40 percent reported that they had a dental exam during the past year. Overall, about one-fifth of these young adults reported that they had not received medical care and a similar proportion reported that they had not received dental care when they thought they needed it during the past year.10 Not having insurance was the main reason cited for not receiving care.11 10 These percentages were higher among the young adults who were not currently insured. Twenty-eight percent of those who lacked health insurance reported that they had not received medical care when they thought they needed it, and 27 percent of those who lacked dental insurance reported that they had not received dental care when they thought they needed it. 11 We only asked about current insurance coverage. As a result, young adults who currently had insurance could still cite lack of insurance as a reason for not receiving care during the past year. 42 Young adults in the Midwest Study were less likely to have health insurance than young adults in Add Health. Moreover, most of the Midwest Study young adults who had health insurance were covered by Medicaid, whereas most of their insured Add Health peers were covered by their parents’ insurance or an employer-provided plan. Interestingly, despite being more likely to have health insurance, young adults in Add Health were more likely to report that there had been a time during the past year when they did not receive needed medical care. Table 39. Insurance Coverage and Access to Health Care: Young Adults in the Midwest Study Compared with Add Health Young Adults Has medical insurance Source of medical insurance Parents’ insurance Spouse’s insurance Employer-provided insurance School-provided insurance Purchase own private insurance Medicaid or medical assistance Other Missing Last physical exam Less than a year ago 1 to 2 years ago More than 2 years ago Did not receive needed medical care Reason(s) did not receive medical care Didn’t know where to go Cost too much No transportation Hours were inconvenient Would lose pay for missing work No insurance Other Has dental insurance Source of dental insurance Parents’ insurance Spouse’s insurance Employer-provided insurance School-provided insurance Midwest Study N # % 578 293 50.7 Add Health N # % 739 562 76.0 8 7 49 5 5 206 13 2 2.7 2.4 16.7 1.7 1.7 70.3 4.4 272 23 170 14 8 55 25 48.4 4.1 30.2 2.5 1.4 9.8 3.6 588 389 106 94 105 66.0 18.0 16.0 17.9 477 103 152 179 65.2 14.1 20.8 24.1 562 15 78 11 6 12 75 15 223 14.3 74.3 10.5 5.7 11.4 71.4 14.3 39.7 5 8 37 3 2.2 3.6 16.6 1.3 589 43 p * 732 743 * Purchase own private insurance Medicaid or medical assistance Other Last dental exam Less than a year ago 1 to 2 years ago More than 2 years ago Did not receive needed dental care Reason(s) did not receive dental care Didn’t know where to go Cost too much No transportation Hours were inconvenient Would lose pay for missing work No insurance Other 3 157 10 1.3 70.4 4.5 236 170 181 105 40.2 29.0 30.8 20.0 15 78 11 6 12 75 15 14.3 74.3 10.5 5.7 11.4 71.4 14.3 * 423 56.9 321 43.1 MENTAL HEALTH AND UTILIZATION OF MENTAL HEALTH SERVICES We asked the young adults in the Midwest Study about their utilization of mental and behavioral health care services since the last time they were interviewed. Eleven percent had received counseling, 13 percent had received psychotropic medication, and 4 percent had received treatment for a substance abuse problem. By comparison, 7 percent of their Add Health counterparts had received counseling and 2 percent had received treatment for a substance abuse problem during the past year. Table 40. Mental and Behavioral Health Care Services Utilization: Young Adults in the Midwest Study Compared with Add Health Young Adults Received psychological or emotional counseling Attended substance abuse treatment program Received medication for emotional problems Ever hospitalized for mental health problems 44 Midwest Study N # % 589 62 10.5 Add Health N # % 743 54 7.3 589 590 589 744 21 75 74 3.6 12.7 12.5 17 2.3 P * Timing of most recent hospitalization Within the past 3 months 4 to 6 months ago 7 to 9 months ago 10 to 12 months ago More than 1 but less than 2 years ago At least 2 years ago 73 5 5 1 4 12 46 6.8 6.8 1.4 5.5 16.4 63.0 Of course, mental and behavioral health care service utilization does not necessarily reflect mental and behavioral health care service needs. Indeed, one might expect the risk of developing mental health or substance use problems to be especially high among young adults making the transition from foster care to independent living, particularly if they do not have adequate social supports after their discharge (Courtney & Hughes Heuring, 2005; Pecora et al., 2003; Pecora et al., 2005). We assessed both mental health and substance use problems among the young adults in the Midwest Study using the 12-month version of the Composite International Diagnostic Interview (CIDI; World Health Organization, 1998). The CIDI is a highly structured interview, designed for use by nonclinicians, which generates psychiatric diagnoses according to the criteria listed in the Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV). Table 41 shows the percentage of young adults in the Midwest Study who met the criteria for various mental health or substance use disorders during the 12 months prior to their wave 3 interview.12 Results are reported separately for males and females because a number of 12 The percentages are lower than the percentages reported in Courtney et al. (2004) and Courtney et al. (2005). However, the latter were based on the lifetime version of the CIDI, not the 12-month version that was used at wave 3. 45 statistically significant gender differences were found. Young men in the Midwest Study were more than twice as likely to have an alcohol or other drug diagnosis as their female counterparts. By contrast, young women in the Midwest Study were far more likely than their male counterparts to have a diagnosis of depression or post traumatic stress disorder (PTSD). Table 41. 12-Month CIDI Diagnoses by Gender Female (n = 314) # % Alcohol dependence 11 3.5 Alcohol abuse 15 4.8 Any alcohol diagnosis 26 8.3 Other drug dependence 3 1.0 Other drug abuse 6 1.9 Any other drug diagnosis 7 2.2 Any alcohol or other drug diagnosis 29 9.2 a Post traumatic stress disorder (PTSD) 24 7.9 Major depression 24 7.6 Dysthymia 0 0.0 Generalized anxiety disorder 0 0.0 Any mental health disorder 42 14.2 a Male (n = 276) # % 32 11.6 26 9.4 58 21.0 14 5.1 16 5.8 26 5.4 64 23.2 10 3.8 3 1.1 0 0.0 0 0.0 12 4.6 P * * * * * * * * * * PTSD diagnosis was indeterminate for eleven females and ten males because of missing data. SEXUAL BEHAVIORS The young adults in our sample were asked a series of questions about their sexual orientation (Table 42) and sexual behaviors (Tables 43 through 46), including questions related to sexuality, “safe” sex practices, and high-risk behaviors. The vast majority identified themselves as heterosexual, but females were somewhat more likely to identify themselves as either bisexual or homosexual than males. 46 Table 42. Self-Reported Sexual Orientation Female # % 220 80.6 26 9.5 12 4.4 4 1.5 5 1.8 3 1.1 3 1.1 41 100% heterosexual Mostly heterosexual Bisexual Mostly homosexual 100% homosexual Not sexually attracted to males or females Don’t know Missinga a Male # % 217 89.7 10 4.1 2 0.8 2 0.8 5 2.1 2 0.8 4 1.6 34 This includes the twenty-four males and twenty-nine females who did not complete the ACASI portion of the interview. More than 90 percent of the young women and young men in the Midwest Study reported that they had ever had sexual intercourse, and most of those young adults had also had sexual intercourse during the past year. Regardless of gender, nearly 60 percent of the young adults who had sexual intercourse during the past year reported using contraception, and nearly half reported using condoms either all or most of the time. The percentages who reported using contraception or condoms the most recent time they had sexual intercourse were very similar. There were no gender differences in the percentage of young adults who had had sexual intercourse or in their use of condoms and birth control. However, males were significantly more likely than females to report both that they had ever been paid by someone to have sex and that they had ever paid someone to have sex. Table 43. Self-Reported Sexual Behaviors by Gender a Ever had sexual intercourse Had sexual intercourse during past year Used birth control most recent sexual intercourse Used birth control all or most of the time past year Used a condom most recent sexual intercourse Used condoms all or most of the time past year 47 N 283 248 191 187 186 186 Females # 266 194 114 113 85 86 % 94.0 78.2 59.7 60.4 45.2 46.3 Males N # 242 219 205 146 138 78 134 76 142 69 142 65 p % 90.5 71.2 56.5 56.8 47.9 45.8 Any sexual partner had an STD past year Ever paid by someone to have sex Ever paid someone to have sex Ever had sex with injection drug user 181 262 263 260 a 30 19 2 5 16.6 7.3 0.8 1.9 127 214 217 217 12 30 13 3 9.4 14.0 6.0 1.4 * * These figures do not include the twenty-four males and twenty-nine females who did not complete the ACASI portion of the interview. There were a number of statistically significant differences in sexual behaviors between the females in the Midwest Study and their Add Health counterparts. The former were more likely to have ever had sexual intercourse. They were also more likely to have engaged in behaviors that put them at high risk for becoming pregnant and contracting an STD. The only exception was that the females in the Midwest Study were more likely to report using condoms. Table 44. Self-Reported Sexual Behaviors: Females in the Midwest Study Compared with