Individual and Institutional Facilitators and Barriers to Reentry Preparedness Among Detained and Committed Youth

Abstract

Youth who are detained and committed to juvenile justice facilities often struggle to successfully reintegrate into their communities following release. Research has found that there are many individual- and institutional-level barriers that can complicate the reentry process. The development of comprehensive transition plans can be helpful as youth navigate the reintegration process and reenroll in school or obtain employment. Using youth surveys and administrative data from the Center for Improving Youth Justice’s Performance-based Standards National Database for Researchers, this study explores the individual- and institutional-level factors affecting youths’ perceptions of reentry preparedness. Results indicate that youth who received assistance with their educational and intrapersonal skills while incarcerated were more likely to feel prepared, whereas youth who faced structural barriers were less likely to feel prepared. We conclude with a discussion of the policy implications for improving the reentry process for youth.

1. Introduction

On any given day, there are approximately 25,000 youth confined in juvenile justice residential facilities across the United States (Puzzanchera 2022). Upon their release and return to the community, many youth will be at risk of recidivating. One-year rearrest rates range between 33% and 55% (Snyder and Sickmund 2006; Robertson et al. 2020; Florida Department of Juvenile Justice 2023) and increase to between 70% and 80% two or three years after release (Mendell 2011). Youth who are reentering the community often lack prosocial support systems, come from low-income communities, and struggle with mental illnesses or substance abuse. Additionally, they have histories of low academic achievement and engagement with school and will return to disadvantaged communities where education and employment opportunities are limited, which likely increases the risk of recidivism (Garfinkel and Nelson 2004; Hsia et al. 2004; Steele et al. 2016).

Despite the challenges faced by this population, educational achievement during residential commitment or detention can be a critical element in the establishment of a prosocial life and desistance. For example, prior research has found that youth who receive a high-quality education and experience academic gains while in residential detention or commitment are less likely to recidivate, more likely to return to school, more likely to obtain employment, and more likely to exhibit a range of positive behavioral outcomes (Blomberg et al. 2011, 2012; Bullis et al. 2002; Elliott 1994; Foley 2001; Griller Clark et al. 2011; Jaggi et al. 2020; Jenson and Howard 1998; Unruh et al. 2009; Cavendish 2014). Importantly, educational attainment can mediate the effects of juvenile justice system involvement on adult employment outcomes. Specifically, youth with lower educational attainment have worsened employment outcomes (Tanner et al. 1999). Thus, completing at least a high school education may serve as a turning point and improve employment prospects (Natsuaki et al. 2008).

Despite making academic gains while in residential detention or commitment, research has shown that these gains may be short-lived if youth do not reenroll in school following release. However, the transition from juvenile justice facilities to the community—back to school in particular—is fraught with challenges and barriers. Notably, only about one-third of youth reenroll in school following release from juvenile confinement (Aizer and Doyle 2015; United States Department of Education 2016). Even if youth do reenroll in school, they regularly struggle to stay, which increases their risk of failure and drop-out and sets them on a path toward negative life outcomes (Hsia et al. 2004; Garfinkel and Nelson 2004; Juvenile Justice Educational Enhancement Program 2006; MacArthur Foundation 2005; Risler and O’Rourke 2009; Steele et al. 2016; Wald and Losen 2003). Although education is an important conduit for gainful employment (Bunting et al. 2019; Lockwood et al. 2015; Tanner et al. 1999), few formerly committed youth will graduate from high school or complete the requirements for a GED, and even fewer will go on to college (Aizer and Doyle 2015; United States Department of Education 2016). For example, an evaluation of eighth- and ninth-grade public school students found only 16% of youth who experienced juvenile detention graduated from high school compared to 72% of non-detained students, and 37% of detained students enrolled in postsecondary education compared with 51% of non-detained students (Gertseva and McCurley 2019). These educational deficits can significantly limit the employment prospects of those who have been involved in the justice system.

One of the most promising initiatives to improve youths’ reentry outcomes has been the development of individualized transition planning and services. Transition planning begins while the juvenile is in confinement and extends beyond their release (Altschuler and Armstrong 1994; Hirschfield 2014; United States Department of Education and United States Department of Justice 2014; Wiebush et al. 2000). To be effective, transition plans should be comprehensive and detail youths’ needs and goals, identify tangible strategies for accomplishing goals, and facilitate necessary community-based connections (Mathur et al. 2021; United States Department of Education and United States Department of Justice 2014). Transition planning may also contribute to youths’ perceptions of readiness for release and community reentry, which may, in turn, facilitate more successful outcomes with independent living, education, and careers (Melkman et al. 2015; Osborn and Belle 2019).

While research on community reintegration and school reentry has identified facilitators and barriers, it is largely dated and has rarely included the perspective of incarcerated juveniles. Youth may have unique experiences and perspectives about their transition plans that can better identify more relevant and meaningful programs and policies (Kirshner et al. 2005; Kubek et al. 2020; Powers and Tiffany 2006; Schubert et al. 2014). Renewed scholarly attention is warranted and could provide needed information on factors that contribute to effective reintegration and, subsequently, improved education and employment outcomes among youth involved in the juvenile justice system (Kupchik and Snyder 2009; Schubert et al. 2014).

To address this gap, this study uses youth surveys and administrative data from the Center for Improving Youth Justice’s Performance-based Standards National Database for Researchers to examine self-reported educational and employment reentry preparedness among youth exiting juvenile residential commitment programs across the United States. Examining youths’ feelings of preparedness is important because, while many youth have high aspirations for their post-release outcomes (see, for example, Abrams 2006; Gardner et al. 2022; Hartwell et al. 2010; Noorman and Brancale 2022), being able to follow through is a challenge. Thus, this study contributes to this gap in the literature by assessing what factors may lead youth to feel more prepared for their release and, subsequently, more likely to follow through with their aspirations.

2. Prior Literature

2.1. Importance of Education and Employment for Successful Post-Release Outcomes

Juvenile justice-involved youth disproportionately have lengthy histories of poor school performance, low attachment to school, low literacy levels, and high rates of intellectual, developmental, learning, and emotional disabilities compared to non-justice-involved youth (Cavendish 2014; Development Services Group, Inc. 2017; Pfannenstiel 1993; Wang et al. 2005). However, research has suggested that providing ample and high-quality educational opportunities and assistance to youth who are in confinement facilities is an effective approach for preventing future delinquency and improving education and employment outcomes following release (Blomberg et al. 2011, 2012; Foley 2001; Jaggi et al. 2020; Jenson and Howard 1998).

