Mentoring for Youth and Young Adults During ReEntry from Confinement
PDF button Facebook button Twitter button

June 2016

This review examines research findings related to mentoring youth and young adults who are reentering their communities after confinement by the justice system. Four areas are addressed for this population: (1) mentoring effectiveness; (2) the moderation of mentoring effectiveness, or the extent to which effectiveness is related to other variables, such as the characteristics of mentors, mentees, or program practices; (3) the mediation of mentoring effectiveness, or the extent to which intervening processes link mentoring to youth outcomes; and (4) the reach, implementation, adoption, and sustainability of programs and other supports for mentoring of reentering youth and young adults.

Unfortunately, few studies have examined any of these questions. While meta-analyses of the findings from evaluations of mentoring programs suggest that mentoring youth at risk for problems such as involvement in the justice system is a promising practice, few comparison group studies were found on outcomes related to mentoring at reentry. Specifically, only one published study and four unpublished studies were found of programs providing mentoring to youth during reentry, and only one published study and no unpublished studies were found on programs providing mentoring to young adults. Mentoring was typically one component in a more comprehensive treatment array that usually included case management and tailored access to needed services. Most mentors were paid, and most had other roles besides mentoring. Significant differences were found between comparison groups in two of the six studies and trends toward significant group differences were found in two additional studies; the differences observed in these four studies were consistent with positive effects of mentoring or the larger intervention of which mentoring was one component. Research addressing the remaining questions was extremely limited and thus did not provide a sufficient basis for conclusions. Insights and recommendations for practice are made based on the available information.

  • Introduction

    There are a variety of lengths for and settings of confinement that a youth might experience when involved with the juvenile justice system, ranging from spending a few hours at a neighborhood police station, to a few days at the local county youth services facility, to a few months in a group home, to many months at a state youth correctional facility. Ideally, each of these consequences is intended to be a corrective experience, to lead to improved outcomes instead of deteriorating ones. However, Snyder1 argued that each such experience has the potential to negatively impact the ability of a youth to succeed in a prosocial manner in his or her community in the future. For example, even during a short stay, a youth might connect with members of a gang, and this could lead to involvement in the gang and the commission of a new set of antisocial behaviors once released. Thus, experiencing any type of confinement might warrant the provision of “reentry” services before and after a return to home that are designed to increase both short- and long-term positive outcomes.

    Unfortunately, reentry is often not a situation that occurs only once. Many youth have repeated episodes of confinement (e.g., 50%2) and thus experience repeated periods of reentry. This occurs despite a large amount of resources being spent on juvenile justice, and particularly toward the highest risk youth. For example, Broder3 estimated that even though California’s juvenile prison system costs the state about $6 billion annually, or $80,000 per youth, 90% of youth recidivated. Given this backdrop, it is not surprising that improving the ability of the juvenile justice system to prevent recidivism among youth who are reentering their communities following periods of confinement is of great interest to funders, program developers, practitioners, and policymakers alike.4, 5

    Interest in reentry has led to a wide range of responses. Two recent initiatives of the Office of Juvenile Justice and Delinquency Prevention (OJJDP) focused on the reentry population are the Juvenile Mentoring Initiative and Strengthening Relationships Between Young Fathers, Young Mothers, and Their Children, both funded under the Second Chance Act. Through these funding streams, OJJDP has provided grants to support mentoring and other transitional services considered essential to reintegrating juvenile offenders into their communities. Attempts also have been made to improve the dissemination of research findings into the practice community. For example, the National Reentry Resource Center was established in 2009 by the U.S. Department of Justice to serve as the primary source of information and guidance in reentry, advancing the use of evidence based practices and policies, and creating a network of practitioners, researchers, and policymakers invested in reducing recidivism. reentry repeatedly, ongoing community-based interventions such as mentoring seem warranted for consideration as a reentry service.

    In this review, research relevant to understanding the potential influence of mentoring programs on youth during reentry from juvenile justice (see Blechman & Bopp6; Chan & Henry7) will be analyzed. Nationally, a one-day census conducted in 2013 (October 23) found that 54,148 juveniles were held in residential placement facilities for offenses; this number has been steadily declining over time, from counts of over 100,000 youth in the late 1990s and 2000.8 Youth in custody are most often male, in their mid-teen years, and those who have been convicted of nonviolent offenses; rates of confinement, furthermore, are notably higher for racial and ethnic minority youth, particularly those who are Black or African-American.8 Most are dealing with a number of significant struggles in their lives, such as poverty, family problems, and difficulties with school, as well as other key challenges, such as substance abuse and mental illness. Although national data on average length of time in custody are not available, most juveniles will be detained (i.e., held in a facility pending an adjudicatory or disposition hearing) for relatively brief periods of time, whereas those who are committed (i.e., placed into a residential facility after adjudication) will experience longer periods of confinement (in the above referenced one-day census, half of the detained juveniles had been in placement fewer than 22 days, whereas half of the committed residents had been in placement longer than 120 days). Regardless, however, the overwhelming majority (over 80% according to one estimate1) will be released within a year of confinement. Thus, most youth who are confined due to juvenile justice involvement do return to their communities, and since many experience reentry repeatedly, ongoing community-based interventions that provide support, such as mentoring, seem warranted for consideration as part of a reentry service package.9 Since most states allow some individuals to remain under the jurisdiction of the juvenile court after the age of 18 years, and since young adulthood is widely viewed today as an “emerging” state10 rather than a qualitatively different state of being from adolescence, we also consider findings on mentoring for young adults (under the age of 25 years) involved with the justice system. This approach is consistent with recently published rigorous reviews of the reentry literature for youth (e.g., James, et al.11).


