eISSN: 1989-9742 © SIPS.
DOI: https://doi.org/10.7179/PSRI_2025.48.04

http://recyt.fecyt.es/index.php/PSRI/

Improving the postsecondary educational outcomes of young people in extended foster care: an evaluation of the effects
of My First Place

Mejorar los resultados educativos postsecundarios de los jóvenes en acogida temporal prolongada: una evaluación de los efectos de My First Place

Melhorar os resultados educacionais pós-secundários de jovens em acolhimento prolongado: uma avaliação dos efeitos do My First Place

Amy DWORSKY https://orcid.org/0000-0001-6538-8550

Amanda M. GRIFFIN https://orcid.org/0000-0002-5971-5954

Chapin Hall

Received date: 19.VIII.2025

Reviewed date: 20.X.2025

Accepted date: 30.X.2025

CONTACT WITH THE AUTHORS

Amy Dworsky. Chapin Hall. 200 W. Madison Street, Chicago, IL, 60606. Email: adworsky@chapinhall.org

The vast majority of young people in foster care aspire to go to college (Courtney et al., 2014; Kirk et al., 2011). Nevertheless, research indicates that young people in foster care are less likely to earn a college degree than their peers in the general population. Based on a review of 13 publications, several of which used data from the same two studies, Okpych and colleagues (2025) found that between 8 and 12 percent of young people who were in foster care on or after their 13th birthday earned a two- or four-year college degree. By comparison, in 2024, 51 percent of 25 to 29-year-olds in the U.S. had at least an associate’s degree, and 40 percent had a bachelor’s degree (U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, 2025).

Increasing postsecondary educational attainment among young people in foster care has been a public policy goal in the United States for the past 40 years. The federal government, state governments, and philanthropic organizations have made significant investments aligned with this goal. One such investment is the Education and Training Voucher (ETV) program. Created in 2001, this federal program provides states with funds that can be used to cover the costs of college attendance, including tuition, fees, or other college-related expenses for young people who experienced foster care at age 14 or older, who aged out of foster care, or who exit foster care through adoption or kinship guardianship at age 16 or older. Young people are eligible for up to $5000 per year for up to five years or until their 26th birthday, whichever comes first, as long as they are making satisfactory academic progress (Fernandes-Alcantara, 2019). Receipt of ETVs is associated with higher rates of college persistence (Hanson et al., 2022; Okpych et al., 2020) and graduation (Hanson et al., 2022). However, many young people who are eligible for an ETV do not receive one (Hanson et al., 2022; Okpych et al., 2020).

State tuition waiver programs are another type of investment designed to promote postsecondary educational attainment among young people in foster care1. These state-funded programs require public colleges and universities to waive the cost of tuition and fees for young people currently or formerly in foster care after other sources of financial aid are applied (Hernandez et al., 2017). Eligibility for these programs varies widely across states. As of 2023, 24 states offered statewide tuition waiver programs specifically for youth who have been in foster care (University of Washington, n.d.). Tuition waivers are associated with an increase in postsecondary educational enrollment (Geiger & Okpych, 2022; Watts et al., 2018). Although one study found that students who received tuition waivers were 3.5 times more likely to graduate than those who did not receive a tuition waiver, many students who were eligible to receive a tuition waiver did not use one (Watt & Faulkner, 2020). Moreover, another study found no relationship between the availability of state tuition waivers and degree completion (Gross et al., 2022).

Unlike ETVs and state tuition waivers, which aim to remove financial barriers to postsecondary education, campus support (or college success) programs aim to increase college retention and graduation rates among students who have experienced foster care by providing a range of academic, psychosocial, and other supports (Dworsky & Perez, 2010; Geiger et al., 2018). Despite a proliferation of campus support programs over the past two and a half decades (Geiger et al., 2018; Okpych et al., 2020), research on these programs is limited (Geiger et al., 2018). Most studies have examined student outcomes rather than program impacts (Dworsky, 2020; Huang et al., 2019; Lenz-Rashid, 2018; Watt et al., 2013). One exception is a recent study by Day and colleagues (2025), which found that students with lived experience in foster care who participated in a campus support program were significantly more likely to be retained than students with lived experience in foster care who did not participate and as likely to be retained as low income, first generation students who had not experienced foster care. Moreover, unlike other studies that have focused on a single program, Okpych and colleagues (2020) found that participation in campus support programs was associated with an increase in college persistence at both 2-year and 4-year schools.