Prior research has consistently demonstrated the positive impacts on post-release outcomes for youth who make academic gains while in detention or commitment, including improvements in behavior, increased likelihood of returning to school, increased likelihood of obtaining employment, and reduced likelihood of recidivism (Ambrose and Lester 1988; Blomberg et al. 2011, 2012; Cavendish 2014; Griller Clark et al. 2011; Elliott 1994; Foley 2001; Jaggi et al. 2020; Jenson and Howard 1998; Steele et al. 2016; Unruh et al. 2009). Similarly, Jaggi et al. (2020) found that school attachment during incarceration was associated with decreased delinquency 12 months following release. Further, youth who return to school after release were less likely to recidivate (Blomberg et al. 2011; Bullis et al. 2002) and, if rearrested, for significantly less serious offenses (Blomberg et al. 2011).

Just as education has been found to reduce recidivism among youth, steady employment has been found to reduce recidivism among adults (Bunting et al. 2019; Lockwood et al. 2015; Nally et al. 2014). Importantly, employment prospects are often directly impacted by and tied to education status. Among the general population, the unemployment rate for those with less than a high school diploma is 5.6%, compared to 3.9% for those with a high school diploma and 2.2% for those with a bachelor’s degree (Bureau of Labor Statistics 2023). Among justice-involved individuals, those with lower levels of education are more likely to be unemployed after release from incarceration and, thereby, are more likely to recidivate (Bunting et al. 2019; Lockwood et al. 2015; Tanner et al. 1999). Thus, ensuring that youth who have been involved in the juvenile justice system are adequately prepared to continue their education following release is an important step toward their subsequent employment attainment.

2.2. Importance of Transition Planning

Transition planning often begins with an assessment to identify areas of psychological, social, and educational needs among youth in detention and residential commitment facilities. Once the assessment has been completed, individualized planning, goal setting, and community connections are established (Platt et al. 2015). Importantly, to be most effective, case workers from the juvenile justice facility should work alongside and involve youths’ families and other support systems in the transition planning and follow-up care (Abrams 2006; United States Department of Education and United States Department of Justice 2014).

Westat, Inc. (1991) and the Research Triangle Institute (1999) found that transition services aimed at assisting youth with returning to school and post-release employment, at best, varied widely across institutions and, at worst, were completely non-existent. Although transition programming is not highly prioritized by states, research has found that supportive programming that begins before the juvenile exits residential commitment can be instrumental in desistance and successful educational and employment pursuits (Mathur and Griller Clark 2013; Platt et al. 2015).

Another important aspect of transition planning may be related to how simply having a plan in place upon release contributes to youths’ overall feelings of preparedness. Only a handful of studies have explored youths’ perceptions of readiness, and all have found that the more ready a youth feels for release, the better their post-release outcomes will be (Clinkinbeard and Zohra 2012; Melkman et al. 2015; Osborn and Belle 2019). While research has found that transition plans are beneficial for post-release success, primarily due to structural factors such as ensuring the youth knows which school to attend after release and providing them with essential documents, gaps remain in our understanding of how prepared youth feel for release and whether youth perceive their plans as helpful. This is an area in need of research.

2.3. Barriers to Community Education Reentry

Research on community reintegration among youth exiting residential commitment programs has focused on living arrangements and education, with less focus on employment, especially immediate job attainment. Therefore, the following section will present research on the individual- and institutional-level barriers to education reentry that may prevent youth from successfully transitioning back to school after their release from a juvenile justice facility.

Although youth have reported high aspirations for their education following release (Abrams 2006; Gardner et al. 2022; Hartwell et al. 2010; Noorman and Brancale 2022; Toldson et al. 2010), few justice-involved youth become engaged in school following release and even fewer remain engaged long-term (Bullis et al. 2002; United States Department of Education 2016). Specifically, only about one-third of youth reenroll in school following release (Aizer and Doyle 2015), and engagement declines over time, with Bullis et al. (2002) reporting a drop in engagement from 47% at 6 months following release to 31% at 12 months following release. Thus, another critical aspect for youths’ post-release success is ensuring effective community transition and reintegration plans and processes are in place. An important line of research involves exploring which factors facilitate or impede the successful transition and reintegration from a juvenile justice facility to a community school.

2.3.1. Individual-Level Factors

Individual-level factors are personal characteristics and experiences that may impact successful transitioning. Prior research has identified several individual-level correlates that may make transitioning back to school more difficult, including a lack of family, school, and social supports, poor academic skills and being far behind in school, negative peer associations, and struggling with mental illness or substance use (Baltodano et al. 2005; Feierman et al. 2010; Gardner et al. 2022; Garwood 2015; Kubek et al. 2020; Marshall et al. 2012; Mathur and Griller Clark 2014; Miller et al. 2019; O’Neill et al. 2017; Siennick and Staff 2008; Sinclair et al. 2021; Unruh 2005; Unruh and Bullis 2005; Wallace 2012). Prior research has also suggested that students lacking the appropriate required records and documents necessary for school enrollment (e.g., birth certificate, residency verification, immunization records) may also have difficulty returning to school (Feierman et al. 2010; Marshall et al. 2012; O’Neill et al. 2017; Wallace 2012).

A recent study by Sinclair et al. (2021) interviewed eight young offenders participating in a reentry program who reported substance abuse, poverty, and gang relations to be their most pervasive barriers to school reentry. Additional barriers included transportation problems, truancy, previous academic issues, and negative community influences. Garwood (2015) interviewed 10 young adults who were incarcerated as juveniles and later earned their high school diploma or GED about the challenges they experienced during reentry. These young adults reported their most pervasive barriers included the lack of a support system, challenges associated with their incarceration experience, and having to overcome their own patterns and habits related to school and delinquency. Gardner et al. (2022) interviewed 90 students in Chicago about their experiences with returning to school after incarceration and found many struggled to balance their desire to return to school with concerns about basic survival, such as maintaining stable housing and providing for children.