      Findings from relevant research studies were sought to answer questions in four key areas about mentoring and reentry: program effectiveness, moderators of program effectiveness, mediators of program effectiveness, and program implementation.

      1. (Effectiveness) What is the demonstrated effectiveness of mentoring for youth who are reentering the community following detention or incarceration within the juvenile justice system (“reentering youth”)?

      2. (Moderators) To what extent do the benefits of mentoring for reentering youth appear likely to be contingent on such considerations as the different characteristics of the youth involved and/or their mentors, the type of outcome(s) being considered, the programmatic practices or approaches that are employed?

      3. (Mediators) What intervening pathways or variables appear likely to be most important in linking mentoring to outcomes for reentering youth?

      4. (Reach, Implementation, and Adoption/Sustainability) What factors appear likely to be effective for reaching and engaging reentering youth in mentoring services, ensuring high-quality implementation of such services, and facilitating their adoption and institutionalization within settings and organizations that serve this population?


      An extensive search was undertaken for published and unpublished studies of mentoring for youth and young adults undergoing reentry. A wide variety of references of relevance to youth mentoring and juvenile justice were found, including individual studies, narrative reviews, and meta-analyses. Unfortunately, few of these focused both on a program that included a mentoring component and participants who were currently experiencing reentry from an extended period of confinement (e.g., six months to one year), which is a typical definition of the word “reentry” in practice. In the end, we included all peer reviewed publications from the past three decades and commonly cited unpublished reports since 1990 of studies that (a) included a sample of youth or young adults who were or had recently been in confinement for any period of time and in any setting at the first assessment point and (b) focused on an intervention that included mentoring. These decisions yielded seven unique research studies, six of which included a comparison group (see Table 1). Unfortunately, most of these studies were unpublished reports, and some were quite difficult to decipher in terms of specific project details of import. A few additional studies were found on youth who were on probation, but those are not included here.

  • What Is the Effectiveness of Mentoring at Reentry for Youth and Young Adults?


      In a variety of meta-analytic and narrative reviews, mentoring has been highlighted as a promising practice for youth. Effects appear to be modest, but stronger for youth living in environments that place them at elevated risk for the commission of problem behaviors, including antisocial and delinquent behaviors, which may lead to involvement with the juvenile justice system. Thus, mentoring seems like a potential intervention for use with youth during reentry. The relatively stronger effects of mentoring on youth in “high risk” circumstances was a primary conclusion in the DuBois, Holloway, Valentine, and Cooper19 meta-analysis of 55 evaluations of the effects of mentoring programs on youth, and again in the DuBois, Portillo, Rhodes, Silverthorn, and Valentine20 meta-analysis of 73 such evaluations. More recently, Tolan et al.21 conducted a meta-analysis focused specifically on mentoring studies of youth at risk for, or already involved in, delinquency and found modest effect sizes on all outcomes examined, including delinquency and aggression. Their findings echoed those of Jolliffe and Farrington22 based on their meta-analysis of effects of mentoring programs on reoffending for youth (see also Lipsey23). Given these reports, mentoring youth during reentry from confinement seems like a strategy worthy of consideration.

      Aftercare programs for youth and young adults who have been confined for offenses have also demonstrated generally small effects on recidivism.11 Interestingly, effects have been more substantial for programs that consist of individual instead of group intervention, with one type of individual intervention being one-on-one mentoring (e.g., Drake & Barnoski13). Work with older populations of offenders has also found that mentoring from community volunteers or clergy appears promising in terms of reducing recidivism.24 Findings such as these offer further reasons for optimism for the potential benefits of mentoring for youth and young adults reentering their communities following periods of confinement.


      As noted previously, seven independent studies were identified in this review (see Table 1). Participants in these studies had been in confinement or a community residential program for varying amounts of time, from three weeks to over one year. Across the studies, and similar to the findings in Snyder1 of the characteristics of the reentry population at large, participants tended to be male and racial and/or ethnic minority. The six studies of juvenile populations resembled Snyder1 in that the average age of each sample was in the mid-teens. Six of the studies examined mentoring within the context of multimodal intervention, which typically included case management and the receipt of other needed services. Two of the studies used volunteer mentors, and the rest used paid mentors, most of whom delivered a variety of other intervention components as well (e.g., case managers).