Finally, although the extension of foster care beyond age 18 was not explicitly designed to increase postsecondary educational attainment, growing evidence suggests that allowing young people to remain in foster care until their 21st birthday is associated with an increase in college enrollment (Courtney & Hook, 2017; Courtney et al., 2018; Dworsky & Courtney, 2010; Okpych & Courtney, 2020). However, extended foster care does not appear to increase persistence or graduation (Courtney et al., 2018; Dworsky & Courtney, 2010; Okpych & Courtney, 2020).

Much has been written about the reasons for the gap in postsecondary educational attainment between young people who have experienced foster care and their peers in the general population (Hayes Piel, 2018; Tobolowsky et al., 2019). One factor that has received relatively little attention is housing instability (Geiger et al., 2018; Kinarsky, 2017; Okpych et al., 2020; Tobolowsky et al., 2019). This is somewhat surprising given the high rate of homelessness among young people who aged out of foster care. Research suggests that between one-quarter and one-third of these young people will experience homelessness after they age out of foster care (Dworsky et al., 2013; Feng et al., 2020; Fowler et al., 2009; Hernandez et al., 2023).

The present study examines the postsecondary educational attainment of young people currently or formerly in extended foster care who participated in a program that provides participants with housing so they can focus on their education and employment goals. The study had three primary objectives. The first was to describe the postsecondary educational outcomes of young people currently or formerly in extended foster care. The second was to examine the relationship between postsecondary educational outcomes and young people’s demographic characteristics. The third was to determine whether program participation improves young people’s postsecondary educational outcomes.

These three study objectives lead to our four research questions:

1. What percentage of young people currently or formerly in extended foster care enroll in college, complete at least one semester of college, and earn a credential?

2. How do the percentage of young people who enroll in college, complete at least one semester of college, and earn a credential vary by gender, race/ethnicity, sexual orientation, parenting status, and county?

3. Are young people who participate in the program more likely to enroll in college, complete at least one semester of college, and earn a credential than young people who were eligible for but did not participate in the program?

4. Does participation in the program increase college enrollment, semester completion, and credential attainment after controlling for other factors?

Before describing the methodology, we used to answer these questions, we briefly describe the program that was the focus of our evaluation.

My First Place

My First Place (MFP) is a program operated by First Place for Youth (FPFY) that provides housing, education and employment supports, and intensive case management services to 18- to 25-year-olds in six California counties (Alameda, Contra Costa, Los Angeles, San Francisco, Santa Clara, and Solano) who are or were in extended foster care. Young people typically remain in the program for between one and three years. In Fiscal Year 2024, the program served 550 youth in California.

Upon program entry, all young people are provided with fully subsidized housing that is chosen through a collaborative process that takes the young person’s needs, neighborhood safety, and proximity to school, work, transportation, and community resources into account. Young people receive a monthly stipend to cover basic needs and are encouraged to contribute to a savings account that is paid out when they exit the program.

Once they are stably housed, young people focus on their education and employment goals and develop independent living that will promote self-sufficiency. Young people meet weekly with a Youth Advocate and biweekly with an Education and Employment Specialist, develop positive relationships with supportive adults, and are encouraged to reframe setbacks as opportunities for growth.

Service delivery is trauma-informed, based on positive youth development and harm-reduction principles, and developmentally and culturally appropriate. During the Engagement phase, which lasts up to six months, staff meet with youth regularly to build relationships, understand their needs, and collaboratively set goals around education, employment, housing, and healthy living. These goals and the steps youth will take to achieve them are captured in the youth’s individualized Action Plan. During the Skill-Building phase, which typically lasts 6 to 18 months, staff support youth as they build skills in household management, financial literacy, interpersonal relations, and responsible tenancy. Youth also work with staff to develop community connections and positive relationships with caring adults, complete high school and/or prepare for postsecondary education, and secure employment related to their career interests. During the Transition to Independence phase, which typically lasts 12 to 24 months, staff engage with youth proactively to develop a plan for securing housing, maintaining relationships with caring adults, accessing needed services, and continuing progress toward education and employment goals.

Methods

Data Sources

Our evaluation relied on data from two sources.

First Place for Youth

The research team received individual-level data for 1.598 MFP participants who exited the program between 2015 and 2022 (“participants”) and 820 young people who were referred to MFP but did not participate either by choice or because the program was at capacity (“non-participants”). The latter were assigned a pseudo exit date based on their referral date and the average length of stay in the program for their county. The data included race/ethnicity, gender identity, sexual orientation, parenting status, history of homelessness, prior legal system involvement, and county. Most of this information came from intake forms completed when young people were referred to the program. For young people who participated in the program, information about parenting status and legal system involvement is updated while they are in the program. For program participants, they also included length of time in the program, exit destination, and termination reason.