Several studies have examined youths’ perceptions of their barriers to successful transition more generally (see, for example, Abrams 2006; Hartwell et al. 2010), with fewer exploring youths’ perceptions of the barriers they face returning to school. The literature examining general transition experiences suggests that while youth make logistical plans for their return to school, relatively few anticipate challenges related to education (Abrams 2006; Hartwell et al. 2010). Rather, youth are generally more concerned about and experience challenges with reengaging with negative peer influences, economic and housing instability, and transportation problems. Notably, youth reported overcoming these challenges often took precedence over enrolling in school (Abrams 2006).

While the literature on youths’ perceptions of their barriers to education reentry is growing, current studies are limited by small and nonrepresentative samples. To the authors’ knowledge, only one recent study has utilized data from a nationally representative sample of incarcerated youth to explore their perceived barriers to school reentry. Using survey data from 447 youth in residential detention and commitment, Noorman and Brancale (2022) explored youths’ self-reported barriers to achieving their educational aspirations. The authors found the strongest predictor for youth not expecting to achieve their educational aspirations was a lack of interest in school. Notably, despite prior research, having a history of poor academic performance was not a significant predictor of lower educational expectations. While there was an improvement in sample size, the generalizability of the findings was limited due to the exclusion of youth with higher educational aspirations and expectations.

2.3.2. Institutional-Level Factors

In addition to the individual-level barriers youth face when attempting to return to school, factors related to the characteristics of juvenile justice facilities and the services they provide may also impact youths’ abilities to successfully return to school. Juvenile justice facilities are required to offer educational services to detained and committed youth on a consistent basis. However, the delivery of these services can vary widely. Further, there are numerous challenges associated with operating educational programs within juvenile justice settings, which have led to reports of low-quality education for the justice-involved youth population for decades (Blomberg and Lucken 2010; Macomber et al. 2010; Pace 2018; Pesta 2012; Pesta and Blomberg 2016; Rothman 2002). For example, most programs are relatively small in size; students have, on average, short lengths of stay; there is high student mobility; and students often have disproportionate educational deficiencies compared to their public school counterparts. Leone and Cutting (2004) identified some of the challenges as being characteristic of the incarcerated youth population, the individualized operation of the facilities, and the limited connection between the juvenile facility and local public schools.

In addition, juvenile justice facilities have historically operated with limited financial resources, inadequate space, a lack of qualified staff and teachers, classrooms with students of varying grade levels and educational abilities, and ineffective accountability measures (Leone et al. 1986; Leone and Cutting 2004; Mechlinski 2001; Pace 2018). State juvenile justice education leaders have also reported challenges in competing with public schools to recruit and retain teachers, a lack of coordination with local community schools to better transition youth to and from juvenile justice facilities, and difficulty in maintaining and measuring student educational progress due to high rates of student mobility (Juvenile Justice NCLB Collaboration Project 2008).

2.4. Gaps in the Research

As described previously, research on perceptions of the barriers youth encounter when returning to the community following release from a juvenile justice facility has been largely missing (Kubek et al. 2020). In a recent systematic review of school reentry practices, Kubek et al. (2020) identified only 27 articles meeting the initial selection criteria, of which only nine met the criteria for a full review and only three were published in the last 10 years. Notably, most of the articles reviewed by Kubek et al. (2020) were based on descriptive case studies and self-reported perspectives of key stakeholders. Largely missing were the voices of students and families—those most directly impacted by detention or commitment and the school and community reentry process. The inclusion of youth voices is important as it can offer deeper insights into youths’ unique experiences and perspectives, which might have otherwise gone unnoticed, unexamined, or misinterpreted by practitioners and policymakers (Kirshner et al. 2005; Schelbe et al. 2015). In turn, these insights can help identify more relevant and meaningful programs and policies to address youths’ needs (Kirshner et al. 2005; Powers and Tiffany 2006; Schubert et al. 2014).

Since the publication of Kubek et al.’s (2020) systematic review, only a handful of studies have attempted to incorporate the voice of youth (see, for example, Gardner et al. 2022; Noorman and Brancale 2022; Sinclair et al. 2021). However, these recent studies did not address sample size limitations or potential lack of generalizability. Research has also not included youth’s perceptions of employment preparation or readiness. In this study, we seek to bring renewed attention to the experiences of confined juveniles by exploring the individual- and institutional-level factors that influence youths’ self-reported preparedness for reentry, with a focus on the perceived helpfulness of their long-term education and employment plans. Importantly, the present study expands upon the available literature and improves the generalizability of findings by using a large national sample of youth incarcerated in facilities across the United States. In doing so, this study is guided by the following research questions:

Research Question 1. Do youth feel their education and employment plans are helpful?

Research Question 2. What individual and institutional characteristics and experiences helped youth feel most prepared for educational and employment reentry?

3. Data and Methods

Data for this study come from the Center for Improving Youth Justice’s Performance-based Standards (PbS) National Database for Researchers. This database offers researchers access to over 10 years of comprehensive quantitative and qualitative data on juvenile justice facility life, practices, services, and programs across more than 300 correction, detention, and assessment facilities located within the United States (Center for Improving Youth Justice n.d.). On a biannual basis, all facilities that participate in the Center for Improving Youth Justice’s PbS program are asked to participate in a series of surveys and administrative data collection to provide performance data in the areas of facility safety; order and security; health, behavioral health, substance use, education, and reentry services; connection with families; and perceptions of fairness and staff-youth relationships. The resulting database includes Administrative Forms, incident reports, outcome measures, staff climate surveys, youth climate surveys, and youth reentry surveys. This study uses data from the Youth Reentry Survey (YRS), Youth Climate Survey (YCS), and Administrative Form.

The YRS is a youth-based self-report survey administered to youth shortly before they leave secure placement or when they end post-placement supervision. The YRS asks youth about their perceptions of their preparedness and readiness to return to their community and live independently. Specifically, the YRS asks youth questions about their perceptions of fairness and safety; the skills they have learned; their relationships with their families and case manager; their sense of connection to the community; their confidence, hope, resiliency, and willingness to show up; and whether or not they feel prepared to take action (Godfrey 2019). This study used data from youth reentry surveys submitted between October 2019 and April 2022 to identify individual-level characteristics and experiences that may facilitate or impede reentry preparedness.