      As can be seen in Table 1, only two of the six comparison group studies found statistically significant differences between the groups on delinquency related outcomes, such as recidivism. In two additional studies, trends in favor of the mentored group were reported; although not statistically significant, it is notable that these differences were evident at follow-ups extending to three or more years. Unfortunately, none of these studies examined other key life outcomes besides delinquency, such as homelessness, completion or continued involvement in education, or employment status. In all, this collection of findings, most of which are unpublished, reflect the “modest” effects of mentoring found in each of the meta-analyses noted above. They also reflect the “substantial” heterogeneity of effects reported by Tolan and colleagues21 in their meta-analysis of the effects of mentoring programs on delinquency and related outcomes. In short, knowledge about the impact of mentoring for youth at reentry is at present extremely sparse despite repeated calls for more rigorous research (e.g., Abrams, Mizel, Nguyen, Shlonsky25).


      1. Insufficient evidence is available to draw conclusions about the effectiveness of mentoring, whether delivered in tandem with other services and supports or not, for youth or young adults during reentry.

  • What Factors Condition or Influence the Effectiveness of Mentoring at Reentry for Youth and Young Adults?


      A “moderator” is a variable that affects the relation between a mentoring program and a particular outcome (e.g., Baron & Kenny26). For example, if both boys and girls were included in a sample of juvenile-justice-involved youth who were randomly assigned to either receive a mentoring program at reentry or not, and the program was found to have an impact on the delinquent behavior of boys but not girls, then sex would be considered a moderator of the effect of the program. While a wide variety of both continuous and categorical variables might be moderators of program effects, even the description of potential moderators is scant in the mentoring literature related to youth delinquency.21

      In this larger body of literature, which subsumes the topic for this review, Tolan et al.21 found enough information to examine the influence of only five moderators: type of sample selection criteria (i.e., individual risk, environmental risk), type of program (i.e., mentoring only, multimodal), motivation to be a mentor (i.e., civic duty, professional development), existence of program quality monitoring, and program emphasis (i.e., modeling/identification, emotional support, teaching, advocacy). Some evidence of moderation was found for three categorical variables: (1) whether or not professional development was a motive for becoming a mentor, (2) whether or not providing emotional support was a program emphasis, and (3) whether or not engaging in advocacy activities for a youth was a program emphasis. Separately, the presence of each of these factors was related to greater program impact. Unfortunately, at the time of the Tolan et al.21 work, there was insufficient information available in the literature to examine interactions between and among these and other potential moderating variables. This remains true today.

      Research on possible moderators of the effects of reentry programs and supports is similarly limited. The broader review of this research referenced previously found that effects were more favorable among older and high-risk youth.11 In the multisite evaluation of the mentoring component of the Ready4Work reentry initiative,27 which included both young and older adults, qualitative interviews were conducted with 31 mentors, eight of whom had been incarcerated. Findings suggested that mentors who had been incarcerated may be in a better position to support their mentees. Specifically, mentors who had never been incarcerated mentioned more frequently that they struggled with getting their mentees to open up and be responsive to offers of help than did mentors who themselves had at some point been incarcerated.

      Different program models further illustrate innovative program practices that might be important in conditioning effective mentoring and associated program supports for youth and young adults reentering their communities. For example, in the Boys & Girls Clubs of America’s Targeted Reentry program (cited in James et al.11), while incarcerated, the youth develops, together with a facility treatment team, an individualized transition plan. As part of this process, the transition specialist assists the youth in establishing a Community Action Team of mentors and local service providers that provide counseling, support, guidance, and assistance in meeting the goals of the transition plan. The transition specialist then works with each youth and his or her Community Action Team to identify and secure resources and services. The approach of this program has much in common with the strategy of youth-initiated mentoring, which has promising evidence of effectiveness for older adolescents (a National Mentoring Resource Center review of the research on this practice can be found here).


      No studies were found on the moderation of outcomes for mentoring programs for youth or young adults during reentry. A few other mentoring studies of samples that included youth involved in juvenile justice have examined the moderation of effects. For example, Seroczynski, Evans, Jobst, Horvath, and Carozza28 reported findings suggestive of greater effects of participation in the Reading for Life program (a diversion program) on rearrest among males, older youth (16 years and older), minority youth, and youth from lower income families. None of these trends, however, reached statistical significance. Moderation is often examined after outcome effects are found for the overall sample, but several of the comparison group studies in this review failed to find such effects. Further, the examination of moderation requires adequate sample sizes so that there is sufficient statistical power to detect such effects. Many of the study samples in this review were quite small.


      1. Insufficient evidence is available to draw conclusions about the moderators of the effects of mentoring provided to youth during reentry from juvenile justice.