Females in Add Healtha Midwest Study Add Health N # % N # % P Ever had sexual intercourse 283 266 94.0 391 342 87.5 * Had sexual intercourse past year 248 194 78.2 388 322 83.0 Used birth control most recent sexual intercourse 191 114 59.7 320 219 68.4 * Used birth control all or most of the time past year 187 113 60.4 219 213 69.9 * Used a condom most recent sexual intercourse 186 85 45.2 320 123 38.4 * Used condoms all or most of the time past year 186 86 46.3 321 121 37.7 * Any sexual partner had an STD past year 181 30 16.6 313 31 9.9 * Ever paid by someone to have sex 262 19 7.3 342 6 1.8 * Ever paid someone to have sex 263 2 0.8 341 3 0.9 Ever had sex with injection drug user 260 5 1.9 339 9 2.7 a The Midwest Study figures do not include the twenty-nine females who did not complete the ACASI portion of the interview. There were also a number of statistically significant differences in sexual behaviors between the males in the Midwest Study and those in Add Health. The former were more likely to have had sexual intercourse during the past year. They were also less likely to have used 48 birth control when they had sexual intercourse and more likely to have been paid by someone to have sex. Table 45. Self-Reported Sexual Behaviors: Males in the Midwest Study Compared with Males in Add Health Midwest Studya Add Health N # % N # % Ever had sexual intercourse 242 219 90.5 344 299 86.9 Had sexual intercourse past year 205 146 71.2 341 277 81.2 Used birth control most recent sexual intercourse 138 78 56.5 273 183 67.0 Used birth control all or most of the time past year 134 76 56.8 274 186 67.9 Used a condom most recent sexual intercourse 142 69 47.9 275 130 47.3 Used condoms all or most of the time past year 142 65 45.8 278 129 46.2 Any sexual partner had an STD past year 127 12 9.4 269 20 7.4 Ever paid by someone to have sex 214 30 14.0 299 18 6.0 Ever paid someone to have sex 217 13 6.0 299 15 5.0 Ever had sex with injection drug user 217 3 1.4 295 6 2.0 P * * * * a The Midwest Study figures do not include the twenty-four males who did not complete the ACASI portion of the interview. Despite other differences, the two samples were quite similar with respect to median age at first sexual intercourse and median number of sexual partners. Table 46. Median Age at First Sexual Intercourse and Number of Sexual Partners by Gender: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study a Female Male Age at first intercourse Number of lifetime sexual partners Number of sexual partners past year (if sexually active past year) Add Health Female Male n 225 200 Md 16.0 3.0 n 187 160 Md 15.0 6.0 n 342 338 Md 16.0 3.0 n 297 296 Md 16.0 5.0 188 1.0 133 2.0 322 1.0 277 2.0 a The Midwest Study figures do not include the twenty-four males and twenty-nine females who did not complete the ACASI portion of the interview. 49 PREGNANCY Seventy-one percent of the young women in the Midwest Study had ever been pregnant, and half had been pregnant since their most recent interview. Repeat pregnancies were more the rule than the exception among those who had ever been pregnant. By comparison, only onethird of the Add Health females had ever been pregnant, and a majority of those females had been pregnant only once. Table 47. Young Women’s Experiences with Pregnancy: Females in the Midwest Study Compared with Females in Add Health Ever pregnantb Total number of pregnancies One Two or more Pregnant since the last interviewb Midwest Studya n # % 261 185 70.9 181 69 38.1 112 61.9 262 133 50.8 Number of pregnancies since last interview One Two or more Add Health n # % 396 134 33.8 134 74 55.2 60 44.8 P * * 129 93 36 a 72.1 27.9 The Midwest Study figures do not include the twenty-nine young women who did not complete the ACASI portion of the interview. b The Midwest Study figures are based on the responses of the young women who answered the pregnancy questions each time they were interviewed. The vast majority of young women in the Midwest Study who had been pregnant since their last interview had received prenatal care during their most recent pregnancy, and three-quarters of those who received prenatal care did so in their first trimester. Somewhat more concerning, 30 percent of these young women wanted to become pregnant, and only a quarter were using birth control around the time that they conceived. Although some of these young women were still pregnant when they were interviewed, most of their pregnancies had resulted in a live birth. 50 The young women in the Midwest Study were not very different from their Add Health counterparts with respect to the characteristics of their most recent pregnancy. Although they were less likely to have been using birth control around the time that they conceived, they were also less likely to have wanted to become pregnant. Table 48. Characteristics of Most Recent Pregnancy: Females in the Midwest Study Compared with Females in Add Health Received prenatal care Trimester first received prenatal care First Second Third Using birth control at time of conception Wanted to get pregnant by partnera Married at time of conceptionb Outcome of pregnancy Still pregnant Live birth Still birth or miscarriage Abortion a Midwest Study (n = 133) n # % 130 116 89.2 106 80 75.5 19 17.9 7 6.6 125 32 25.6 122 37 30.4 129 11 8.5 133 24 18.0 84 63.2 16 12.0 9 6.8 Add Health (n = 134) n # % 131 107 81.7 90 77 85.6 9 10.0 4 4.4 131 52 39.7 130 59 45.4 p * * 134 21 81 16 16 15.7 60.5 11.9 11.9 Includes females who responded “definitely or probably yes.” Add Health asked the young women if they were married at the time they gave birth. b Half of the young men in the Midwest Study reported that they had ever gotten a female pregnant, compared with 19 percent of their Add Health counterparts. In fact, 38 percent had gotten a female pregnant since their most recent interview. 51 Table 49. Young Men’s Experiences with Pregnancy: Males in the Midwest Study Compared with Males in Add Health Midwest Studya N # % 242 119 49.2 118 86 72.9 20 16.9 12 10.2 Any female partner became pregnant Number who became pregnant 1 2 3 or more Any female partner became pregnant since last interview Number who became pregnant 1 2 3 or more 242 90 a 90 37.2 69 15 6 76.7 16.7 6.7 Add Health N # % 349 67 19.2 P * The Midwest Study figures do not include the twenty-four young men who did not complete the ACASI portion of the interview. The young men who had gotten a female pregnant since their last interview were asked about the most recent pregnancy. The vast majority reported that the female who they had gotten pregnant received prenatal care, generally beginning in their first trimester. A majority of the pregnancies had resulted in a live birth. These young men were less likely than their Add Health counterparts to report that they and their female partner had been using birth control around the time that she conceived but no more likely to report that they had wanted their female partner to become pregnant. Table 50. Characteristics of Most Recent Pregnancy: Males in the Midwest Study Compared with Males in Add Health Midwest Study (N = 90) n # % 84 70 83.3 49 38 77.6 5 10.2 6 12.2 86 12 14.0 Impregnated girl received prenatal care Trimester first received care First Second Third Using birth control at time of conception 52 Add Health (N = 67) n # % 60 48 80.0 64 27 42.2 P * Wanted partner to get pregnant Married to partner at time of conception Outcome of pregnancy Still pregnant Live birth Still birth or miscarriage Abortion Missing 86 88 35 2 40.7 2.3 12 55 12 11 0 13.3 61.1 13.3 12.2 64 28 43.8 8 37 8 13 1 12.1 56.1 12.1 19.7 We also asked the young adults in the Midwest Study about pregnancy prevention. Although their responses varied depending on the wording of the question, only one-third of the females and one-fifth of the males had received either family planning services or information about birth control since their last interview. Table 51. Receipt of Family Planning Services and Birth Control Information Since Last Interview Females Males P n # % n # % Received family planning services 258 35 11.1 233 5 1.8 * Received information about birth control 251 52 19.3 233 45 20.7 Either 276 83 32.2 314 50 21.5 * MARRIAGE, COHABITATION, AND RELATIONSHIPS Nearly one-third of the young women and over one-fifth of the young men in the Midwest Study were either married or cohabiting (i.e., living with a partner in a marriage-like relationship). Although they were as likely to be married or cohabiting as males and females in Add Health, Midwest Study young adults were more likely to be cohabiting than married, whereas Add Health young adults were more likely to be married than cohabiting. Among those who had never been married, young adults in Add Health were more likely to regard marrying some day as very important, although only the difference between female samples was statistically significant. 