National Student Clearinghouse

We obtained individual-level data from the National Student Clearinghouse (NSC), a nonprofit organization that receives college enrollment and graduation data from more than 36,000 colleges and universities across the U.S. We provided the NSC with a data file that include the first name, last name, birthdate, and unique identification number for each of the 2,418 young people in our sample. The NSC matched the names and birthdates of the young people in our sample against its database and returned a file with college enrollment and graduation data through October 2024 for the young people with matching records. Each record included enrollment start and end dates, the postsecondary institution’s name and state, the type of institution (i.e., public or private, 2-year or 4-year), and, if relevant, the graduation date and credential earned. Then we linked the NSC data to the data we had received from First Place for Youth using the unique identification number.

Analysis

Our analysis focused on college enrollment, semester completion, and credential attainment. Young people were counted as having completed a semester if they did not withdraw or take a leave of absence. Credentials included certifications, associate’s degrees, bachelor’s degrees, and master’s degrees. We examined whether these outcomes varied by gender, sexual orientation, race/ethnicity, parenting status, cohort, region, and program participation, and ran chi-square tests to determine whether the between-group differences were statistically significant.

Because any differences in outcomes between the young people who participated in the program and the young people who did not could potentially be explained by pre-existing differences between the two groups, we used propensity score matching (PSM) to create comparable participant and non-participant groups. We estimated propensity scores by regressing participation status on gender, race/ethnicity, age, parenting status, and region. We did not include sexual orientation, homelessness history, or legal system involvement in the model because data were missing for a non-trivial percentage of the young people.

We experimented with different matching algorithms. Based on the balance diagnostics, we used one-to-one greedy matching (Austin, 2011). With this algorithm, one treatment group member is randomly selected and matched to the comparison group member with the closest propensity score (even if that comparison group member would be a better match for a subsequent treatment group member). We also required that the absolute difference in propensity scores between the matched pairs be below .2 (i.e., the caliper distance). Treatment group members were excluded from the matched sample if no comparison group member had a propensity score whose difference from the treatment group member’s propensity score was within the caliper distance. After constructing the matched groups, we tested whether they were roughly equivalent on observable measures.

Because MFP participants could have enrolled in college, completed a semester of college, or earned a credential before entering MFP, we could not use logistic regression to examine the effects of MFP on our outcome measures. Instead, we used survival analysis, a set of methods for analyzing “time to event” data and estimated a set of proportional hazard models. The hazard is the instantaneous risk that an event will occur at a particular time, given that the event has not already occurred. Hazard models allow for right censoring (i.e., when the event in question does not occur before the end of the observation period) and can incorporate time-varying covariates whose values change over time (Allison, 1984, 1995). In this case, MFP entry can be considered a time-varying covariate (i.e., coded zero before entry and coded one thereafter).

To estimate the effect of the program on the hazard of first enrolling in college, our dependent variable was the number of days between each young person’s 16th birthday and the date on which they first enrolled in college (if they ever enrolled) or the last enrollment date recorded in the data (i.e., September 3, 2024). We used the young people’s 16th birthday as a starting point because none of the young people had enrolled in college before age 16. To estimate the effect of the program on the hazard of first completing one semester, our dependent variable was the number of days between each young person’s 16th birthday and the date on which they completed their first semester (if they completed at least one) or the last semester completion date recorded in the data (i.e., March 21, 2025). Finally, to estimate the effect of the program on the hazard of earning a credential, our dependent variable was the number of days between the first day on which a young person enrolled in college and their first (or only) graduation date (if they earned a credential) or the last graduation date (i.e., September 25, 2024). This analysis was limited to the sample of young people who enrolled in college.

All three models included gender, race/ethnicity, county, and parenting status, and MFP entry as covariates, and MFP entry could vary over time. The time-varying covariate was initially coded zero for all the young people. The value of that covariate switched to one on the date the young people enrolled in MFP if they enrolled. Exponentiating the parameter estimates from the hazard models converts the coefficients into estimated hazard ratios. A hazard ratio greater than one means that the hazard is higher for the group whose value is one than for the group whose value is zero; a hazard ratio less than one means that the hazard is higher for the group whose value is zero than for the group whose value is one (Yamaguchi, 1991).