  • What Intervening Processes Are Most Important in Linking Mentoring at Reentry to Outcomes for Youth and Young Adults?


      A “mediator” is a variable that accounts for the relationship between a mentoring program and an outcome of interest.26 In mentoring programs, one type of mediator of interest is typically some characteristic or aspect of the relationship between the mentor and the youth, such as “relationship quality.” More generally, it is through processes that occur within this relationship, or that are initiated by this relationship, that mentoring has been thought to impact outcomes such as youth delinquency.29 Unfortunately, examinations of the mediation of effects in the mentoring literature at large are rare, and almost nonexistent in the literature on juvenile delinquency (see discussion in Tolan et al.21).

      Several studies have hypothesized about factors that might be related to mediation, or a combination of mediation and moderation (i.e., factors that may serve as intervening processes for those with particular characteristics or backgrounds). For example, Moore30 hypothesized that mentor qualities impacted outcomes in his study of a counseling program using “citizen volunteers” with youth on probation:

      “Severe demands were made on both the time and the talents of the unpaid volunteers, who functioned as crisis counselors, resource brokers, tutors, advocates, and friends, and even arranged employment. In the present study the volunteers were exceptionally capable, mature, energetic, and dedicated people who had counseled several young offenders successfully in the past. Volunteers who are less talented, less dedicated, and less experienced may be less successful.” (p. 828)

      Possibly the background, experience, and commitment of these mentors (moderating variables) led to higher quality relationships (a mediating variable), as well as meaningful outcomes, such as increasing the social capital of mentees and helping problem solve key issues along the way.

      In a notable examination of mediation within the broader mentoring and reentry literature, the evaluation of the Ready4Work initiative referred to previously27 found some support for two potential pathways through which mentoring during reentry was related to improved employment and recidivism outcomes for participants: through program retention and through participation in program services (i.e., wraparound and job placement/retention services). Research from youth mentoring suggests that such pathways may be most viable for younger reentry populations when there is ample programmatic support for mentoring (e.g., training) and it is integrated with other services rather than simply as an “add on.”31


      No studies were found that examined mediators of the effects of mentoring programs for youth or young adults during reentry. One study, however, did find that when a mentor-mentee match was related to a “high intensity” or “closer” mentoring relationship, better outcomes were found12 (see also Gur & Miller32). In this research by Jarjoura, among those matches in which “mentor and youth worked closely together,” only 13% of youth were reincarcerated by one year post-release and only 28% by four years post-release. This is in contrast to corresponding rates of 39% and 62%, respectively, in a services-as-usual control condition. Interpretation of these types of results is complicated by a lack of clear definitions of what quality relationships look like during reentry as well as by the potential for youth identified as having strong relationships with their mentors to have differed in other ways that were consequential prior to mentoring (e.g., stronger social skills).


      1. Insufficient evidence is available to draw conclusions about mediators of outcomes related to mentoring youth and young adults at reentry.

  • Have Mentoring Supports and Services at Reentry Following Confinement Been Implemented with High Quality and Sustained?


      Guidance around program implementation is vital for the effective delivery of mentoring programs. As noted by Tolan et al.,21 one explanation for the significant variation in effect sizes for mentoring programs is that there are better ways to provide mentoring services than others. Additional key concerns include engaging targeted populations of youth in programs as well as the extent to which available programs and services are likely to be adopted and implemented on a sustained basis by relevant host settings within and outside of the juvenile justice system.

      In a few instances, researchers have commented on implementation issues. For example, in a study of youth on probation, Lane, Turner, Fain, and Sehgal33 noted the conflicts that can occur for staff and mentors when working within a justice system. The family and community-centered focus of their program was at times at odds with the orientation and practices of the juvenile justice system. This might have led to a less than optimal implementation of a program, decreased likelihood of program success, and increased likelihood that staff and mentors left their program. Given these possibilities, it may be vital that program administrators and their staff as well as justice system administrators and their staff work closely together, reach an understanding of what is possible and what is not within an existing system, and develop policies, practices, and support systems that enable what is possible to actually occur.

      A key part of such a set of policies, practices, and support could potentially revolve around good record keeping so that implementation can be monitored by practitioners and administrators and documented by researchers.21 In their studies of mentoring, both Lane et al.33 and Bouffard and Bergseth14 noted that record keeping was a problem. Reliable and valid administrative records that document not only youth demographics and background characteristics, but mentoring program and juvenile justice system processes, may be just as important as records that document outcomes, such as recidivism.