53 Table 52. Marriage and Cohabitation by Gender: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study Add Health Female Male Female Male n = 314 n = 276 n = 396 n = 347 # % # % # % # % Ever married (ABC) 36 11.5 14 5.1 71 17.9 35 10.1 Currently married (AB) 35 11.1 12 4.3 64 16.2 30 8.6 Currently living with spouse (ABC) 29 9.2 9 3.3 60 15.2 28 8.1 64 22.6 48 18.0 66 16.7 47 13.5 Currently cohabiting Either married or cohabiting (A) 99 31.6 59 21.4 129 32.7 77 22.2 Very important to marry someday (C) 118 42.8 109 41.8 182 56.0 150 48.1 (if never married) A = Statistically significant difference between Midwest Study males and females B = Statistically significant difference between Midwest Study and Add Health males C = Statistically significant difference between Midwest Study and Add Health females Just over half of the young women and young men in the Midwest Study who were neither married nor cohabiting were involved in some type of relationship, and many of them were dating one partner exclusively. Table 53. Other Intimate Partner Relationships by Gender Females (n = 221) Currently involved in a relationship Type of relationship Dating exclusively Dating frequently Dating once in a while Only having sex Missing Males (n = 219) # 124 % 56.1 # 114 % 52.1 93 12 17 1 1 75.6 9.8 13.8 0.8 80 7 16 9 2 71.4 6.3 14.3 8.0 54 CHILDREN AND PARENTING More than half of the young women and nearly one-third of the young men in the Midwest Study had at least one living child at age 21. Nearly all of these young women, but just over one-third of these young men, reported that one or more of their children were living with them. Conversely, two-thirds of these young men, but only 1 in 10 of these young women, reported that one or more of their children were living somewhere else. Both male and female young adults in the Midwest Study were more than twice as likely to have at least one living child as their Add Health counterparts. However, the Add Health males and females were more likely to be living with one or more of their children if they had at least one. They were also less likely to have one or more children living somewhere else. Table 54. Parenthood by Gender: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study Add Health Female Male Female Male # % # % # % # % At least one living child (ABC) 176 56.1 83 30.2 93 23.5 40 11.5 Living with any children (ABC) 160 90.9 30 36.1 93 100 26 65.0 26 14.8 56 67.5 Any nonresident children (ABC) 1 1.1 3 11.5 A = Statistically significant difference between Midwest males and females B = Statistically significant difference between Midwest and Add Health males C = Statistically significant difference between Midwest and Add Health females Most of the young women and young men in the Midwest Study who had at least one living child had only one. Although there was no gender difference in the number of children these young men and women had, the young women had more children of their own living with them. 55 Table 55. Number of Children and Resident Children by Parent Gender Females Number of children 1 2 3 or more Mean number of children Number of “resident” children 0 1 2 3 or more Mean number of resident children* Males # % # % 110 50 16 62.5 28.4 9.6 63 13 7 75.9 15.7 8.4 1.50 16 104 46 10 1.35 9.1 59.1 26.1 6.5 1.30 53 23 5 2 63.9 27.7 6.0 2.4 .47 We asked the young adults whose children were not living with them who the children were living with. Their responses varied greatly by gender. Forty-six percent of the young women reported that at least one nonresident child was living with foster or adoptive parents, compared with only 4 percent of the young men. Conversely, nearly all of the young men reported that at least one nonresident child was living with the child’s other parent, compared with just under one-fifth of the young women. Regardless of gender, a significant minority of these young adults reported that at least one nonresident child was living with grandparents or other relatives. Table 56. Current Living Circumstances and Frequency of Visits with Nonresident Children during the Past Year by Parent Gender Female Male # % # % At least one nonresident child 26 14.7 56 67.4 Has at least one nonresident child living witha Child’s other parent 5 19.2 54 96.4 Maternal grandparents or other maternal relatives 6 23.1 17 30.4 Paternal grandparents or other paternal relatives 5 19.2 1 1.8 Adoptive parents 9 34.6 0 0.0 Foster parents 3 11.5 2 3.6 Other 0 2 3.6 56 Has at least one nonresident child who they visitedb Never Less than once a month Once a month Two or three times a month Once a week Every day a 8 2 1 4 9 4 30.8 7.7 3.8 15.4 34.6 15.4 7 13 5 6 11 19 12.5 23.2 8.9 10.7 19.6 33.9 Percentages sum to more than 100 because some children were living with more than one other person (e.g., other parent and maternal grandparents) and because children with the same parent could be living with different people. b Percentages sum to more than 100 because parents with more than one nonresident child could visit them with different frequencies. Relatively few of the young adults who had at least one child reported that a child had health problems or disabilities. Table 57. Child Well-Being by Parent Gender n At least one living child One or more resident children Any child fair or poor health Any resident child fair or poor health Any child learning disability Any resident child learning disability Any child disability limits activities Any resident child disability limits activities 167 158 165 158 167 158 Female # % 176 56.1 160 50.8 11 6.6 10 6.3 13 7.9 11 7.0 12 7.2 12 7.6 N 82 31 81 31 81 31 Male # 83 30 5 1 3 1 5 1 % 30.2 11.2 6.1 3.2 3.7 3.2 6.2 3.2 We asked the young parents who were working or in school a number of questions about childcare. Well over half reported that the child(ren)’s other parent or other relative provided childcare. Another 27 percent relied on more formal providers, including day care centers, nursery schools, and pre-K. Nearly two-thirds of these young parents reported that finding someone to care for their children was not difficult at all, and three-quarters had not changed childcare providers within the past 6 months. Although just over one-third of these young parents were receiving any childcare assistance, half reported that they paid nothing out of 57 pocket for their childcare. This probably reflects the fact that their children were often being cared for by the other parent or a relative. Table 58. Childcare among Parents Currently Working or in School (N = 104)a # % Childcare provider while working or going to school Other parent 25 24.3 Grandparent 24 23.3 Other relative 12 11.7 Neighbor or baby-sitter 7 6.8 Day care center, nursery school, or pre-K 28 27.2 Other 7 6.8 Missing 1 Difficulty of finding someone to care for child(ren) while working or going to school Very difficult Somewhat difficult Not at all difficult Missing Times missed work or school during the past 6 months because of lack of childcare Never Once or twice Three or more times Missing Times changed childcare providers during the past 6 months Never Once or twice Three or more times Missing Currently receiving childcare assistance from government agency (missing = 5) Usual weekly out-of-pocket cost for childcare (not counting any childcare assistance) $0 $1 - $50 $51 - $100 More than $100 Missing a 11 25 67 1 10.7 24.3 65.0 63 27 10 4 63.0 27.0 10.0 78 19 5 2 76.5 18.6 4.9 35 35.4 43 19 12 11 19 50.6 22.3 14.1 12.9 Data were missing for an additional eighteen parents (fifteen female and three male) who did not complete the ACASI portion of the interview. 58 Many of these young parents identified their biological mother or another relative as both a source of information about parenting and someone who had taught them how to be a good parent. Others identified their foster mother or a friend. No statistically significant gender differences were found. Table 59. Information about Parenting (N = 172) Received information about parenting from Biological mother Biological father Foster mother Foster father Grandparent Other relative Friend Social worker/caseworker Book/parenting magazine Parenting class Other Missing a Learned how to be a good parent from Biological mother Biological father Foster mother Foster father Grandparent Other relative Friend Social worker/caseworker Book/parenting magazine Parenting class Other Missing a a # % 29 3 18 1 30 32 21 1 4 4 17 30 18.1 1.9 11.3 0.6 18.8 20.0 13.1 0.6 2.5 2.5 10.6 25 3 24 2 26 25 6 1 6 10 32 30 15.6 1.9 15.0 1.3 16.3 15.6 3.8 0.6 3.8 6.3 20.0 Includes the eighteen parents (fifteen female and three male) who did not complete the ACASI portion of the interview 59 We asked these young parents a series of nine questions designed to measure their level of parenting stress.13 For each question, parents indicate how frequently their child causes them to feel a particular way, using a 5-point scale that ranges from 1 = “not at all” to 5 = “very true.” Parents who had more than one child living with them were instructed to think about the eldest. A parenting stress score was constructed by summing their responses to these questions and taking the mean. The scale exhibited good reliability (alpha = .78), meaning that all of the items seem to be measuring the same underlying construct. In general, these young parents were not experiencing high levels of parenting stress. Their mean score on the scale was 1.58 out of a possible 5, with 5 corresponding to high levels of stress. With only one exception, a majority responded “not at all” to each of the items. Nevertheless, most also acknowledged that being a parent was harder than they had expected. There was no difference in scores on the parenting stress scale between the young women (mean = 1.6) and the young men (1.4). Table 60. Parenting Stressa N Feel I am giving up my life to meet my child’s needs Not at all true Moderately or a little true Mostly or very true Feel trapped by my responsibilities as a parent Not at all true Moderately or a little true Mostly or very true Taking care of my child is more work than pleasure Not at all true Moderately or a little true Mostly or very true Child seems much harder to care for than most 13 # % 96 39 24 60.4 24.5 15.1 119 32 12 73.0 19.6 7.4 102 45 14 63.4 28.0 8.7 P 159 163 161 162 This scale has been used in studies of other low-income parents (Bos, Polit, & Quint, 1997; Courtney et al., 2005; Dworsky et al., 2007; Huston et al., 2003). 