The YCS is a biannual self-report survey administered to a random selection of 30 youths residing within each facility to collect feedback on living conditions/climate, understanding rules and rights, facility programs, family contact, safety and security, and staff and justice (PbS Learning Institute 2022). The PbS Administrative Form is collected biannually for each facility and contains both general and specific information about each facility, including the number of youths and staff, types of assessments, and facility programs using volunteers (PbS Learning Institute 2022). Data from the YCS and Administrative Form submitted between April 2017 and October 2021 were used to identify institutional-level characteristics and experiences that may facilitate or impede reentry preparedness. To create the institutional-level measures described below, an average of each measure across the five-year data collection period was created for each juvenile residential facility.

3.1. Analytic Sample

Drawing from a national sample of juvenile residential facilities, this study focused on youth exiting a juvenile residential treatment program or facility between October 2019 and April 2022. The analytic sample for this study included 5175 youth across 104 juvenile residential facilities.1 Most youth were between the ages of 16 and 18, and about 90% were male. Most of the youth identified themselves as Black (35.75%), White (31.52%), and Hispanic (19.94%). Most youth were exiting a correctional facility compared to a detention or assessment facility. Over half of the youth exited facilities located in rural areas, and about 40% of youth exited facilities located in the Western United States.2

3.2. Dependent Variable

The current study examines how prepared youth feel to return to education and employment upon release from secure placement. Specifically, the dependent variable for this study comes from a series of questions on the YRS about youths’ reentry plans and asked youth how much they agree or disagree with the statement that they “have a plan for their long-term education and employment that is helpful”. Responses were measured on a four-point Likert scale ranging from Strongly Disagree to Strongly Agree. Due to the small number of responses in the Disagree and Strongly Disagree categories, these two response categories were collapsed into one.3 Therefore, the dependent variable is coded 1 = Disagree/Strongly Disagree, 2 = Agree, and 3 = Strongly Agree.

3.3. Individual-Level Variables

The individual-level variables for the current study come from the YRS and include the following domains: helpful experiences, reentry plan, case management, essential documents possession, living arrangements, aftercare programming, and community activities. The YRS survey question measuring helpful experiences asked youth to identify which experiences with staff and case managers helped the youth be most ready for reentry. The present study included three variables from this question: “helped me with my education/GED”, “helped me better understand my strengths and talents”, and “taught me job skills”. These variables were coded dichotomously (1 = Yes, 0 = No). Feeling supported by staff in schoolwork and life skills has been shown to increase school attachment (Reed and Wexler 2014; Scales et al. 2020), which can improve school attendance and reduce delinquency following release (Jaggi et al. 2020; Jaggi and Kliewer 2020). Furthermore, a lack of formal job training can contribute to higher unemployment and recidivism rates (Lockwood et al. 2015; Nally et al. 2014).

The YRS survey questions measuring reentry plan asked youth how much they agree or disagree with a set of statements about their reentry plan. The present study included seven variables from this series of questions: “I have transportation to get to school and/or work”, “It will be easy to pay my rent/living expenses”, “I have enough money to buy food and clothing”, “I have the supports I need for a successful reentry”, “I am confident I will achieve my reentry goals”, “I understand what is expected of me when I leave”, and “I can comply with/meet the expectations of my reentry plan”. These variables were measured on a four-point Likert scale: 1 = Strongly Disagree, 2 = Disagree, 3 = Agree, and 4 = Strongly Agree. The reference category for the analysis was 4 = Strongly Agree. Transition planning has been identified as a promising practice to improve reentry outcomes (Altschuler and Armstrong 1994; Hirschfield 2014; Wiebush et al. 2000). Among these transition planning measures, prior research has found that addressing challenges related to economic stability and transportation often takes precedence over returning to school (Abrams 2006; Gardner et al. 2022; Sinclair et al. 2021). In addition to lacking structural support, lacking a social support system has been associated with barriers to reentry (Garwood 2015; Hartwell et al. 2010). Lastly, feeling confident about goals and expectations and being able to make concrete investments toward achieving those goals are likely to increase youths’ ability to reach their educational aspirations (Mahler et al. 2018; Siennick and Staff 2008).

The YRS survey questions measuring case management asked the youth how true the provided statements were about their case manager. These statements included topics about feeling supported and cared for by their case managers and receiving appropriate assistance. The present study focused on the assistance received by case managers and included two variables from this series of questions: “My case manager helps me connect with people and resources that will help me succeed” and “My case manager helps me work through barriers that could stop me from achieving my goals”. These variables were measured on a four-point Likert scale: 1 = True, 2 = Mostly True, 3 = Somewhat True, and 4 = Not True At All. The reference category for the analysis was 1 = True.

The YRS survey question measuring essential document possession asked youth to identify which items from a list they have in their possession or can easily obtain if they need them. The listed items included a valid ID (license, state ID, or school ID), birth certificate, Social Security card, passport, medical records, immunization records, prescriptions, and a cell phone. Each of these was included in the analyses and was coded dichotomously (1 = Yes, 0 = No). Due to the identification and medical requirements for public school enrollment, lacking these essential documents may result in youth not being able to reenroll in school (Feierman et al. 2010; Wallace 2012).

Living arrangements were measured in the YRS by asking youth where they will sleep most of the time after release. Having unstable housing has been identified as a barrier to reentry among youth, and addressing housing concerns may outweigh reenrolling in school (Abrams 2006; Gardner et al. 2022). Response categories included at home with family, friend’s house, with significant other, shelter, couch surfing, outdoors, car, transitional housing, programs/facilities, and other place. Factor analysis was conducted to determine what, if any, underlying structure exists for the measures. Principal components analysis produced a three-component solution. Component 1 included the variables of sleep at a shelter, couch surfing, outside, and in a car. This component was labeled unstable living arrangements. Component 2 included sleeping at transitional housing, at a program/facility, and at home with family. This component was labeled stable living arrangements. While transitional housing may seem inherently unstable, it is often characterized as a longer-term shelter that provides services to support individuals in their transition to self-sufficient living (Family & Youth Services Bureau 2023). Component 3 included sleeping at a friend’s house and with a significant other. This component was labeled semi-stable living arrangements. Component scores were calculated using the post-estimation predict command and were mean centered for inclusion in the analysis. The component scores are as follows: unstable living arrangements (3.59 × 10⁻⁹, SD = 1.36, min = −1.18, max = 6.93), semi-stable living arrangements (2.16 × 10⁻⁸, SD = 1.15, min = −1.96, max = 4.87), and stable living arrangements (3.34 × 10⁻⁹, SD = 1.34, min = −0.61, max = 17.13).