      Another area of potential importance highlighted by researchers relates to the fundamental nature of mentoring programs. For example, Blechman, Maurice, Buecker, and Helberg34 compared a skills training program with a mentor program on recidivism outcomes for youth (see also Davidson, Redner, Blakely, Mitchell, and Emshoff35). These researchers suggested that logistical problems for mentor programs present unique challenges that contribute to heterogeneity of program effects. For example, in skills training programs, instructor qualifications and class content are usually specified and formalized, and what it means to “implement” a program is clear to all, including juvenile justice staff members. This may not be the case with a mentoring program. Not only are mentor specifications often lacking, but issues such as recruiting mentors who are mutually acceptable matches with mentees and who endure throughout the course of the program can be difficult. These types of issues could pose serious problems for implementation and for mentor program effectiveness.30, 34

      Yet it’s worth trying to overcome these challenges. In the aforementioned review of aftercare programs11 effects on recidivism were stronger when programs were well-implemented. A key consideration for mentoring programs in this regard may be assessing the extent to which a program, as actually delivered, aligns with both the developmental and intervention models underlying the program (see Eddy, Dishion, and Stoolmiller36). This would include not only a careful examination of how mentoring was delivered, but the training, supports, and supervision that mentors receive (versus what they are supposed to receive).


      No rigorous research studies were found on any of the above concerns with respect to mentoring programs and services for youth or young adults during reentry. Although the Ready4Work initiative involved a broader age range of adults (slightly less than half of the participants were between ages 18 and 25 years), several of its findings are nonetheless noteworthy. On a positive note, sites were generally successful in recruiting enough mentors. Congregations were a major source of mentors, with 54% being recruited from this source; furthermore, although recruitment efforts did not deliberately target formerly incarcerated individuals, almost one third of the mentors had spent time incarcerated. On the other hand, the sites were successful in engaging only about half of those enrolled in the initiative in the mentoring component. Among the barriers to engagement identified that seem most potentially relevant to youth and young adults were competing demands on time, a perception that mentoring was just one more form of institutional requirement, and a reluctance to discuss problems with strangers. Sites also experienced ongoing struggles to keep participants engaged in the mentoring component, such that only about a third of participants received more than three months of mentoring. Of additional note, younger participants were significantly less likely to become involved in the mentoring component of the program.27 These findings notwithstanding, an in-depth and longitudinal qualitative examination of resilience among adolescents who were formerly incarcerated revealed evidence of a general openness to adult mentoring.37 Furthermore, whereas results highlighted the apparent benefits of relationships with caring and committed adults that youth experienced while confined, this contrasted starkly with a lack of adults available to youth post-confinement to provide support with skill development in areas of critical importance to their overall well-being and adaptation in community settings (e.g., coping, problem-solving, substance abuse).


      1. Insufficient evidence is available to draw conclusions about any aspect of implementing mentoring programs and supports for youth or young adults during reentry following confinement; this is also true for related considerations of successfully engaging targeted youth in such programs or support and their adoption and sustained utilization by host settings.

  • Implications for Practice

    (Mike Garringer, MENTOR: The National Mentoring Partnership)

    There is a long history in the United States of charitable and religious organizations doing work not only behind prison walls, but also providing various forms of support to adults upon their release from correctional facilities. This work correctly recognizes that incarceration and removal from a community is a deeply unsettling experience and that individuals will need many forms of support as they integrate back into communities and begin the hard work of staying on the right side of the law. Increasingly, these types of reentry services are also being offered to juveniles who are coming out of incarceration or other, less disruptive, periods of engagement with the juvenile justice system⎯offered by community organizations or government agencies themselves in an attempt to curb recidivism and future adult system involvement by juvenile offenders. As noted in this review above, mentoring, either by an individual or in a group setting, is one of the more prominent strategies employed by these programs.

    Serving this population with mentors, whether as a stand-alone service or in conjunction with other services, can be a challenging proposition for a variety of reasons: the multifaceted needs of the youth; the skills needed for mentors to build a strong mutual relationship at this critical moment in time; the communication between, and coordination of, services with the many players in the “system.” But as researcher Roger Jarjoura has noted, “mentoring may be the missing ingredient”38 in helping these youth break old habits, find stable footing, and start rebuilding their lives39. So what are some keys to success for serving reentering juvenile offenders well?


      The review above offers a few potential keys to success in its discussion of moderators of program impact (Question 2). There were three factors that were associated with stronger outcomes for programs in the studies examined here:

      • Mentors who were using the mentoring experience for professional development at some level. This finding implies a few things: Much of the mentoring offered by programs serving reentering youth is delivered by undergraduate or graduate students or similar young professionals looking for practical field experience. Many programs utilize students who are pursuing careers in law enforcement, juvenile justice, psychology, social work, or other helping professions. These mentors not only bring considerable skill from their classroom learning and other experience, but are also likely to have a support structure of faculty or other clinicians who can effectively solve problems in their work with mentees and help them hone their abilities to provide appropriate support. It remains to be seen whether average citizens can step into this mentoring role as well as someone who is using mentoring to further, or prepare for, a related career. That combination of skill and motivation would be hard for any layperson to beat, regardless of the level of training provided by the program.