60 Not at all true Moderately or a little true Mostly or very true Child does things that really bother me a lot Not at all true Moderately or a little true Mostly or very true 134 22 6 82.7 13.6 3.7 97 58 8 59.5 35.6 4.9 163 Sometimes lose patience with child Not at all true Moderately or a little true Mostly or very true Often feel angry with my child Not at all true Moderately or a little true Mostly or very true Being a parent is harder than expected Not at all true Moderately or a little true Mostly or very true Child has been a lot of trouble to raise Not at all true Moderately or a little true Mostly or very true Mean * 165 122 43 0 73.9 26.1 0.0 134 28 2 81.7 17.1 1.2 46 72 46 28.0 43.9 28.0 134 31 0 81.2 18.8 0.0 164 164 165 1.58 a Data were missing for the eighteen parents (fifteen female and three male) who did not complete the ACASI portion of the interview. We also administered the revised Child Parent Conflict Tactics Scale (Strauss et al., 1998). This measure has been used in many studies to assess the extent to which parents employ various modes of discipline (i.e., nonviolent discipline, psychological aggression, minor physical assault, severe physical assault, and very severe physical assault) with their children. Parents are asked to rate how frequently they have taken twenty-two specific actions to discipline their child during the past year, using a 7-point scale that ranges from 0 = “never” to 6 = “more than 20 times.” 61 Because we were concerned that young parents in the Midwest Study might be reluctant to report some of the actions they had taken to discipline their child, we included the Child Parent Conflict Tactics Scale items in the ACASI portion of the interview. Although some of these disciplinary actions may still have been underreported, the Audio CASI interview format should have reduced that possibility. Table 61 shows the percentage of young parents in the Midwest Study who reported taking a specific action to discipline their child during the past year.14 These young parents were most likely to report using nonviolent modes of discipline as well as “shouting, screaming, or yelling.” However, the percentages were consistently higher for the young women than the young men. The most common type of physical discipline, spanking a child with a bare hand, was reported by nearly half of the young women and one-third of the young men. Very few of these young parents reported using the more severe types of physical discipline. Table 61. Disciplinary Actions Taken during the Past 12 Months by Parent Gender a Nonviolent Discipline Explained why something was wrong* Put child in a time out or sent child to room* Took away privileges or grounded child Gave child something else to do* Psychological Aggression Threatened to spank or hit child but didn’t do it Shouted, screamed, or yelled at child* Swore or cursed at child* Called child dumb or lazy or some other name Threatened to send child away or kick him or her out of the house 14 n Female # n Male # % % 126 134 143 130 94 94 51 95 74.6 70.1 35.7 73.1 22 26 27 23 11 12 7 11 50.0 46.2 25.9 47.8 142 136 141 142 79 97 40 7 55.6 71.3 28.4 4.9 27 26 27 27 3 12 2 1 12 46.2 7.4 3.7 143 7 4.9 26 0 0.0 The seven categories were never, once, twice, three to five times, six to ten times, 11 to 20 times, and more than 20 times. As recommended by Strauss et al. (1998), medians were calculated using the midpoint of the category for categories 4 through 6 and using 25 for the last category. 62 Minor Physical Assault Spanked child on the bottom with a bare hand Hit child on the bottom with a belt or hard object Slapped child on the hand, arm, or leg Pinched child Shook child (if child > 2 years old) Severe Physical Assault Slapped child on the face, head, or ears Hit child somewhere other than on the bottom with a belt or hard object Threw or knocked child down Hit child with a fist or kicked the child hard Very Severe Physical Assault Beat child over and over Grabbed child around the neck and choked him or her Burned or scalded child on purpose Threatened child with a knife or gun Shook child (if child < 2 years old) 138 138 141 143 12 64 26 49 12 0 46.4 18.8 34.8 8.4 3.6 27 27 27 26 11 0 9 2 10 1 2 33.3 7.4 37.0 3.8 18.2 139 10 7.2 26 0 0.0 142 143 143 5 6 3 3.5 4.2 2.1 27 27 27 1 0 1 3.7 0.0 3.7 143 1 0.7 27 0 0.0 143 2 1.4 27 0 0.0 143 1 0.7 27 0 0.0 143 2 1.4 27 0 0.0 59 4 6.8 18 0 0.0 a Data were missing for the eighteen parents (fifteen female and three male) who did not complete the ACASI portion of the interview. The revised Child Parent Conflict Tactics Scale also includes five items designed to measure parental neglect. Parents use the same 7-point scale to rate how frequently they engaged in a particular neglectful behavior. Most of the young parents in the Midwest Study had not engaged in any of these behaviors according to their self-reports. Table 62. Neglectful Behaviors during the Past 12 Months by Parent Gender a Left child home alone even when some adult should be with him or her Not able to show or tell child you loved him or her due to being so caught up with own problems Not able to make sure child was fed Not able to make sure child got to a doctor or hospital Problem taking care of child due to being drunk or high n Female # % 142 3 2.1 n Male # % 27 0 0 139 16 11.5 27 1 141 8 5.7 26 1 139 4 2.9 26 1 141 0 0.0 26 0 a Data were missing for the eighteen parents (fifteen female and three male) who did not complete the ACASI portion of the interview. 63 3.7 3.8 3.8 0.0 CRIMINAL BEHAVIOR AND CRIMINAL JUSTICE SYSTEM INVOLVEMENT We asked the young adults in the Midwest Study a series of questions about their participation in criminal behaviors during the past 12 months and then compared their responses with the behaviors reported by the nationally representative sample of young adults who participated in Add Health. In general, males were more likely to report engaging in these behaviors than were females, and nearly all of these gender differences were statistically significant. Young men in the Midwest Study were most likely to report belonging to a gang and taking part in a group fight; young women were most likely to report belonging to a gang and deliberately damaging someone else’s property. Where statistically significant differences between the young men in the Midwest Study and their Add Health counterparts were found, the criminal behaviors were more likely to have been reported by the former foster youth. The only statistically significant difference between young women in the Midwest Study and young women in Add Health is that the former were more likely to report having pulled a knife or gun on someone. Table 63. Self-Reported Criminal Behavior by Gender: Young Adults in the Midwest Study Compared with Add Health Young Adults Males p Females Midwest Add Midwest Add Study Health Study Health (n = 347) (n = 285)a (n = 396) (n = 223)a # % # % # % # % Ever belonged to a named gang 53 23.8 52 15.0 * 28 9.8 55 13.9 Took part in a fight involving one group against another 46 20.6 74 21.3 17 6.0 16 4.1 Deliberately damaged someone’s property 38 17 52 15.0 25 8.8 21 5.3 Hurt someone so badly in a fight that medical treatment was required 35 15.7 51 14.7 10 3.5 8 2 Sold marijuana or other drugs 32 14.3 44 12.7 14 4.9 16 4.0 Stole something worth < $50 25 11.2 41 11.8 10 3.5 18 4.5 64 p Bought, sold, or held stolen property Stole something worth > $50 Entered a house or building to steal something Became so injured in a fight that medical treatment was required Deliberately wrote a bad check Used a weapon in a fight Pulled a knife or gun on someone Used or threatened to use a weapon to get something from someone Carried a handgun to school or work Used someone’s credit card or bank card without their permission or knowledge Shot or stabbed someone a 22 20 9.9 9.0 25 13 7.2 3.7 * 6 8 2.1 2.8 4 9 1.0 2.3 14 6.3 7 2.0 * 3 1.1 3 0.8 14 13 13 13 6.3 5.8 5.8 5.8 26 15 12 8 7.5 4.3 3.5 2.3 * 8 17 9 12 2.8 6.0 3.2 4.2 7 17 6 2 1.8 4.3 1.5 0.5 7 6 3.1 2.7 10 9 2.9 2.6 1 0 0.4 0.0 3 2 0.8 0.5 3 2 1.3 0.9 8 2 2.3 0.6 0 3 0.0 1.1 3 1 0.8 0.3 * Data were missing for the fifty-three young men and twenty-nine young women who were incarcerated and/or did not complete the ACASI portion of the interview. While not all criminal behavior results in criminal justice system involvement, young adults in the Midwest Study reported a high level of involvement with the criminal justice system since their most recent interview. Thirty one percent reported being arrested, 15 percent reported being convicted of a crime, and 30 percent reported being incarcerated. However, the level of criminal justice involvement was significantly higher among the young men. We asked those who were arrested, convicted, or incarcerated whether this was the result of a violent crime, a property crime, or a drug-related crime. The response categories were neither mutually exclusive nor exhaustive. For example, young adults could report being arrested for more than one type of crime or, alternatively, could report that the crime they were arrested for did not fall into any of the three categories. There were significant gender differences with respect to the types of crimes that led to their criminal justice system involvement. Males were more likely to be arrested, convicted, and 65 incarcerated for drug-related and violent crimes than for property crimes; females were more likely to be arrested, convicted, and incarcerated for property and violent crimes than for drugrelated crimes. However, a fairly large percentage of the criminal justice system involvement that these young adults reported was for other reasons, such as probation violations or trafficrelated offenses.15 Table 64. Self-Report of Arrest, Conviction, and Incarceration Since Last Interview Males Females Total (n = 287) a (N = 544) a (n = 257) a # % # % # % Arrested since last interview bc 113 45.7 52 18.6 165 31.3 Arrested for