Hazard ratios can be interpreted as a percentage change in risk for the group exposed to the treatment as compared to the unexposed group. The percentage change can be calculated by subtracting 1 from the hazard and multiplying by 100. For example, if the hazard ratio is 1.5, the increased risk for the exposed group is (1.5-1) * 100 or 50 percent. Similarly, if the hazard ratio is .5, the decreased risk for the exposed group is (.5-1) * 100 or -50 percent.

Ethics

The study was approved by the study team’s institutional review board. The study team received a waiver of informed consent to analyze the First Place for Youth and NSC data.

Sample Characteristics

Table 1 compares the characteristics of the young people who participated in the program to the characteristics of the young people who did not. The percentage of young people in Los Angeles County was much lower in the participant group than in the non-participant group. This reflects the fact that Los Angeles County has more young people in extended foster care than any other county in the United States and the number of young people being referred to MFP in Los Angeles is far higher than the number of young people the program can serve. Most of the young people in both groups are either Black or Latino. In both groups, young women outnumber young men. About one in eight young people in both groups identify as gay, lesbian, or bisexual, but young people in the earlier exit cohorts were not asked about their sexual orientation. Participants were more likely to be parents than non-participants, which could be because information about parenting status is updated while young people are in the program. About half of both groups reported a history of homelessness, and about one in ten young people in both groups reported being arrested or incarcerated. However, data were far more likely to be missing for non-participants than for participants for both variables.

Results

Sixty-one percent (n = 1,465) of the young people whose records we shared with the NSC ever enrolled in college. The vast majority of these young people enrolled in a two-year public (i.e., community) college. Of those who ever enrolled in college, 89 percent (n = 1,302) first enrolled in a community college while 5 percent (n = 79) first enrolled in a four-year public college or university. When both first and sequent enrollments are included, 95 percent (n = 1,389) of the young people who ever enrolled in college ever enrolled in a community college and 12 percent (n = 170) ever enrolled in a four-year public college or university. The difference between the percentage who first enrolled and the percentage who ever enrolled reflects the fact that 52 percent (n = 767) of the young people who ever enrolled in college enrolled in more than one. Fifty-eight percent (n = 1,377) of the young people whose records we shared with NSC, or 94 percent of the young people who ever enrolled, completed at least one semester.

Overall, 130 young people earned a postsecondary credential. These young people represent 5 percentage of the sample or 9 percentage of the young people who ever enrolled in college. They include 56 young people who earned a certificate, 71 who earned an associate’s degree, 52 who earned a bachelor’s degree, and 3 who earned a master’s degree. These numbers sum to more than 130 because some young people earned more than one credential. The highest credential earned was a certificate for 23 young people, an associate’s degree for 55 young people, a bachelor’s degree for 49 young people, and a master’s degree for 3 young people.

Program Participation and Postsecondary Educational Outcomes

Sixty-eight percent of MFP participants enrolled in college, and 64 percent completed at least one semester, compared to 49 percent and 45 percent, respectively, of non-participants. Both of these differences are statistically significant. However, among young people who enrolled in college, 10 percent of MFP participants and 7 percent of non-participants earned a postsecondary credential. This difference was not statistically significant.

Exit Cohort and Postsecondary Educational Outcomes

Although early exit cohorts tended to have higher enrollment and semester completion rates than later cohorts, this was in part because they had more time to pursue postsecondary education. However, these differences were not statistically significant. Moreover, young people were equally likely to have earned a postsecondary credential regardless of the year in which they exited.

Demographic Characteristics and Postsecondary Educational Outcomes

Table 6 shows the relationship between each of our postsecondary educational outcomes and an array of demographic characteristics.

Gender. Young women were significantly more likely to enroll in college, to complete at least one semester, and to earn a postsecondary credential than young men3.

Sexual orientation. Young people who identified as lesbian, gay, or bisexual were as likely to enroll, to complete at least one semester, and to earn a postsecondary credential as young people who identified as straight.

Race/ethnicity4. Latino young people were less likely to have enrolled in college and less likely to have completed a semester than young people who were Black or White. However, only the differences between Latino young people and Black young people were statistically significant after applying the Bonferroni corrections. Latino young people were also more likely to have earned a postsecondary credential than Black young people and multiracial young people, but only the difference between Black young people and Latino young people was statistically significant after applying a Bonferroni correction.

Parenting. Parents were as likely to enroll in college, to complete at least one semester, and to earn a postsecondary credential as non-parents5.