        However, there may be a downside to using students or professionals-in-training in this role: limitations on their time and availability. Most of these students would be unlikely to be in a position to continue their mentoring relationship beyond a school year and may have restrictions on the types of support they can provide (and when they can provide it) as a way of teaching how to set boundaries and manage liability in their future careers. One hopes that nine months of mentoring by a highly skilled graduate student will be enough to get a youth back on track after reentry or juvenile justice involvement, but perhaps it’s not for some youth. There is a general sentiment in the mentoring field that mentoring relationships grow stronger and deeper with time. But that’s time that is inherently not available to matches that use a student or professional-in-training in this role. They may find themselves having to end their match at a critical point in the mentee’s transition. However, difficulties caused by this limited duration might be offset by clear expectation setting up front, meaningful match closure procedures, or even policies that allow the youth to continue with a new student mentor.

      • Mentors providing emotional support, and mentors providing advocacy on behalf of their mentee. These two findings are interesting together, as they speak to the need to address both the social-emotional (soft skills) and instrumental (concrete help) aspects of mentoring relationships for reentry youth. Mentoring programs serving these youth may feel considerable pressure, and rightly so, about providing tangible problem solving and networking support to these youth. But the emotional support reentering mentees receive is likely just as critical to their long-term success. There are likely to be many ups and downs as a young person attempts to get their life back on track and the emotional support of a mentor might be essential to helping a mentee stick with their life plan. The need to provide both forms of support hints that programs must give mentors meaningful time to get to know and bond with the youth they will be supporting in this journey. It may be wise to begin these relationships while the youth is still incarcerated (or heavily involved in probation or diversion) so that the mentor is already viewed as a stable and trusted presence at the time of release. The close emotional bond that mentors can form not only paves the way for the more instrumental forms of support (e.g., help finding employment) that follow, but can also help the mentee grow cognitively and support identity development (reinventing oneself) upon release. (In fact, one prominent reentry program emphasizes this cognitive development in the work of mentors, using them to help change the way these youth think about themselves and their decision making.) Programs should not take these close emotional bonds for granted in serving these youth⎯a program mentor may offer a chance for a genuine relationship that is unavailable from another person in their community or those in the juvenile justice system itself.


      As noted earlier, during reentry, youth are likely to need help in many areas of life, many beyond what a typical mentoring program, or even a multicomponent youth development organization, can supply. Providing referrals and assisting in the access to other services is a given for serving this population well.

      Mentors will need to be able to make recommendations or provide support around life areas such as:

      • Employment, career exploration, job training
      • Housing assistance
      • Remedial education and credit recovery
      • Postsecondary education exploration and application
      • Financial decisions and planning
      • Drug and alcohol treatments
      • Mental health services
      • Goal-setting and planning for the future

      Since no one organization, or one mentor, can address all of these areas perfectly, programs should establish referral agreements with relevant service providers in the community and train mentors on how and when to access these services. A mentor should not have to feel like they need to do it all for these mentees, but they do need to know where their role begins and ends in relation to other services and how other providers compliment their efforts.


      Mentoring programs should do careful planning around how they will mitigate challenges in serving reentry youth, such as:

      • Interacting effectively with the juvenile justice system and the many other professionals who are working with the mentee. Programs will need to determine how coordinated their services will be with other support and services that are being provided through juvenile justice services, parole and probation, court-appointed advocates, and others. This was noted as a common implementation challenge in the review, but there may be opportunities for information sharing and coordinated support provided the program and the mentor can establish relationships with these other professionals and service providers and get written agreements in place that allow their organizations to partner effectively.

      • Tracking mentees. Unfortunately, life after release is often chaotic for reentry youth⎯ they may be transitioning into different living arrangements, new jobs, or a fresh start on their education, or they may just be busy trying to get things “back on track” in many other areas of their life. This can make finding time and space to meet with mentees a challenge for mentors. Programs may want to offer more frequent check-ins than is recommended in the Elements of Effective Practice for MentoringTM40 and be prepared to offer additional support to mentors in finding and scheduling meetings with their mentees (this is another area where established lines of communication with parole and probation employees can really help).

      • Establishing appropriate boundaries for the relationship. As noted above, these youth will likely need help in many areas of life and may be facing some very challenging circumstances around their return to their community. It will be critical for mentors to understand the boundaries and limitations of their role. Programs will need clearly communicated policies around transportation, limits on forms of support (financial, for example), confidentiality and information sharing, and when to hand off some aspect of mentee support to another organization or service provider. As noted previously, this may look quite different for programs that are using students or professionals-in-training versus programs that are using community or faith-based volunteers. Additionally, as with all mentoring relationships, mentors will also need to make sure they don’t take a prescriptive approach here and impose their own solutions on the mentee, but rather, work with the mentee to find and pursue their own solutions to their reentry challenges. An overall philosophy of empowerment and selfdetermination is a good fit for programs and mentors serving these youth.