violent crime 20 17.7 6 11.5 26 15.8 Arrested for property crime 10 8.8 7 13.5 17 10.3 Arrested for drug-related crime 26 23.0 4 7.7 30 18.2 Convicted of a crime since last interview bc Convicted of violent crime Convicted of property crime Convicted of drug-related crime 60 16 7 15 24.0 26.7 1217 25.0 20 3 4 2 7.1 15.0 20.0 10.0 80 19 11 17 15.1 23.8 13.8 21.3 Spent at least one night in jail, prison, other correctional facility since last interview bc Incarcerated for violent crime Incarcerated for property crime Incarcerated for drug-related crime 111 26 17 25 44.6 23.4 15.3 22.5 46 11 10 7 16.4 23.9 21.7 15.2 157 37 27 32 29.7 23.6 17.2 20.2 a p * * Data were missing for nineteen nonincarcerated young men and twenty-seven nonincarcerated young women who did not complete the ACASI portion of the interview. b Five incarcerated young men and two incarcerated young women who did not complete the ACASI portion of the interview were coded as having been arrested, having been convicted and having been incarcerated since their most recent interview. c Data on arrests were missing for eleven young men and six young women, data on convictions were missing for eight young men and six young women and data on incarcerations were missing for nine young men and six young women who did complete the ACASI portion of the interview. Although there were few differences between young adults in the Midwest Study and their Add Health counterparts with respect to self-reported criminal behaviors, both males and females in the Midwest Study reported significantly higher levels of criminal justice system 15 In fact, preliminary analysis of official arrest data suggests that many arrests are for traffic-related offenses or probation violations. 66 involvement than males and females in Add Health. Specifically, they were more likely to report ever being arrested, ever being convicted, and ever being arrested as an adult. In fact, females in the Midwest Study were significantly more likely than males in Add Health to report ever being arrested (57% vs. 20%), ever being convicted (25% vs. 12%), and ever being arrested as an adult (33% vs. 8%). Table 65. Self-Reported Arrests and Convictions by Gender: Young Adults in the Midwest Study Compared with Add Health Young Adults Ever arrested ab Arrested as an adultabc Ever convicted ab Convicted as an adultabc a Males Midwest Add Study Health (n = 348) (n = 270) # % # % 212 79.4 70 20.1 150 59.3 26 7.5 133 52.6 42 12.1 98 39.7 36 10.3 p * * * * Females Midwest Add Study Health (n = 297) (n = 396) # % # % 165 56.7 17 4.3 90 32.5 2 .5 68 24.5 5 1.3 38 13.9 5 1.3 p * * * * Data on “arrested ever” were missing for ten young men and twenty-two young women, data on “arrested as adult” were missing for twenty-four young men and thirty-six young women, data on “convicted ever” were missing for twenty-four young men and thirty-five young women and data on “convicted as an adult” were missing for thirty young men and forty young women. b Five incarcerated young men and two incarcerated young women who did not complete the ACASI portion of the interview were coded as having been arrested and having been convicted both ever and as an adult. c .The Add Health figures reflect arrests and convictions since age 18. The Midwest Study figures represent arrests and convictions since the wave 1 interview, when 62 percent of the young adults in the wave 3 sample were still 17 years old. VICTIMIZATION Young adults in the Midwest Study were asked two sets of questions about victimization they may have experienced since their last interview. The first set of questions focused on violent crime. Twenty six percent of the males and 10 percent of the females reported having been a victim of a violent crime.16 Generally speaking, the young adults in the Midwest Sample reported rates of victimization similar to those reported by their counterparts in Add 16 These percentages do not include the young adults who saw someone shot or stabbed. 67 Health. However, males in the Midwest Study were more likely to report being cut or stabbed by someone and seeing someone shot or stabbed. Females in the Midwest Study were more likely to report seeing someone shot or stabbed and being beaten up with nothing stolen. Table 66. Self-Report of Victimization by Gender: Young Adults in the Midwest Study Compared with Add Health Young Adults Males Females Midwest Add Health p Midwest Add Health p Study Study (n = 348) (n = 285)a (n = 396) (n = 252)a # % # % # % # % Saw someone being shot or stabbed 39 15.5 33 9.5 * 19 6.7 11 2.8 * Someone pulled a knife on you 39 15.5 36 10.3 10 3.5 9 2.3 Someone pulled a gun on you 27 10.7 26 7.5 9 3.2 9 2.3 Shot by someone 6 2.4 2 .6 2 .7 2 .5 Cut or stabbed by someone 9 3.6 4 1.1 * 4 1.4 3 .8 Beaten up with nothing stolen 16 6.3 14 4.0 17 6.0 11 2.8 * Beaten up and belongings stolen 8 3.2 5 1.4 6 2.1 3 .8 a Data were missing for the twenty-four young men and twenty-nine young women who did not complete the ACASI portion of the interview. The second set of questions dealt with sexual victimization. Seven items adapted from the Lifetime Experiences Questionnaire (Rose, Abramson, & Kaupie, 2000) were used. Each item describes a specific way in which someone could be sexually victimized. Young adults in the Midwest Study were asked if they had experienced each type of sexual victimization since their last interview. There was little difference in the incidence of sexual victimization between the young women and the young men. Nine percent of the young women and 7 percent of the young men reported that they had experienced at least one of the seven types of sexual victimization about which they were asked. 68 Table 67. Sexual Victimization Since Last Interview Femalesa N # Male inserted sexual body part inside 279 12 private sexual part, anus, or mouth when not desired Individual inserted fingers or objects inside 277 8 private parts or anus when not desired Individual put their mouth on private parts 276 5 when not desired Individual touched private sexual parts 278 9 when not desired Coerced to touch an individual’s private 275 7 sexual parts Individual touched other private sexual 275 13 parts when not desired Female put private sexual part inside her -body when not desired Experienced any of the above 24 a % N Malesa # 4.3 247 3 1.2 2.9 248 4 1.6 1.8 248 5 2.0 3.2 248 8 3.2 2.5 248 2 .8 4.7 249 7 2.8 -- 247 18 7.4 8.8 % P * Data were missing for the twenty-four young men and twenty-nine young women who did not complete the ACASI portion of the interview. CIVIC PARTICIPATION We asked the young adults in the Midwest Study a series of questions about their civic participation that young adults in the Add Health Study had also been asked. Young adults in the Midwest Study were less likely than their Add Health counterparts to report performing any unpaid volunteer or community service work during the past 12 months. Those who did perform any unpaid volunteer or community service work were most likely to have done something involving church groups, community centers, or youth organizations. There was relatively little difference between the two samples in their level of political participation. Nearly three-quarters of the young adults in both samples were registered to vote, but less than half had voted in the 2004 election. Very few young adults in either sample had contributed 69 money to a political party or candidate, contacted a government official, or attended a political rally. Table 68. Civic Participation during Past 12 Months: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study Add Health p (N = 590) (N = 744) # % # % Performed unpaid volunteer or community service 120 20.3 217 29.2 * Type of service performed: Youth organizations (e.g., Scouts) Service organizations (e.g., Big Brothers) Political clubs or organizations Ethnic-support groups (e.g., NAACP) Church groups Community centers Social action groups Educational organizations Environmental groups (e.g., Sierra Club) (n = 120) 28 23.3 15 12.5 6 5.0 6 5.0 41 34.2 34 28.3 9 7.5 22 18.3 12 10.0 (n = 217) 59 27.3 29 13.4 17 7.8 11 5.1 73 33.6 65 30.0 37 17.1 63 29.0 18 8.3 Registered to vote Voted in 2004 presidential election Contributed money to political party or candidate Contacted government official Attended a political rally or march 413 255 14 18 24 550 309 12 20 23 70.0 43.2 2.4 3.1 4.1 73.9 41.5 1.6 2.7 3.1 Young adults in the Midwest Study were also asked about their political beliefs. Compared with their Add Health counterparts, young adults in the Midwest Study were less likely to report trusting the government and more likely to be uncertain or ambivalent about their political ideology and party identification. 