The YRS question measuring aftercare programming asked the youth if they will be going to any community services, aftercare, or other kind of program (e.g., counseling, drug programs, or job services) when they leave the facility. The response categories were coded 1 = Yes, 0 = No/Don’t Know. The YRS question measuring educational financial advising asked youth if, over the past year or so, they have ever been told about ways to pay for college (e.g., financial aid packages and loans). This variable was coded dichotomously (1 = Yes, 0 = No).

3.4. Institutional-Level Variables

The quality of education services received by youth while in juvenile justice settings may improve or worsen youths’ educational outcomes (Hjalmarsson 2008). Specifically, education while incarcerated may improve educational outcomes for those youth coming from low-quality schools; however, educational outcomes may worsen if the quality of the education inside the juvenile justice setting is poor. There are a variety of institutional-level factors that may impact the quality of education youth receive in juvenile justice settings, including program enrollment, having a disproportionate number of youths with educational deficiencies, average lengths of stay, and hiring and retaining qualified teachers (Leone et al. 1986; Leone and Cutting 2004; Mechlinski 2001; Pace 2018). The institutional-level variables included in the analysis aim to capture these factors.

The institutional-level variables for the study come from the YCS and the Administrative Form. These variables were created to gauge programming, youth population, and staffing at each facility. Because individual responses could not be matched to the YRS, this study used youth-level measures to create an aggregate of each variable from April 2017 to October 2021 for each of the 104 juvenile residential facilities identified in the YRS.4 The first set of variables from the YCS measure program enrollment. The first enrollment measure we created was the proportion of youth in each facility who attend school by summing the number of youths in each facility that indicated they have been attending school since entering their facility and taking the mean. Our measure of the proportion of youth in each facility with a GED/high school diploma was created by summing the number of youths in each facility who indicated that they have a GED or high school diploma and taking the mean. Our final enrollment measure was the proportion of youth in each facility with a treatment or service plan, created by summing the number of youths in each facility who indicated that they have a treatment or service plan and taking the mean.

In addition to measures of program enrollment, we also used information from the YCS to create measures reflecting program quality at each facility. Our first program quality measure was the proportion of youth in each facility who indicated the facility has a good school program, which was created by summing the number of youths in each facility who indicated their facility has a good school program and taking the mean. We also created a measure of the proportion of youth in each facility who felt the programming helped them understand how to succeed. This measure was created by summing and then taking the mean of the number of youths in each facility who responded True or Somewhat True to the statement that the programming (i.e., school, counseling, or other program) they go to is helping them to understand what they need to do to succeed when they return home. Lastly, we created an index measuring the average staff climate in each facility, with higher scores indicating a better climate. This index was created by first summing youth responses to a series of questions about treatment by staff members—with response categories including 1 = No, 2 = Sometimes, and 3 = Yes—and then creating an average of the youth responses at each facility.

Using the Administrative Form, we created measures of facility characteristics that may influence program availability and quality. The first measure was the average population size at each facility, created by taking the mean of the youth population size at each facility from 2017 to 2021. Next, we created a measure of the average length of stay in days per facility by taking the mean of the average length of stay at each facility from 2017 to 2021. To gauge facility staffing levels, we created measures of the average number of new staff per facility and the average number of staff leaving each facility. To create these measures, we took the average of the number of direct care staff who began work or left the facility in the last 6 months, respectively, across reporting periods from 2017 to 2021.

3.5. Control Variables

The present study controls for factors at both the individual-level and institutional-level that may impact our outcome. At the individual level, we used data from the YRS to control for youth demographics, including age, race/ethnicity, and gender. Age was a continuous variable, ranging from 8 to 24, and was mean centered for the analysis.5 Racial/ethnic categories included White (alone), Black (alone), Hispanic (any), American Indian/Alaskan, Native/Asian/Native Hawaiian/Other Pacific Islander, and Other. For the analysis, White served as the reference group. Gender was coded dichotomously (1 = Male, 0 = Female).

At the institutional level, data from the YRS were used to control for the type of facility and facility location. Facility type included correctional, detention, and assessment; facility area included rural, suburban, and urban; and facility region included Midwest, Northeast, South, and West. For the analysis, correction facility, rural, and Midwest served as reference categories for facility type, facility area, and facility region, respectively.

3.6. Analytic Strategy

The analytic methods used in this study included descriptive statistics and multilevel ordinal logistic regression. Descriptive statistics were used to identify the level of agreeableness among youth in their preparedness for reentry. Given that the youth were nested in juvenile residential facilities (i.e., two levels of data), multilevel/hierarchical models were the most appropriate technique to use (Raudenbush and Bryk 2002). Since the dependent variable is an ordinal measure, we used multilevel ordinal logistic regression to identify which individual- and institutional-level factors helped youth feel more prepared for reentry. Specifically, we used Stata 18’s meologit command to estimate a two-level random intercept model, using the Facility ID as the variable for the random-effects equation (StataCorp 2023). We first estimated the unconditional model, which showed that there is statistically significant between-facility variation in reentry preparedness (χ² = 145.52, p < 0.001) and about 6.7% of the total variance can be explained by variation between facilities. We then estimated two conditional models, introducing the individual-level variables (Model 1) and then adding the institutional-level variables (Model 2).6

4. Results

Descriptive statistics for the sample are presented in

Table 1. Descriptive statistics.