County. Young people in Los Angeles County were less likely to have enrolled in college than young people in Solano, San Francisco/Alameda, and Contra Costa Counties. Young people in Los Angeles were less likely to have completed at least one semester than young people in Solano and San Francisco/Alameda Counties, and young people in Contra Costa County were less likely to have completed at least one semester than young people in San Francisco/Alameda Counties. Among young people who enrolled in college, young people in Los Angeles County were less likely to have earned a postsecondary credential than young people in Santa Clara County.

Timing of Postsecondary Educational Outcomes

Sixty-five percent of the 1,078 MFP program young people who enrolled in college first enrolled in college before they entered the program, and 56 percent of the 1,025 MFP program young people who completed at least one semester completed their first semester before they entered the program. However, only 8 percent of the 104 MFP young people who earned a credential earned their first, and in some cases, only credential before they entered the program.

Effect of Program on Postsecondary Educational Attainment

Table 7 compares the MFP participants to the non-participants before and after PSM. PSM eliminated all but one of the statistically significant differences between the two groups. After PSM, the MFP participant group included proportionately fewer parents than the non-participant group, even though parent status was one of the variables included in the model we used to generate the propensity scores.

Table 8 shows the results of the three hazard models. Consistent with our bivariate results, MFP participation increased the hazard of enrolling in college by 32 percent. Also consistent with our bivariate results, the hazard of enrolling in college was significantly higher for young women than for young men and for young people in San Francisco/Alameda Counties than for young people in Los Angeles County. Contrary to our bivariate results, the hazard of first enrolling in college was lower for young parents than for their non-parenting peers. Also, contrary to our bivariate results, the hazard of first enrolling in college did not vary by race/ethnicity.

Consistent with our bivariate results, MFP participation increased the hazard of completing a semester in college by 39 percent. Also consistent with our bivariate results, the hazard of completing a semester was significantly higher for young women than for young men and for young people in San Francisco/Alameda Counties than for young people in Los Angeles County. Contrary to our bivariate results, the hazard of completing a first semester was lower for young parents than for their non-parenting peers. Also, contrary to our bivariate results, the hazard of first completing a semester did not vary by race/ethnicity.

Consistent with our bivariate results, MFP participation had no effect on the hazard of earning a postsecondary credential. Also consistent with our bivariate results, the hazard of earning a postsecondary credential was significantly higher for young women than for young men. Contrary to our bivariate results, the hazard of earning a first credential did not vary by race/ethnicity.

Discussion

We undertook this study to examine the effects of MFP on the postsecondary educational outcomes of young people. This question is important given that postsecondary educational attainment is positively correlated with employment and earnings among young people who age out of foster care, and the effects of postsecondary educational attainment on employment and earnings are more pronounced among young people who age out of foster care than among their peers in the general population (Okpych & Courtney, 2014). Moreover, the economic benefits of postsecondary education are not limited to those who earn a degree. Completing at least some college is associated with an increased likelihood of being employed and higher earnings, although to a much lesser degree than having a two- and especially a four-year degree (Okpych & Courtney, 2014).

Our findings suggest that many young people who age out of extended foster care do pursue some type of postsecondary education. Sixty-one percent of the young people ever enrolled in college, and 58 percent completed at least one semester. Consistent with the results of prior studies, we found that the vast majority of young people who pursued postsecondary education enrolled in a community college, even if they later enrolled in a 4-year school. Also consistent with prior studies, we found that only 9 percent of the young people who enrolled in college earned a credential, even though we defined credential broadly to include certificates as well as associate or bachelor’s degrees.

Young people who enrolled in MFP were significantly more likely to have enrolled in college and more likely to have completed at least one semester than young people who did not enroll in MFP. However, about two-thirds of the MFP participants who enrolled in college first enrolled in college, and over half of the MFP participants who completed at least one semester completed their first semester before they entered the program. Thus, their first college enrollment and semester completion cannot be attributed to their program. Moreover, although young people who enrolled in MFP were more likely to have earned a postsecondary credential than young people who did not enroll in MFP, this was largely because young people who enrolled in MFP were significantly more likely to have enrolled in college. Among young people who enrolled in college, MFP participants were no more likely to have earned a credential than non-participants.

We also examined the relationship between postsecondary educational attainment and other factors. Consistent with trends in the broader population, young women were significantly more likely to have enrolled in college, to have completed at least one semester, and to have earned a credential than young men. The relationship between postsecondary educational attainment and race/ethnicity was less straightforward. Latino young people were less likely to have enrolled in college than Black young people. However, if they enrolled in college, Latino young people were more likely than Black young people to have earned a postsecondary credential. We also observed regional differences in postsecondary educational attainment. Young people in the Los Angeles region were, in general, less likely to have enrolled in college, less likely to have completed a semester, and less likely to have earned a credential than young people in the other regions. However, not all of those differences were statistically significant.