70 Table 69. Political Beliefs and Identification: Former Foster Youth Compared with Add Health Young Adults Midwest Study Add Health p (N = 590) (N = 744) # % # % Strongly agree or agree: I trust the federal government 184 30.2 439 45.6 * I trust my state government 234 39.5 371 49.8 * I trust my local government 239 40.5 356 47.9 * Political ideology Very conservative Conservative Middle-of-the-road Liberal Very liberal Don’t know/refused/NA * 21 123 193 82 34 137 Political party identificationa None Democrat Republican Other a 3.6 20.8 32.7 13.9 5.8 23.2 21 114 406 118 21 64 2.8 15.3 54.6 15.9 2.8 8.6 * 428 124 21 8 72.5 21.0 3.6 1.4 486 134 102 10 65.3 18.0 13.7 1.3 Add Health percentages may not add up to 100% because of a small amount of missing data. RELIGION Young adults in the Midwest Study were asked about their religious faith and participation. They were much less likely to have attended religious services during the past 12 months than their Add Health counterparts, but just over half of both samples reported that their religious faith was at least very important. In fact, young adults in the Midwest Study were more likely to report that their religious faith was more important than anything else. 71 Table 70. Religious Participation and Faith: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study Add Health (N = 590) (N = 744) # % # % Number of times attended a religious service during the past 12 monthsa Never 255 43.2 214 28.8 A few times 145 24.6 186 25.0 Several times 54 9.2 92 12.4 Once a month 17 2.9 48 6.5 Two or three times a month 44 7.5 73 9.8 Once a week 41 6.9 91 12.2 More than once a week 32 5.4 34 4.6 Importance of religious faitha Not important Somewhat important Very important More important than anything else a p * * 75 199 222 90 12.7 33.7 37.6 15.3 112 235 329 62 15.1 31.6 44.2 8.3 Add Health percentages may not add up to 100% because of a small amount of missing data. FEELINGS ABOUT THE TRANSITION TO ADULTHOOD The transition from adolescence to adulthood has become longer, more complex, and less orderly (Setterstein et al., 2005). Because much of the research on this transition has focused on youth in the general population, less is known about how it is experienced by vulnerable populations such as youth exiting foster care. For this reason, we asked the young adults in the Midwest Study a series of questions about how they experienced the transition to adulthood and compared their responses to the responses of their peers in Add Health. Approximately two-thirds of the young adults in the Midwest Study thought they became socially mature and took on adult responsibilities faster than others their age. In this respect, they were not very different from their Add Health peers. They were, however, less likely than 72 their Add Health peers to think that they became socially mature and took on adult responsibilities more slowly than others their age, and they were more likely to think of themselves as being adults most or all of the time. Table 71. Experiences with the Transition to Adulthood: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study Add Health P (N = 590) (N = 744) # % # % Became socially mature * Faster than others 386 66.2 473 63.7 About the same rate as others 165 28.3 59 8.0 Slower than others 32 5.5 210 28.3 Missing 7 2 Took on adult responsibilities * Faster than others 398 67.9 506 68.2 About the same rate as others 155 26.5 54 7.3 Slower than others 33 5.6 182 24.5 Missing 4 2 Think of self as an adult * Never or seldom 29 4.9 69 9.3 Sometimes 51 8.7 135 18.2 Most or all of the time 507 86.3 539 72.6 Missing 3 1 LIFE SATISFACTION AND FUTURE ORIENTATION We also asked the young adults in the Midwest Study a series of questions about their lives and their futures. Slightly more than three-quarters reported feeling satisfied or very satisfied with their lives as a whole. More than half reported that life has been better or much better since they exited foster care; relatively few reported that it had gotten worse or much worse. Most also reported feeling fairly to very optimistic about their futures. 73 Table 72. Life Satisfaction Satisfaction with life as a whole Satisfied or very satisfied Neither satisfied nor dissatisfied Dissatisfied or very dissatisfied Missing Life since exiting foster care Better or much better Sometimes better/sometimes worse Worse or much worse Optimism about the future Very optimistic Fairly optimistic Not very or not at all optimistic N 586 # % 443 89 54 2 75.6 15.2 9.2 327 215 43 55.9 36.8 7.4 322 191 70 55.2 32.8 12.0 583 Another way of looking at the direction in which these young adults think their lives are headed is to consider their responses to a set of questions that asked them to rate their likelihood of experiencing a particular event. Responses could range from 1 = almost no chance to 5 = almost certain. Although young adults in the Midwest Study were relatively optimistic about their prospects for the future, they were significantly less optimistic than their Add Health counterparts. Table 73. Orientation toward the Future: Young Adults in the Midwest Study Compared with Add Health Young Adults Midwest Study Add Health p N Mean S.D. N Mean S.D. * Live to 35 584 4.4 .79 741 4.7 .62 * Divorced by 35 566 1.8 1.1 719 1.6 .94 * Married within the next 10 years 536 3.4 1.4 644 3.9 1.1 * Middle-class income by age 30 583 3.6 1.1 724 4.1 .99 * More than middle-class income by age 30 580 3.3 1.2 735 3.5 1.1 * MENTORING We asked the young adults in the Midwest Study about mentoring relationships they may have had. Although a majority of these young adults reported that they had maintained a 74 positive relationship with a caring adult since age 14, they were less likely to do so than their Add Health counterparts. Young adults in both samples who did have a mentor were most likely to describe their mentor as a friend, a family member, or a teacher/counselor/coach. Seventy percent of the young adults in the Midwest Study who had a mentor reported that they still had telephone or e-mail contact with their mentor at least once a month, and more than half had in-person contact that frequently. Given this level of contact, it is probably not surprising that nearly three-quarters of the Midwest Study young adults who had a mentor felt quite or very close to him or her. Table 74. Mentoring Relationships Midwest Study (n = 590) # % Maintained a positive relationship with a caring adult since age 14 Relationship to mentor Sibling Grandparent or uncle/aunt Teacher, counselor, coach Clergy member Employer or co-worker Friend Neighbor or parent of friend Volunteer from mentoring program (e.g., Big Brothers, Big Sisters) Social worker Other E-mail or telephone contact with mentor Not at all to once a year or less Every few months Monthly to every few weeks Weekly or more Add Health (n = 744) # % 352 60.3 572 77.4 18 93 46 6 6 117 20 5.2 27.1 13.3 1.7 1.8 34.0 5.8 71 120 148 23 42 88 31 12.4 21.0 26.0 4.0 7.3 15.4 5.4 18 5.2 0 0 20 0 5.8 - 3 45 0.5 7.9 66 39 50 195 18.9 11.1 14.3 55.7 172 67 71 226 32.1 12.5 13.2 42.2 P * * * * * * * * * 75 In-person contact with mentor Not at all to once a year or less Every few months Monthly to every few weeks Weekly or more * 103 43 55 149 Closeness to mentor Not at all to a little close Somewhat close Very or quite close 29.4 12.3 15.7 42.6 162 98 52 228 30.0 18.1 9.6 42.2 * 48 46 253 13.8 13.3 72.9 121 130 290 22.4 24.0 53.6 CONNECTEDNESS Because youth aging out of foster care have been identified as being at high risk of becoming disconnected young adults, that is, neither working nor enrolled in school (Haveman & Wolfe, 1994; Levin-Epstein & Greenberg, 2003; Sheehy et al., 2001; Sum et al., 2003; Youth Transition Funders Group, 2004), we looked at the percentage of young adults in the Midwest Study who were connected to employment or to education at ages 19 and 21. Females were more likely to be connected and experienced a greater increase in connectedness over time. At age 19, 54 percent of the males and 59 percent of the females were either working or enrolled in school. By age 21, these figures had risen to 60 percent and 69 percent, respectively. Although many young adults combine work or school with parenthood, we broadened our definition of connectedness to include young adults who were parenting (i.e., living with one or more of their own children). With this more inclusive definition, the percentage of young women who were connected was considerably higher at both points in time. It has a much 76 smaller effect on connectedness among the young men, because they were much less likely to be parenting even if they had a child. Table 75: Connected at Ages 19 and 21 by Gender (N = 590) Females Age 19 Age 21 # % # % Working or enrolled in school 164 58.6 194 69.3 Working, enrolled in school, or 215 76.8 247 88.2 parenting Males Age 19 Age 21 # % # % 125 53.9 138 59.5 127 54.7 147 63.4 TRENDS OVER TIME We have been tracking the outcomes of the young adults in the Midwest Study between the ages of 17 or 18 and age 21. An obvious question is whether any trends can be observed. To address this question we limited our analysis to the 512 young adults (70 percent of the original sample) who were interviewed at all three waves. We selected nine outcomes that are often cited as important markers during the transition to adulthood. Figures 1 through 6 demonstrate how the situation of these young adults has changed over time with respect to each outcome. Because one would expect to find gender differences in some of these outcomes, results are shown separately for females and for males. Trends in school enrollment and educational attainment. There was a large increase in the percentage of young adults who had a high school diploma or GED over time (See Figures 1 and 2). By age 21, 81 percent of the young women and three quarters of the young men had a high school diploma or a GED. There was a much smaller increase (in absolute terms) in the percentage of young adults who had ever attended college, 77 and the percentage enrolled in college peaked at age 19. Young women (38%) were more likely than young men (23%) to have ever attended college by age 21. The percentage of young adults enrolled in school or in a training program fell over time, but the decline was considerably larger among the young men. Figure 1. Trends in Young Women's School Enrollment and Educational Attainment 90 81.4 80.4 80 Percentage of Females 70 66.8 60 48.2 50 37.8 40 35 27.5 30 20 25 19.3 11.8 10 6.8 1.2 0 High school diploma or GED Any college Currently enrolled in school Currently enrolled in college or training program Wave 1 78 Wave 2 Wave 3 Figure 2. Trends in Young Men's School Enrollment and Educational Attainment 90 84.1 80 75.0 Percentage of Males 70 58.6 60 50 44.4 40 30 20 23.3 23.3 18.5 16.4 13.4 9.1 10 6.9 1.3 0 High school diploma or GED Any college Currently enrolled in school Currently enrolled in college or training program Wave 1 Wave 2 Wave 3 Trends in employment There was an increase in the percentage of both young women and young men who were currently employed over time (See Figure 3). Notwithstanding this increase, a significant percentage of these young adults were still not working at wave 3 when they were 21 years old. 79 Figure 3. Trends in Current Employment by Gender 60 57.1 49.6 Percentage of Males 50 40.7 39.7 40 41.4 33.6 30 20 10 0 Male Female Wave 1 Wave 2 Wave 3 Trends in family formation. Only a small percentage of these young adults were married by age 21, and females were more likely to be married than males at both waves 2 and 3 (See Figures 4 and 5). Cohabitation was much more common, especially at wave 3, when 32 percent of the young women and 22 percent of the young men were married or cohabiting.17 About 20 percent of the young women and 6 percent of the young men were already parents at wave 1. Those percentages had more than doubled for the young women to 55 percent and more than quadrupled for the young men to 29 percent by age 21. 