	Mean, %	SD	Minimum	Maximum
Individual-level variables
Education plan	2.53	0.55	1	3
Strongly Agree	55.77
Agree	41.64
Disagree/Strongly Disagree	2.59
Helpful experiences
Education and GED	0.58	0.49	0	1
Talents and strengths	0.64	0.48	0	1
Job skills	0.61	0.49	0	1
Reentry plan
Transportation	3.38	0.65	1	4
Strongly Agree	46.49
Agree	46.63
Disagree	5.62
Strongly Disagree	1.26
Living expenses	3.13	0.72	1	4
Strongly Agree	31.69
Agree	51.90
Disagree	14.53
Strongly Disagree	1.87
Food expenses	3.30	0.70	1	4
Strongly Agree	42.59
Agree	46.98
Disagree	8.64
Strongly Disagree	1.80
Supports	3.52	0.55	1	4
Strongly Agree	54.72
Agree	43.19
Disagree	1.70
Strongly Disagree	0.39
Confident in goals	3.55	0.54	1	4
Strongly Agree	57.26
Agree	41.14
Disagree	1.29
Strongly Disagree	0.31
Understand expectations	3.65	0.50	1	4
Strongly Agree	66.14
Agree	33.02
Disagree	0.64
Strongly Disagree	0.19
Comply with expectations	3.58	0.53	1	4
Strongly Agree	59.56
Agree	39.13
Disagree	1.06
Strongly Disagree	0.25
Case manager
Connect with resources	1.50	0.84	1	4
True	68.54
Mostly True	17.33
Somewhat True	10.01
Not True At All	4.12
Barriers	1.51	0.85	1	4
True	68.64
Mostly True	16.58
Somewhat True	10.32
Not True At All	4.46
Essential documents
Valid ID	0.716	0.45	0	1
Birth certificate	0.793	0.41	0	1
Social security card	0.760	0.43	0	1
Medical records	0.502	0.50	0	1
Immunization records	0.331	0.47	0	1
Prescriptions	0.368	0.48	0	1
Cell phone	0.699	0.46	0	1
Living arrangements
At home with family	0.858	0.350	0	1
At friend’s house	0.166	0.372	0	1
With significant other	0.135	0.341	0	1
At a shelter	0.010	0.100	0	1
Couch surfing	0.015	0.120	0	1
Outdoors	0.016	0.126	0	1
In a car	0.013	0.114	0	1
Transitional housing	0.071	0.257	0	1
Program/facility	0.086	0.120	0	1
Other	0.039	0.193	0	1
Community activities
Educational financial advising	0.76	0.43	0	1
Aftercare programming
Community program	0.57	0.50	0	1
Demographics
Age	17.09	1.45	8	24
Race/Ethnicity	2.22	1.18	1	5
White	31.52
Black	35.75
Hispanic	19.94
American Indian/Alaskan Native	4.31
Other	8.48
Gender (male = 1)	0.90	0.30	0	1
Facility-level variables
Facility type	1.04	0.23	1	3
Correction	97.37
Detention	1.78
Assessment	0.85
Facility area	1.61	0.77	1	3
Rural	56.62
Suburban	25.78
Urban	17.60
Facility region	2.71	1.28	1	4
Midwest	28.87
Northeast	13.26
South	15.65
West	42.22
Proportion good school	0.76	0.13	0.375	1
Proportion attend school	0.88	0.09	0.368	1
Proportion youth has GED/diploma	0.06	0.06	0	0.412
Proportion find school helpful	0.80	0.11	0.279	1
Proportion with treatment plan	0.67	0.18	0.191	1
Proportion felt help understand	0.88	0.07	0.651	1
Facility staff climate	28.94	5.06	20	45.947
Average population size	2429.16	1501.76	37	6624.556
Average length of stay	246.01	115.96	11.255	912.85
Average staff hired	19.29	15.59	0	54.889
Average staff leaving	18.00	14.03	0	74

Note. Individuals (n = 5175); facilities (n = 104).

. Most youth strongly agreed (55.77%) or agreed (41.64%) that they have a plan for their long-term education and employment that is helpful. Notably, very few youths disagreed/strongly disagreed (2.59%) that they have a helpful long-term plan. Over half of the youth found receiving help with their education/GED (58%), receiving help to better understand their strengths and talents (64%), and being taught job skills (61%) helped them feel most ready for reentry. Further, most youth had a positive outlook on the components of their reentry plan. For example, 46.49% and 46.63% of youth strongly agreed or agreed, respectively, that they had transportation to travel to school and/or work. Similarly, 57.26% and 41.14% of youth strongly agreed or agreed, respectively, that they were confident they could achieve their reentry goals. Additionally, 76% of youth reported receiving educational financial advice, and 57% reported they would be attending some type of aftercare community program (e.g., counseling, drug program, or job services).

The results of the multilevel ordinal logistic regression model predicting the odds of being in higher agreement on reentry preparedness are presented in

Table 2. Multilevel ordinal logistic regression results for educational and employment reentry preparedness.

Variable	Model 1		Model 2
	OR	SE	OR	SE
Helpful experiences
Education and GED	1.335 ***	0.120	1.344 ***	0.122
Talents and strengths	1.283 **	0.119	1.306 **	0.122
Job skills	1.065	0.100	1.092	0.352
Reentry plan a
Transportation
Agree	0.542 ***	0.069	0.545 ***	0.069
Disagree	0.518 ***	0.104	0.513 ***	0.104
Strongly Disagree	0.490	0.194	0.485	0.193
Living expenses
Agree	0.499 ***	0.064	0.496 ***	0.064
Disagree	0.432 ***	0.069	0.427 ***	0.069
Strongly Disagree	0.632	0.209	0.637	0.212
Food expenses
Agree	0.987	0.130	0.991	0.131
Disagree	1.054	0.195	1.061	0.197
Strongly Disagree	0.809	0.292	0.804	0.292
Supports
Agree	0.479 ***	0.055	0.477 ***	0.055
Disagree	0.152 ***	0.050	0.149 ***	0.049
Strongly Disagree	0.490	0.417	0.495	0.425
Confident in goals
Agree	0.286 ***	0.031	0.285 ***	0.031
Disagree	0.073 ***	0.029	0.074 ***	0.029
Strongly Disagree	0.073 **	0.069	0.071 **	0.067
Understand expectations
Agree	0.417 ***	0.049	0.413 ***	0.049
Disagree	0.258 **	0.131	0.253 **	0.129
Strongly Disagree	0.017 ***	0.019	0.017 ***	0.020
Comply with expectations
Agree	0.279 ***	0.032	0.273 ***	0.032
Disagree	0.306 **	0.131	0.305 **	0.131
Strongly Disagree	0.654	0.776	0.648	0.769
Case manager b
Connect with resources
Mostly True	0.873	0.113	0.850	0.110
Somewhat True	0.764	0.133	0.769	0.135
Not True At All	0.773	0.202	0.790	0.207
Barriers
Mostly True	0.911	0.119	0.921	0.121
Somewhat True	0.943	0.163	0.950	0.164
Not True At All	1.202	0.302	1.190	0.300
Essential documents
Valid ID	1.065	0.102	1.052	0.102
Birth certificate	0.979	0.131	0.982	0.133
Social security card	0.914	0.117	0.906	0.117
Medical records	1.155	0.119	1.173	0.121
Immunization records	1.069	0.123	1.087	0.125
Prescriptions	1.036	0.107	1.045	0.108
Cell phone	0.907	0.087	0.911	0.088
Living arrangements
Stable living arrangement	1.043	0.033	1.045	0.033
Semi-stable living arrangement	0.939	0.034	0.933	0.034
Unstable living arrangement	0.987	0.033	0.994	0.033
Community activities
Educational financial advising	1.459 ***	0.138	1.491 ***	0.143
Aftercare programming
Community program	1.171	0.099	1.187 *	0.103
Demographics
Age	0.971	0.030	0.966	0.032
Race/Ethnicity c
Black	1.075	0.113	1.047	0.114
Hispanic	1.294 *	0.161	1.344 *	0.173
American Indian/Alaskan Native	0.841	0.169	0.931	0.191
Other	0.900	0.141	0.923	0.145
Gender	0.860	0.129	0.808	0.123
Facility type d
Detention	---	---	0.768	0.252
Assessment	---	---	0.948	0.458
Facility area e
Suburban	---	---	1.071	0.131
Urban	---	---	1.101	0.134
Facility region f
Northeast	---	---	0.950	0.190
South	---	---	1.054	0.167
West	---	---	0.932	0.127
Proportion good school	---	---	0.884	0.427
Proportion attend school	---	---	2.276	2.866
Proportion youth has GED/diploma	---	---	3.861	6.815
Proportion with treatment plan	---	---	0.918	0.422
Proportion felt help understand	---	---	0.725	1.109
Facility staff climate	---	---	1.012	0.019
Average population size	---	---	1.000	0.000
Average length of stay	---	---	0.999	0.001
Average staff hired	---	---	1.001	0.008
Average staff leaving	---	---	1.005	0.009
Random Effects	Variance Component		Variance Component
	0.005		2.04 × 10−34