Because MFP participation was not randomized, we estimated the effects of the program on postsecondary educational attainment using propensity score-matched treatment and comparison groups. Consistent with the results of our bivariate analysis, we found that MFP enrollment increased the hazard of enrolling in college by 32 percent and the hazard of completing a semester in college by 39 percent. Also consistent with the results of our bivariate analysis, we found that the hazard of enrolling in college and the hazard of completing at least one semester were significantly higher for young women than for young men and for young people in San Francisco/Alameda Counties than for young people in Los Angeles County. Although our bivariate analysis revealed no differences in college enrollment or semester completion related to parenting, the hazard of first enrolling in college and the hazard of completing a first semester were lower for young parents than for their non-parenting peers. Importantly, and contrary to our bivariate analyses, we found no relationship between race/ethnicity and the hazard of first enrolling in college or the hazard of first completing a semester. Finally, consistent with our bivariate analysis, we found no relationship between MFP enrollment and the hazard of earning a first postsecondary credential. The hazard of earning a first postsecondary credential was significantly higher for young women than young men but did not vary by region or race/ethnicity.

Limitations

Our study has several limitations that we wish to acknowledge. First, we do not know the extent to which the data we received from NSC include false positives (i.e., young people who did not enroll in college matched to NSC records) or false negatives (i.e., young people who did not enroll in college not matched to NSC records). Second, although nearly all colleges and universities in the United States report enrollment and/or graduation data to the NSC, the NSC does not receive data from institutions with postsecondary education programs that lead to vocational or industry-recognized certificates, which MFP encourages participants to pursue. Thus, the program may have had a larger impact on postsecondary educational outcomes than our results suggest. Third, the NSC data do not provide information about credit accumulation, an important measure of progress towards a postsecondary educational credential. Fourth, some of the pre-existing differences between the “participant” and “non-participant” groups involved characteristics that were related to postsecondary educational attainment. Although our analysis of postsecondary educational attainment used propensity score-matched groups, PSM is an imperfect substitute for randomization. Finally, limiting the analysis of earning a credential to the sample of young people who ever enrolled in college introduces the potential for sample selection bias, and this can result in biased parameter estimates.

Policy and Practice Implications

Although our data are agnostic as to why participation in MFP had a positive effect on college enrollment and semester completion, at least two explanations are worth considering. One possibility is that MFP participants benefited from the support they received from their Education and Employment Specialist and Youth Advocate. These professionals may have encouraged them or helped remove barriers to enrollment, such as by assisting with financial aid applications. To the extent that this type of support makes a difference, then child welfare systems need to think about how they can continue to provide support to young people after they age out of extended foster care by, for example, contracting with nonprofit organizations like First Place for Youth. Not all young people would avail themselves of these supports, but they would be an important resource for those who need them.

Paying for these additional supports could be a challenge. The Chafee Foster Care Program for Successful Transition to Adulthood (Chafee) program provides funds that states can use to prepare young people in foster care for the transition to adulthood and to support them during that transition. Under the Family First Prevention Services Act (FFPSA) of 2018, states that have extended foster care until age 21 can seek approval from the Children’s Bureau to use their federal Chafee dollars to fund services for young people who aged out of foster care until their 23rd birthday. However, that approval does not come with additional funding. In essence, this means that states must provide services to a larger population with the same amount of funds. One solution would be for Congress to increase the annual authorization for the Chafee program and allocate that increase to states that choose to provide services to young people who age out of extended foster care.

Another potential explanation for why MFP participation was associated with an increase in college enrollment and semester completion is that all the MFP participants had stable housing, and hence, were better able to concentrate on pursuing their postsecondary educational goals. Consistent with this explanation, research suggests that young people whose housing is unstable may have less cognitive capacity to focus on their education and may face a variety of barriers to enrolling or persisting in school (Bowers & O’Neill, 2019; Hallett & Freas, 2018; Silva et al., 2017).

MFP may help improve the postsecondary educational outcomes among young people who are or were in extended foster care by providing them with fully subsidized housing. More such programs are essential, given the high rate of homelessness and housing instability among young people who have aged out of foster care. That said, additional research is needed to determine whether other participation in other programs that provide young people who have aged out of extended foster care with fully subsidized housing is associated with higher rates of college enrollment or semester completion.