17 We do not have information about cohabitation at wave one. 80 Figure 4. Trends in Young Women's Family Formation 60 55 Percentage of Females 50 40 32.5 32.1 30 20.4 20 14.3 11.1 10 3.6 0 Married Married/Cohabiting Wave 1 81 Wave 2 Wave 3 Has at least one child Figure 5. Trends in Young Men's Family Formation 35 29.3 30 Percentage of Males 25 22.0 20 15 12.5 10 5.6 5.2 5 3.9 0 Married Married/Cohabiting Wave 1 Wave 2 Has at least one child Wave 3 Trends in criminal justice system involvement Both young women and young men were much more likely to have been arrested prior to their wave 1 interview than between their wave 1 interview and their interview at age 19 (See Figure 6). They were even less likely to have been arrested between their interviews at age 19 and age 21. Although young adults of both genders were also more likely to have been incarcerated prior to their wave 1 interview than between their wave 1 interview and their interview at age 19, this downward trend did not continue between waves 2 and 3.. 82 Figure 6. Trends in Criminal Justice Involvement by Gender 70 60 60.3 Percentage of Males 50 42.7 40 40 38.8 35.8 31 30 26.6 24.3 19.6 20 16.8 14.6 11.1 10 0 Males Arrested Females Arrested Wave 1 Males Incarcerated Wave 2 Females Incarcerated Wave 3 DISCUSSION AND NEXT STEPS How should the descriptive findings presented here be interpreted? Are they evidence of the need for a call to action on behalf of foster youth making the transition to adulthood or, alternatively, are the outcomes of these young people all that should be expected given the difficulties they experienced prior to entering out-of-home care? We believe that our findings illustrate the inadequacy of current efforts to prepare young people in state care for a successful transition to adulthood. If the outcomes of these young adults were assessed through the same lens that most U.S. parents would use to view the progress of their own children, the findings presented here should be very troubling. On many dimensions that would be of concern to the typical parent, these young people are faring poorly as a group. In comparison with their peers, 83 they are, on average, less likely to have a high school diploma, less likely to be pursuing higher education, less likely to be earning a living wage, more likely to have experienced economic hardships, more likely to have had a child outside of wedlock, and more likely to have become involved with the criminal justice system. However, concern about these young adults should not be based solely on how they are faring relative to their peers. Rather, we should also be concerned about the large percentage who have experienced outcomes that do not bode well for their future or the future of their children. For example, too many of the young men have been involved with the criminal justice system as adults. Similarly, too many of the young women who are raising children are doing so on their own and are dependent on needs-based government support. To be sure, the young adults who have been participating in the Midwest Study are not a monolithic group. Some have made significant progress toward self-sufficiency. They are working or continuing their education. They have a stable place to live and have avoided both criminal justice system involvement and early parenthood. It is also important to acknowledge the strengths that many of these former foster youth appear to share. As a group, they continue to exhibit extraordinary optimism and high aspirations. In addition, many have maintained close relationships with members of their biological family and with adult mentors. This report is descriptive rather than explanatory in nature. Future directions for policy and practice will become clearer as we examine predictors of transition outcomes for the young people in our study and identify potentially effective avenues for intervention. Nevertheless, 84 we believe that two observations warrant the immediate attention of the child welfare services community. First, there is sobering evidence that foster youth are not are not acquiring the life skills they will need during the transition to adulthood. Although the language of the Foster Care Independence Act makes clear that states should continue to provide independent living services to young people through age 21, even if they are no longer in care, our results suggest that all too often this is not happening, and that the implementation of that legislation may be falling far short of its intent. Young people in our study reported little in the way of independent living services receipt between 19 and 21. Moreover, most of what they received was provided before they were discharged from care (i.e., to Illinois youth who remained in care past their twentieth birthday). To be sure, evidence of the effectiveness of independent living services is virtually nonexistent, calling into question whether provision of such services would improve the outcomes of youth transitioning to adulthood from foster care (Montgomery, Donkoh, & Underhill, 2006). It may also be the case that many foster youth in transition do not make use of services that are made available to them. Still, nearly two-fifths of the young people report that there was some kind of assistance not offered to them that would have helped them prepare for independence. Second, our data highlight the need for child welfare practice and policy to pay closer attention to the family connections of foster youth. One-quarter of the young people in our 85 study were living with a parent or other relative at age 21. Three-quarters felt very close to at least one biological family member and more than four-fifths were in contact with a member of their biological family at least once a week. In addition, many of these young people were receiving concrete assistance from their families. Family members often provided child care for the young parents in our study and homes for their nonresident children. One need not believe that relationships between these young people and their families are uniformly beneficial to conclude that policymakers, child welfare practitioners, and the courts should give more consideration to these family ties. That these ties are not being given adequate consideration is evidenced by the fact that nearly half of the young people in our study reported that the rights of their parents had been terminated by the juvenile court. The descriptive findings presented in this report raise further questions about the transition to adulthood among young people aging out of foster care. First, what are the predictors of transition outcomes for this vulnerable population? What risk or protective factors distinguish those on track to become self-sufficient young adults from those who seem to be headed down less promising pathways? What internal or external resources allowed the former to overcome the various challenges that typically confront youth aging out of foster care—in other words, what made them resilient? We will continue to conduct analyses to help answer these questions, and these answers may suggest avenues for intervention Second, do foster youth benefit during the transition to adulthood if the state continues its role as parent beyond age 18? Currently states are not entitled to IV-E federal reimbursement 86 for services provided to foster youth who are older than 18. However, pending legislation (i.e., S. 1512) would extend the IV-E entitlement to foster youth between the ages of 18 and 21. The introduction of this legislation suggests that policy makers are beginning to rethink the government’s responsibility to support foster youth during the transition to adulthood. The Midwest Study has much to contribute to the debate that is likely to arise over this issue. Indeed, it is the only that study that can compare the outcomes of young adults in a state where foster youth can remain under the care and supervision of the child welfare system until age 21 (i.e., Illinois) to those of young adults in states where that has not been an option (i.e., Iowa and Wisconsin).18 The comparisons we made in our wave 2 report between the young adults who were still in care at age 19 and those who had already left suggested that foster youth stood to benefit if they were allowed to remain under the care and supervision of the child welfare system beyond age 18,. We are currently in the process of examining whether extending care beyond age 18 leads to better outcomes during the transition to adulthood, and, if so, whether the benefits of doing so outweigh the costs. Preliminary analysis suggests that at least some of the apparent benefits of extending care continue through at least age 21.19 . . However, for at least some outcomes of interest, it may be too soon to observe an effect. One reason is that many of the Illinois youth did not exit care until just a few months before 18 In 2006, Iowa’s child welfare statute was amended to create a program that provides continuing support to former foster youth who are at least 18 years old but not yet 21 under certain circumstances. However, the Iowa youth participating in the Midwest Study were too old to benefit from this program when it was implemented. 