Note. Results are based on 5175 youth within 104 juvenile residential facilities. * p < 0.05, ** p < 0.01, *** p < 0.001. ^a Reference group is Strongly Agree. ^b Reference group is True. ^c Reference group is White. ^d Reference group is Correction. ^e Reference group is Rural. ^f Reference group is Midwest.

. Model 1 in Table 2 shows the relationship between youth-level factors and feeling prepared for reentry. The results show that youth who received help with their education/GED (OR = 1.335, p = 0.001) and received help to better understand their strengths and talents (OR = 1.306, p = 0.007) were significantly more likely to feel that their long-term education and employment plan is helpful, whereas being taught job skills (OR = 1.092, p = 0.502) did not significantly affect the feeling that their long-term education and employment plan is helpful. After controlling for other individual-level covariates, youth who received help with their education/GED have 1.335 times higher odds of being in higher agreement that their long-term education and employment plan is helpful compared to youth who did not receive help with their education/GED. Additionally, youth who received help to better understand their strengths and talents have 1.306 times higher odds of being in higher agreement that their long-term education and employment plan is helpful compared to youth who did not receive help understanding their strengths and talents. Relatedly, youth who had been told about ways to pay for college (e.g., financial aid packages and loans) were significantly more likely (OR = 1.459, p = 0.000) than youth who did not receive this information to feel as if their long-term education and employment plan is helpful. Specifically, the odds of being in higher agreement that their long-term education and employment plan is helpful are 1.459 times higher for youth who received information on financial aid assistance than those who did not receive this information.

Within the reentry plan domain, the results suggest that youth who were in lower agreement on the reentry plan variables were less likely to be prepared for reentry. Specifically, youth who were less likely to have transportation, able to pay their living expenses, have support for successful reentry, have confidence in their reentry goals, understand the expectations of their reentry, and comply with the expectations of their reentry were significantly less likely to feel that their long-term education and employment plan is helpful. For example, the odds of being in higher agreement that their long-term education and employment plan is helpful is 0.542 (p = 0.000) and 0.518 (p = 0.001) times lower for youth who agree or disagree, respectively, that they have transportation to travel to school and/or work compared to youth who strongly agree they have transportation. Similarly, the odds of being in higher agreement that their long-term education and employment plan is helpful is 0.286 (p = 0.000), 0.073 (p = 0.000), and 0.073 (p = 0.005) times lower for youth who agree, disagree, or strongly disagree, respectively, that they are confident in their reentry goals compared to youth who strongly agree they are confident in their reentry goals.

Results also show that Hispanic youth were significantly more likely to feel that their long-term education and employment plan is helpful. Specifically, Hispanic youth had 1.294 times higher odds (p = 0.038) of being in higher agreement that their long-term education and employment plan is helpful than White youth. More research is needed to understand these racial/ethnic differences. While the effect was not significant, older youth were less likely to feel that their long-term education and employment plan is helpful. Notably, possession of essential documents and living arrangements were not significantly related to feeling more prepared for reentry.

Model 2 shows regression results for reentry preparedness when institutional-level covariates are included in the model. Notably, except for community programming, there were no substantive changes in the statistical significance and effect sizes of the individual-level variables. Without controlling for the institutional-level covariates, the effect of community programming on youth feeling that their long-term education and employment plan is helpful was marginally significant (OR = 1.171, p = 0.062). After controlling for the institutional-level covariates, the effect of community programming reached significance (p = 0.048) and indicates that youth who plan to attend community programming have 1.187 times higher odds of being in higher agreement that their long-term education and employment plan is helpful than youth who will not be attending community programming. Contrary to prior research, none of the institutional-level covariates were significantly associated with reentry preparedness. For instance, despite prior research linking poor juvenile justice education quality with poor academic outcomes, our indicators of the quality of education in the facilities were not predictive of feeling more prepared for reentry.