Although participation in MFP had a positive effect on college enrollment and semester completion, it had no effect on earning a postsecondary educational credential. One explanation for this null finding is that MFP participants “aged out” of the program or exited for other reasons before they were able to complete the requirements for a certificate or degree. This implies that allowing longer stays in the program might lead to higher rates of degree or certificate attainment.

Another possibility is that the supports that enabled MFP participants to enroll in college and complete at least one semester are different from the supports that are needed to persist in college long enough to earn a certificate or degree. Additional research is needed to determine what supports are needed to increase degree or certificate attainment.

Conclusion

The findings from our study highlight both the promise and the limitations of programs like MFP for improving the postsecondary educational outcomes of young people transitioning out of extended foster care. We found meaningful effects of the program on college enrollment and semester completion, but not on credential attainment. This suggests that increasing postsecondary educational attainment among this population may require longer-term, targeted supports.

Contributions

Funding

This research was funded by First Place for Youth.

Conflict of Interest

The authors declare that they have no conflicts of interest.

Bibliographic References

Austin, P. (2011). An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behavioral Research, 46, 399-424. https://doi.org/10.1080/00273171.2011.568786

Bowers, P., & O’Neill, M. (2019). The lived experience of being a homeless college student: A qualitative interpretive meta-synthesis (QIMS). Journal of Children and Poverty, 25, 114-130. https://doi.org/10.1080/10796126.2019.1629580

Courtney, M., Charles, P., Okpych, N., Napolitano, L., & Halsted, K. (2014). Findings from the California Youth Transitions to Adulthood Study (CalYOUTH): Conditions of foster youth at age 17. Chapin Hall at the University of Chicago.

Courtney M., & Hook J. (2017). The potential educational benefits of extending foster care to young persons: Findings from a natural experiment. Children and Youth Services Review, 72, 124-132. https://doi.org/10.1016/j.childyouth.2016.09.030

Courtney M., Okpych N., Park K., Harty J., Feng H., Torres-Garcia A., Sayed S. (2018). Findings from the California Youth Transitions to Adulthood Study (CalYOUTH): Conditions of youth at age 21. Chapin Hall at the University of Chicago. https://www.chapinhall.org/research/improved-outcomes-at-age-21-for-youth-in-extended-foster-care/

Day, A., Choi, Y., & Jackson, M. (2025). Transition to Independence (TIP) Wayne State: Effects on academic outcomes of college-enrolled students with lived experience in foster care. Journal of Diversity in Higher Education. https://doi.org/10.1037/dhe0000600

Dworsky, A., & Courtney, M. (2010). Does extending foster care beyond age 18 promote postsecondary educational attainment? Chapin Hall at the University of Chicago.

Dworsky, A., & Perez, A. (2010). Helping former foster youth graduate from college through campus support programs. Children and Youth Services Review, 32, 255-263. https://doi.org/10.1016/j.childyouth.2009.09.004

Dworsky, A., Napolitano, L., & Courtney, M. (2013). Homelessness during the transition from foster care to adulthood. American Journal of Public Health, 103, S318–S323. https://doi.org/10.2105/AJPH.2013.301455

Feng, H., Courtney, M., & Okpych, N. (2020). Memo from CalYOUTH. Chapin Hall at the University of Chicago.

Fowler, P., Toro, P., & Miles, B. (2009). Pathways to and from homelessness and associated psychosocial outcomes among adolescents leaving the foster care system. American Journal of Public Health, 99, 1453-1458. https://doi.org/10.2105/AJPH.2008.142547

Geiger, J., Hayes Piel, M., Day, A., & Schelbe, L. (2018). A descriptive analysis of programs serving foster care alumni in higher education: Challenges and opportunities. Children and Youth Services Review, 85, 287-294. https://doi.org/10.1016/j.childyouth.2018.01.001

Geiger J., & Okpych N. (2022). Connected after care: Youth characteristics, policy, and programs associated with postsecondary education and employment for youth with foster care histories. Child Maltreatment, 27, 658-670. https://doi.org/10.1177/10775595211034763

Gross, J., Geiger, J., Uhls, E., & King, G. (2022). The relationship between financial aid and postsecondary completion among youth formerly in foster care. Child and Adolescent Social Work Journal, 40, 207-219. https://doi.org/10.1007/s10560-022-00884-y

Hallett, R., & Freas, A. (2018). Community college students’ experiences with homelessness and housing insecurity. Community College Journal of Research and Practice, 42, 724-739. https://doi.org/10.1080/10668926.2017.1356764