19 See [Issue Brief] 87 their wave 3 interview. Another is that certain benefits may only be realized after other outcomes have been achieved. In particular, our earlier analyses suggested that young adults were more likely to be enrolled in college if they were still in care at age 19. Although college enrollment should have positive effects on employment and earnings over the long term, there may be a tradeoff between postsecondary education and labor market outcomes at age 21, particularly if the college-educated young adults are still in school. Thus, it may be necessary to follow these young adults for several more years before any valid conclusions can be drawn about the potential benefits of extending foster care 88 REFERENCES American Psychiatric Association (1994). Diagnostic and Statistical Manual, 4th ed Arlington, VA: APA. Bickel, G., Nord, M., Price, C., Hamilton, W., & Cook, J. (2000). Guide to measuring household food security. Washington, DC: United States Department of Agriculture, Food and Nutrition Service, Office of Nutrition, Analysis and Evaluation. Blome, W. (1997). What happens to foster kids: Educational experiences of a random sample of foster care youth and a matched group of non-foster care youth. Child and Adolescent Social Work, 14, 41–53. Blumberg, S., Bialostosky, K., Hamilton, W., & Briefel, R. (1999). The effectiveness of a short form of the household food security scale. American Journal of Public Health, 89, 1231-34. Bos, J., Polit, D., & Quint, J. (1997). New Chance: Final Report on a Comprehensive Program for Young Mothers in Poverty and Their Children. New York: MDRC. Courtney, M.E., Piliavin, I., Grogan-Kaylor, A., & Nesmith, A. (2001). Foster youth in transitions to adulthood: A longitudinal view of youth leaving care. Child Welfare, 80(6), 685-717. Courtney, M.E., Dworsky, A., Ruth, G., Keller, T., Havlicek, J., & Bost, N. (2005). Midwest Evaluation of the Adult Functioning of Former Foster Youth: Outcomes at Age 19. Chicago: Chapin Hall Center for Children at the University of Chicago. Courtney, M.E., & Hughes Heuring, D. (2005). The transition to adulthood for youth “aging out” of the foster care system. In D. W. Osgood, E. M. Foster, C. Flanagan, & G. Ruth (Eds.), On Your Own Without a Net: The Transition to Adulthood for Vulnerable Populations. Chicago: The University of Chicago Press. Courtney, M.E., Terao, S., & Bost, N. (2004). Midwest Evaluation of the Adult Functioning of Former Foster Youth: Conditions of the Youth Preparing to Leave State Care. Chicago: Chapin Hall Center for Children at the University of Chicago. Dworsky, A., Courtney, M.E., & Zinn, A. (2007). Child, parent and family level predictors of child welfare services involvement among TANF applicant families. Children and Youth Services Review, 29, 802-820. Fields, J. (2003). America’s Families and Living Arrangements: 2003. Current Population Reports, P20-553. U.S. Census Bureau, Washington, DC. 89 Goldscheider, F., & Goldscheider, C. (1999). The changing transition to adulthood: Leaving and returning home. Thousand Oaks, CA: Sage Publications. Gribble, J. N., Miller, H. G., Rogers, S. M., & Turner, C. F. (1999). Interview mode and measurement of sexual behaviors: Methodological issues. The Journal of Sex Research, 36, 16-24. Harris, K., Florey, F., Tabor, J., Bearman, P., Jones, J., & Udry, J. R. (2003). The National Longitudinal Study of Adolescent Health: Research Design. Available: http://www.cpc.unc.edu/projects/addhealth/design. Haveman, R., & Wolfe, B. (1994). Succeeding Generations: On the Effects of Investing in Children. New York: Russell Sage Foundation. Huston, A., Miller, C., Richburg-Hayes, L., Duncan, G., Eldred, C., Weisner, T., Lowe, E., McLoyd, E., Crosby, D., Ripke, M., & Redcross, C. (2003). New Hope for families and children: Five-year results of a program to reduce poverty and reform welfare. New York: MDRC. Levin-Epstein, J., & Greenberg, M. (2003). Leave no youth behind: Opportunities for Congress to reach disconnected youth. Washington, DC: Center for Law and Social Policy. McMillan, C. J., & Tucker, J. (1999) The status of older adolescents at exit from out-of-home care. Child Welfare 78(3), 339-359. Montgomery, P., Donkoh, C., & Underhill, K. (2006). Independent living programs for young people leaving the care system: The state of the evidence. Children and Youth Services Review, 28, 1435-1448. Pecora, P. J., Kessler, R. C., Williams, J., O’Brien, K., Downs, A. C., English, D., Hiripi, E., White, C. R., Wiggins, T., & Holmes, K. E. (2005). Improving family foster care: Findings from the Northwest Foster Care Alumni Study. Seattle, WA: Casey Family Programs. Pecora, P. J., Williams, J., Kessler, R. J., Downs, A. C., O’Brien, K., Hiripi, E., & Morello, S. (2003). Assessing the effects of foster care: Early results from the Casey National Alumni Study. Seattle, WA: Casey Family Programs. Rose, D. T., Abramson, L. Y., & Kaupie, C. A. (2000). The Lifetime Experiences Questionnaire: A Measure of History of Emotional, Physical, and Sexual Maltreatment. Manuscript in preparation, University of Wisconsin-Madison, Madison, Wisconsin. Schoeni, R. & Ross, K. (2004). Material assistance received from families during the transition to adulthood. In R. Settersten, Jr., F. Furstenberg, Jr., and R. Rumbaut (Eds.). On the 90 Frontier of Adulthood: Theory, Research, and Public Policy. Chicago: University of Chicago Press. Settersten, R., Furstenberg, F. F., & Rumbaut, R. G. (Eds.) (2005). On the frontier of adulthood: Theory, research, and public policy. Chicago: University of Chicago Press. Sheehy, A., Oldham, E., Zanghi, M., Ansell, D., Crreia, P., & Copeland, R. (2001). Promising practices: Supporting transition of youth served by the foster care system. National Foster Care Awareness Project. Sherbourne, C., & Stewart, A. (1991). The MOS Social Support Survey. Social Science Medicine, 32 (6), 705-714. Straus, M. A., Hamby, S. L., Finkelhor, D., Moore, D. W., & Runyan, D. (1998). Identification of child maltreatment with the Parent-Child Conflict Tactics Scales: Development and psychometric data for a national sample of American parents. Child Abuse and Neglect, 22(4), 249-270. Sum, A., Khatiwada, I., Pond, N., Trub’skyy, M., Fogg, N., & Palma, S. (2002). Left behind in the labor market: Labor market problems of the nation’s out-of-school young adult populations. Center for Labor Market Studies at Northeastern University. Turner, C. F., Ku, L., Rogers, S. M., Linderg, L. D., Pleck, J. H., & Sonenstein, F. L. (1998). Adolescent sexual behavior, drug use, and violence: Increased reporting with computer survey technology. Science, 280, 867-873. U.S. Department of Health and Human Services (2007). Temporary Assistance for Needy Families Separate State Program-Maintenance of Effort Aid to Families with Dependent Children Caseload Data. Washington, DC: Office of Family Assistance. Retrieved May 21, 2007 from http://www.acf.hhs.gov/programs/ofa/caseload/caseloadindex.htm U.S. Department of Health and Human Services (2006). The AFCARS report: Preliminary estimates for FY 2005 as of September 2006. Administration for Children and Families, Administration on Children, Youth and Families, Children's Bureau. Retrieved August 9, 2007 from http://www.acf.hhs.gov/programs/cb/stats_research/afcars/tar/report13.htm U.S. House of Representatives (2004). 2004 Green Book: Overview of Entitlement Programs. Washington, DC: Government Printing Office. Wald, M., & Martinez, T. (2003). Connected by 25: Improving the life chances of the country’s most vulnerable 14 – 24 year olds. Hewlett Foundation Working Paper. World Health Organization (1998). The Composite International Diagnostic Interview (CIDI). Geneva, Switzerland. 91 Youth Transition Funders Group (2004). Connected by 25: A plan for investing in successful futures for foster youth. Takoma Park, MD: Youth Transition Funders Group. 92 Appendix A Outcome of Baseline Field Period Completed interviews Eligible but not interviewed Care provider refusal Respondent refusal Contact with care provider or informant but not respondent Unable to reach respondent after prior contact Respondent no-show for appointment Respondent out of state or country after start of field period Respondent runaway after start of field period Not interviewed and eligibility unknown No attempt to contact respondent Unable to reach respondent Unable to locate address or valid contact information not available Not eligible to be interviewed Respondent physically or mentally unable to complete interview Respondent runaway or missing prior to start of field period Respondent out of state prior to start of field period Respondent incarcerated prior to start of field period Other eligibility issue Total 93 IL 474 IA 63 WI 195 Total 732 2 5 6 2 1 2 2 20 0 1 0 0 0 0 0 1 1 1 2 1 0 0 0 5 3 7 8 3 1 2 2 26 1 0 2 3 0 1 4 5 0 0 1 1 1 1 7 9 17 13 11 38 5 84 1 1 0 1 2 5 16 1 1 1 1 20 34 15 12 40 8 109 604 80 227 911 Chapin Hall Center for Children at the University of Chicago 1313 East 60th Street Chicago, Illinois 60637 www.chapinhall.org phone: 773/753-5900 fax: 773/753-5940