5. Summary and Discussion

Educational achievement during detention and commitment has been identified as an important turning point for juvenile justice-involved youth in the establishment of careers, a prosocial life, and desistance. However, these gains may be short-lived if youth do not reenroll in school following release, a process that can be especially challenging. Additionally, research has consistently shown that youth are motivated to further their education and careers (see, for example, Gardner et al. 2022; Hartwell et al. 2010; Noorman and Brancale 2022); however, following through with these goals is the challenge. Therefore, understanding the transition process, including the barriers and facilitators, is a needed focus of research. However, to date, much of the research in this area has rarely included the perspective of youth who may have unique experiences and perspectives that can better identify more relevant and meaningful programs and policies to address their needs (Kirshner et al. 2005; Kubek et al. 2020; Powers and Tiffany 2006; Schubert et al. 2014). Research has also not systematically considered whether and how much institutional-level factors, such as the quality of education, staffing, or transition services, influence the likelihood that youth will be prepared to return to school following release.

This study contributes to the literature by bringing renewed attention to the reentry experiences of confined juveniles using a national sample of youth from a unique data source. Specifically, we explored the individual- and institutional-level factors that influenced youths’ self-reported preparedness for education and employment reentry. We found that while most youth agree or strongly agree that they have a plan for their long-term education and employment that is helpful, there are several facilitators and barriers that youth experience that make them feel more or less prepared. Consistent with Siennick and Staff (2008), we found youth who made concrete investments toward achieving their goals—namely, by receiving help with their education/GED, understanding their strengths and talents, and receiving information about ways to pay for college while in a commitment program—were more likely to agree that they had a helpful plan for education and employment reentry.

Our findings also highlight the importance of transition planning. Specifically, we found that youth who were less confident in aspects of their reentry plan—namely, understanding their reentry goals and expectations and having support—were less likely to agree that they had a helpful plan for their long-term education and employment. Consistent with prior research (see, for example, Abrams 2006; Gardner et al. 2022; Sinclair et al. 2021), we found youth who reported likely experiencing challenges with transportation and having enough money to pay for living expenses were less likely to agree they were prepared for reentry. Contrary to prior research, institutional-level factors related to the quality of education in the facilities were not predictive of youths’ reported perception of reentry preparedness.

5.1. Limitations

There are several limitations to this study’s data that are important to mention. First, the dependent variable did not distinguish between education and employment. The survey question asked youth whether they had a plan for their long-term education and employment that was helpful. It is possible that while some of the non-significant factors are important for educational preparedness, they may have no relationship with employment preparedness. There is theoretical and empirical evidence to suggest that education has important implications for employment outcomes; however, we were unable to disentangle this relationship. Another potential limitation lies in the coding of the dependent variable. Specifically, youth showed high levels of preparedness, with most of the sample either strongly agreeing or agreeing that their education and employment plans were helpful and very few disagreeing or strongly disagreeing. The decision to code the variable as ordinal, rather than dichotomize to Strongly Agree/Agree and Strongly Disagree/Disagree was made for both empirical and theoretical reasons. Empirically, it is not recommended to collapse categories that are selected regularly and provide additional information, as doing so may bias results by reducing power and effect sizes (Van Dusen and Nissen 2020; Stromberg 1996). Theoretically, there is an important distinction between analyzing how prepared youth feel to continue their education/employment versus if they are prepared. By coding the dependent variable as an ordinal measure, we were able to assess what factors led youth to feel more prepared for their reentry.

The limited variability across responses could be attributed to when the survey was administered (e.g., close to release) and potential concerns from the youth that they could be confined longer or prohibited from being released if they indicated they did not feel prepared. Further, we were unable to assess what the youths’ reentry plans consisted of and why the youth found them to be helpful. Another limitation is that the data did not include information about youths’ community schools. Specifically, we were unable to account for any community school factors or communication between the juvenile justice facility and the school in which the youth would be reenrolling after their release.

Last, while the current study drew on a larger national sample compared to prior research, there are limitations that may limit the generalizability of the findings. Specifically, the data are over-representative of long-term secure facilities, with 97% of youth exiting a correctional facility. Therefore, the findings may not be representative of youth exiting other types of juvenile justice facilities, such as short-term detention. Given that long-term confinement may provide more opportunities for youth to receive assistance with reentry planning than short-term confinement, it is important for future research to identify potential differences in reentry preparedness for youth returning from different types of juvenile justice facilities.

5.2. Implications for Research and Policy

This study’s limitations notwithstanding, the findings support the need for additional research and policy attention to the reentry experiences of youth confined in residential facilities in the United States. Although the size of the confined juvenile population has been declining in recent years, and the quality of services has been under scrutiny from advocacy organizations and governmental bodies, more focused research on the barriers and facilitators to reentry is warranted. Specifically, attention is needed to characterize the types of transition services that are provided to youth, how these may vary for different populations of youth (e.g., boys and girls, those with disabilities), and mechanisms of communication between juvenile justice schools and community schools (Hirschfield 2014; Richardson et al. 2012; Thomas 2014). Research is also needed to explore the factors that most prepare youth for immediate and long-term success. For example, while vocational training is emphasized and common among adult prison populations, it remains unknown whether juvenile confinement facilities should place the same emphasis on employment. It can be argued that juvenile facilities should prepare youth for educational reentry or taking the GED exam rather than focus on career readiness, as many jobs and careers will be automatically unattainable without certain education credentials.

Additionally, understanding the transition process from the youths’ perspective is important to better identify and address the challenges this population faces in returning to school and the community. This study found that Hispanic youth were more likely than their White peers to feel prepared for release. Future research should explore potential differences in feelings of preparedness among different demographic groups.

The findings from this study highlight the importance of supporting youth while they are in juvenile commitment and ensuring they have a strong transition plan for reentry. Youth who reported receiving assistance with their educational and intrapersonal skills were more likely than those who did not to feel that their post-release educational and employment plan was helpful. Further, youth who reported structural barriers (e.g., not having reliable transportation or inability to pay for their living expenses) were less likely to feel that they had a helpful education and employment plan. These findings suggest that policymakers and practitioners should place an emphasis on ensuring that youth receive quality interaction from staff in juvenile justice facilities that aims to help them develop a meaningful plan for their future.

6. Conclusions

Research and policy attention to the education and employment goals and prospects of detained and committed youth has been largely lacking. Little is known from the perspective of youth—what they believe is helpful or harmful for their post-release success, for example. In this study, we found that most youth believe they have a helpful long-term plan for their education and employment and identified salient facilitators and barriers to their reentry preparedness. However, future research is needed to understand what constitutes the youths’ plan, why it is helpful, and whether youth who report having a helpful plan differentially enroll in school following release and have better or worse outcomes than those who do not report having a helpful plan.