      • Facilitating access to other community supports. The effectiveness of a mentoring program serving reentry youth may be somewhat dependent on the availability of other services and support in a community. Direct help with “big-picture” needs, such housing, employment, and substance abuse treatment, will likely come from outside of the mentoring relationship or program. But these services may be missing or already overwhelmed in some communities. Programs serving reentry youth should map the network of providers in their area and understand how and when mentees can take advantage of those other services (as well as working across organizations to fill gaps).

      • Defining success. This last challenge may be a bit philosophical, but it’s one that’s important to consider in terms of program evaluation and funder relations. Obviously, no mentoring program will keep every reentry youth from future juvenile justice involvement, but setting truly realistic outcome targets may be challenging given the lack of rigorously studied program models noted in this review. Programs should make sure that their theory of change, and any associated benchmarks or indicators, are carefully considered. Asking key questions can help define success: Is a certain reduction in recidivism a stated goal of the program or are other goals emphasized? Is that reduction bound by time (e.g., making it a year without reoffending) or status (what if they re-offend as an adult down the road?). Are there goals tied to reductions in the severity of offenses? Are there shorter-term precursors to reduced recidivism that need to be measured? How can the program assess progress with a young person in the short term?

        As one example of defining success noted in this review, an evaluation of the Aftercare for Indiana through Mentoring program (AIM) has reported a reduction in the recidivism rate for youth who participate fully in the program to 13% after one year and 28% after four years.13 Assuming these findings are reliable, it still may be the case that these impressive results are not achievable by all, or even most, programs. Programs should put careful thought into how they, and their funders, co-define success, recognizing that it takes a number of services and support working effectively together for most youth to fully turn their lives around during the reentry process.

  • References

    1.Snyder, H. N. (2004). An empirical portrait of the youth reentry population. Youth Violence and Juvenile Justice, 2, 39-55.

    2. Dembo, R., Turner, G., Chin, C. S., Schmeidler, J., Borden, P., & Manning, D. (1995). Predictors of recidivism to a juvenile assessment center. The International Journal of the Addictions, 30, 1425–1452.

    3. Broder, J. M. (2004, February 15). Dismal California prisons hold juvenile offenders. New York Times. Retrieved from

    4. Gies, S. V. (2003). Aftercare services. Washington, D.C.: US Department of Justice, Office of Justice Programs, OJJDP. Available at

    5. Eddy, J. M., & Swanson Gribskov, L. (1998). Juvenile justice and delinquency prevention in the United States: The influence of theories and traditions on policies and practices. In T. P. Gullota, G. R. Adams, & R. Montemayor (Eds.), Delinquent violent youth (pp. 12–52). Thousand Oaks, CA: SAGE.

    6. Blechman, E. A., & Bopp, J. (2005). Juvenile offenders. In D. L. Dubois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 454–466). Thousand Oaks, CA: SAGE.

    7. Chan, W. Y., & Henry, D. B. (2014). Juvenile offenders. In D. L. Dubois & M. J. Karcher (Eds.), Handbook of Youth Mentoring (2nd ed., pp. 315–324). Thousand Oaks, CA: SAGE.

    8. OJJDP Statistical Briefing Book. (2015, April 27). Online. Retrieved from

    9. Anthony, E. K., Samples, M. D., de Kervor, D. N., Ituarte, S., Lee, C., & Austin, M. J. (2010). Coming back home: The reintegration of formerly incarcerated youth with service implications. Children and Youth Services Review, 32, 1271–1277.

    10. Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55, 469–480.

    11. James, C., Stams, G. J., Asscher, J. J., De Roo, A. K., & van der Laan, P. H. (2013). Aftercare programs for reducing recidivism among juvenile and young adult offenders: A meta-analytic review. Clinical Psychology Review, 33, 263–274.

    12. Jarjoura, G. R. (2004). AIM — Indiana’s juvenile reentry program, aftercare for Indiana through mentoring. Indianapolis, IN: AIM.

    13. Drake, E. K., & Barnoski, R. (2006). Recidivism findings for the Juvenile Rehabilitation Administration’s mentoring program: Final report. Olympia, WA: Washington State Institute for Public Policy.

    14. Bouffard, J. A., & Bergseth, K. J. (2008). The impact of reentry services on juvenile offenders’ recidivism. Youth Violence and Juvenile Justice, 6, 295–318.

    15. Braga, A. A., Piehl, A. M., & Hureau, D. (2009). Controlling violent offenders released to the community: An evaluation of the Boston Reentry Initiative. Journal of Research in Crime and Delinquency, 46, 411–436.

    16. Gray, E., Taylor, E., Roberts, C., Merrington, S., Fernandez, R., & Moore, R. (2005). Intensive Supervision and Surveillance Programme: Final report. Oxford, UK: University of Oxford.

    17. Greenwood, P. W., Deschenes, E. P., & Adams, J. (1993). Chronic juvenile offenders: Final results from the Skillman Aftercare Experiment. Santa Monica, CA: The RAND Corporation.