Hanson D., Pergamit M., Packard Tucker L., Thomas K., Gedo S. (2022). Do education and training vouchers make a difference for young persons in foster care? OPRE Report #2022-140. Office of Planning, Research and Evaluation (OPRE), Administration for Children and Families, US Department of Health and Human Services. https://www.urban.org/research/publication/do-education-and-training-vouchers-make-difference-young-adults-foster-care

Hayes Piel, M. (2018). Challenges in the transition to higher education for foster care youth. New Directions for Community Colleges, 181, 21-28.

https://doi.org/10.1002/cc.20288

Hernandez, L., Day, A., & Henson, M. (2017). Increasing college access and retention rates of youth in foster care: An analysis of the impact of 22 state tuition waiver programs. Journal of Policy Practice, 16, 397-414. https://doi.org/10.1080/15588742.2017.1311819

Hernandez, H., Danielson, T., Mayfield, J., Black, C., & Felver, B. (2023). Homelessness among youth exiting systems of care in Washington State. Washington State Department of Social and Health Services, Research Data Analysis Division.

Huang, H., Fernandez, S., Rhoden, M., & Joseph, R. (2019). Elements of a college support program that matter: A case study. Research on Social Work Practice, 29, 949-960. https://doi.org/10.1177/1049731519832105

Kirk, C., Lewis, R., Nilsen, C., & Colvin, D. (2011). Foster care and college: The educational aspirations and expectations of youth in the foster care system. Youth & Society, 45, 307-323. https://doi.org/10.1177/0044118X11417734

Kinarsky, A. (2017). Fostering success: Understanding the experience of foster youth undergraduates. Children and Youth Services Review, 81, 220-228. https://doi.org/10.1016/j.childyouth.2017.08.016

Lenz-Rashid, S. (2018). An urban university campus support program for students from foster care: Services and outcomes. Children and Youth Services Review, 94, 180-185. https://doi.org/10.1016/j.childyouth.2018.09.033

Okpych, N., Whitman, K., Lee, J., Neria-Piña, L., Jackson, L., & Day, M. (2025). Secondary and postsecondary education outcomes of students with experience in foster care: Systematic review of the literature from 2000-2023. AERA Open, 11, 23328584251331454.

Okpych, N., & Courtney, M. (2020). The relationship between extended foster care and college outcomes for foster care alumni. Journal of Public Child Welfare, 14, 254-276. https://doi.org/10.1080/15548732.2019.16088

Okpych, N., & Courtney, M. (2014). Does education pay for youth formerly in foster care? Comparison of employment outcomes with a national sample. Children and Youth Services Review, 43, 18-28. https://doi.org/10.1016/j.childyouth.2014.04.013

Okpych N., Park S., Sayed S., Courtney M. (2020). The roles of campus-support programs (CSPs) and Education and Training Vouchers (ETVs) on college persistence for youth with foster care histories. Children and Youth Services Review, 111, Article 104891. https://doi.org/10.1016/j.childyouth.2020.104891

Silva, M., Kleinert, W., Sheppard, A., Cantrell, K., Freeman-Coppadge, D., Tsoy, E., Roberts, T., & Pearrow, M. (2017). The relationship between food security, housing stability, and school performance among college students in an urban university. Journal of College Student Retention: Research, Theory & Practice, 19, 284-299. https://doi.org/10.1177/1521025115621918

Tobolowsky, B., Scannapieco, M., Aguiniga, D., & Madden, E. (2019). Former foster youth experiences with higher education: Opportunities and challenges. Children and Youth Services Review, 104, 1-11. https://doi.org/10.1016/j.childyouth.2019.05.039

University of Washington. (n.d.). Tuition waivers by state. Retrieved August 15, 2025, from https://depts.washington.edu/fostered/tuition-waivers-state

Watt T., & Faulkner M. (2020). The Texas tuition and fee waiver program for youth who have experienced foster care: An assessment of waiver utilization and impact. Children and Youth Services Review, 117, 105285. https://doi.org/10.1016/j.childyouth.2020.105285

Watt, T., Kim, S., & Garrison, K. (2018). The relationship between state supports and post-secondary enrollment among youth aging out of foster care. Child Welfare, 96, 1-20. https://www.jstor.org/stable/4862361

Watt, T., Norton, C., & Jones, C. (2013). Designing a campus support program for foster care alumni: Preliminary evidence for a strengths framework. Children and Youth Services Review, 35, 1408-1417. https://doi.org/10.1016/j.childyouth.2013.06.002

Yamaguchi, K. (1991). Event history analysis. Sage Publications.

HOW TO CITE THE PAPER

AUTHOR’S ADDRESS

ACADEMIC PROFILE