    18. Bergseth, K. J., & McDonald, T. D. (2007). Reentry services: An evaluation of a pilot project in Clay County, MN. Fargo, ND: North Dakota State University.

    19. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A metaanalytic review. American Journal of Community Psychology, 30, 157–197.

    20. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorne, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91.

    21. Tolan, P. H., Henry, D. B., Schoeny, M.S., Lovegrove, P., & Nichols, E. (2014). Mentoring programs to affect delinquency and associated outcomes of youth at risk: A comprehensive meta-analytic review. Journal of Experimental Criminology, 10, 179–206.

    22. Jolliffe, D., & Farrington, D. P. (2007). A rapid evidence assessment of the impact of mentoring on re-offending: A summary. London, UK: Home Office.

    23. Lipsey, M. W. (1992). Juvenile delinquency treatment: A meta-analytic inquiry into the variability of effects. In T. D. Cook, H. Cooper et al. (Eds.), Meta-analysis for explanation: A casebook (pp. 83–127). New York: Russell Sage Foundation.

    24. Duwe, G., & Johnson, B. R. (2015). The effects of prison visits from community volunteers on offender recidivism. The Prison Journal, 96, 279–303.

    25. Abrams, L. S., Mizel, M. L., Nguyen, V., & Shlonsky, A. (2014). Juvenile reentry and aftercare interventions: Is mentoring a promising direction. Journal of Evidence-based Social Work: Advances in Practice, Programming, Research, and Policy, 11, 404–415.

    26. Baron, R. M., & Kenny, D.A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182.

    27. Bauldry, S., Korom-Djakovic, D., McClanahan, W. S., McMaken, J., & Kotloff, L. J. (2009). Mentoring formerly incarcerated adults: Insights from the Ready4Work reentry initiative. Philadelphia, PA: Public/Private Ventures.

    28. Seroczynski, A.D., Evans, W. N., Jobst, A. D., Horvath, L., & Carozza, G. (2016). Reading for Life and adolescent re-arrest: Evaluating a unique juvenile diversion program. Journal of Policy Analysis and Management, 35, 662–682.

    29. Rhodes, J. E. (2005). A theoretical model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30–43).Thousand Oakes, CA: SAGE.

    30. Moore, R.H. (1987). Effectiveness of citizen volunteers functioning as counselors for high-risk young male offenders. Psychological Reports, 61, 823–830.

    31. Kuperminc, G.P., Thomason, J., DiMeo, M., & Broomfield-Massey, K. (2011). Cool Girls, Inc.: Promoting the positive development of urban preadolescent and early adolescent girls. The Journal of Primary Prevention, 32, 171–183.

    32. Gur, M., & Miller, L. (2004). Mentoring improves acceptance of a community intervention for court-referred male persons in need of supervision (PINS). Child and Adolescent Social Work Journal, 21, 573–591.

    33. Lane, J., Turner, S., Fain, T., & Sehgal, A. (2005). Evaluating an experimental intensive juvenile probation program: Supervision and official outcomes. Crime & Delinquency, 51, 26–52.

    34. Blechman, E. A., Maurice, A., Buecker, B., & Helberg, C. (2000). Can mentoring or skill training reduce recidivism? Observational study with propensity analysis. Prevention Science, 1, 139–155.

    35. Davidson, W. S., Redner, R., Blakely, C. H., Mitchell, C. M., & Emshoff, J. G. (1987). Diversion of juvenile offenders: An experimental comparison. Journal of Consulting and Clinical Psychology, 55, 68–75.

    36. Eddy, J. M., Dishion, T. J., & Stoolmiller, M. (1998). The analysis of intervention change in children and families: Methodological and conceptual issues embedded in intervention studies. Journal of Abnormal Child Psychology, 26, 53–69.

    37. Todis, B., Bullis, M., Waintrup, M., Schultz, R., & D’Ambrosio, R. (2001). Overcoming the odds: Qualitative examination of resilience among formerly incarcerated adolescents. Exceptional Children, 68, 119-139.

    38. Summer Institute on Youth Mentoring. (2011). It may be the missing piece: Exploring the mentoring of youth in systems of care. Reflections from the 2011 Summer Institute on Youth Mentoring.

    39. MENTOR: The National Mentoring Partnership. (2014). Restorative relationships: Elements of Effective Mentoring for Adjudicated Youth (Part of the Collaborative Mentoring Series) [Video]. Retrieved from

    40. Garringer, M., Kupersmidt, J., Rhodes, J., Stelter, R., & Tai, T. (2015). Elements of effective practice for mentoring (4th ed.). Boston: MENTOR: The National Mentoring Partnership. Retrieved from

  • Tables

You can also find several tools and activity guides that can support mentoring in the Resources section of the NMRC website. And remember that you can always request NMRC technical assistance to help start or improve a mentoring program.

Request no-cost help for your program

Advanced Search