Mentoring Model/Population Reviews

Mentoring Model/Population Reviews (18)

Mentoring for EABB
PDF button Facebook button Twitter button

March 2021

This review examines research that addresses the potential influence of mentoring for youth on their educational attitudes, beliefs, and behaviors EABBs). In general, from experimental and meta-analytic studies, the effects of mentoring on educational attitudes and beliefs were small and inconsistent across studies and forms of mentoring (e.g. naturally occurring vs. program sponsored). Yet, there is evidence that mentoring has the potential to influence a range of EABBs, including self-esteem, school connectedness, school engagement, and attitudes toward school. The cumulative literature provides some insight into how programs and institutions that offer mentoring might better support mentoring relationships to expand and improve EABB outcomes. These factors include cultural, environmental, social experiences of youth, the strategic selection of program practices that are aligned with EABBs (e.g., setting goals with youth, teaching youth how to cope with stress), and the use of more carefully designed experiments that focus on measuring and improving EABBs. Finally, the review suggests that attention to implementing these enhancements is limited; but that when programs do adopt and implement these programmatic enhancements, mentoring can have a greater positive effect on EABBs.

In addition to the formal review of research on mentoring and EABBs, Implications for Practice based on this research are also included. These practice recommendations focus on actions that mentors or program staff could take to support development of positive EABBs, including the identification of root causes of negative EABBs, supporting growth mindsets and persistence skills, providing emotional support and encouragement, facilitating referrals to tutoring or other direct academic supports, working collaboratively with parents around academic challenges, and both direct advocacy on behalf of the child within schools and teaching youth to advocate for themselves to address points of disconnection. Links to relevant resources and training are provided when relevant.

  • Introduction

    Youth mentoring programs are predicated on the notion that a supportive, trusting non-familial adult (i.e., a mentor) can help facilitate the youth’s social, academic, and behavioral development in positive ways.1 Educational attitudes, beliefs, and behaviors (EABBs) refer to the attitudes, beliefs youth have about their experiences in school (e.g., academic self-efficacy, connectedness to peers and teachers) as well as the behaviors (e.g., classroom participation, prosocial helping) they exhibit during the school day. Because EABBs are thought to be causal and maintaining factors that influence academic (e.g., grade point averages, test scores) and behavioral (e.g., delinquency, behavioral referrals) outcomes, they are often targeted by mentoring programs as a means to influencing supporting youths’ performance in school.

    This review of mentoring research was conducted to examine four questions:

    1. What are the effects of mentoring on educational attitudes, beliefs, and behaviors (EABB) among youth?

    2. What factors condition or shape the effects of mentoring on EABB?

    3. What intervening processes are most important for linking mentoring to beneficial effects on EABB?

    4. To what extent have efforts to provide mentoring to youth with EABB as a priority outcome reached and engaged the intended youth, been implemented with high quality, and been adopted and sustained by host organizations and settings?

    For this review, we limited EABBs to the positive constructs: School belonging, school connectedness, academic self-efficacy, growth mindset, grit, self-discipline, homework, school value, study skills, self-regulated learning, goal-setting, non-cognitive skills, academic mindset, mastery beliefs, and academic motivation. In this review, we focus on positive EABBs given their long-term association to positive school outcomes like grades, school-related behavior, and economic success.2

  • What are the effects of mentoring on educational attitudes, beliefs, and behaviors (EABB) among youth?


      Historically, mentoring programs targeting EABBs have been viewed as a service to prevent and address negative school-related outcomes (e.g., dropout, suspension, and school failure). The earliest forms of formal mentoring programs, for example, paired non-familial adults with youth identified by schools or communities as having a problem behavior that a mentor could address. In particular, prevention-focused mentoring programs were thought to be one way to address students' deficits in schools by interrupting harmful causal influences and increasing their connection with other non-familial adult (or peer) mentors who could provide emotional support and teach youth skills necessary for success in school.1

      In addition to being a means for preventing and addressing negative school-related outcomes, EABBs have also been viewed as a way to promote positive youth development in school. This shift from mentoring programs as preventing unwanted EABBs to promoting EABBs coincides with similar shifts in the fields of education, human development, and psychology. Collectively, these fields have begun to recognize the importance of not only alleviating problems but also promoting positive youth development because positive outcomes are associated with long-term social, academic, and economic success.2,3,4 This is also consistent with a common motto of proponents of positive youth development: “Problem free is not fully prepared”.5 Students who report high levels of school engagement, feel connected with peers and teachers, and believe they can be successful in school are more likely to stay in school, be suspended less frequently, and exhibit fewer behavioral challenges.6,7 In addition, EABBs are associated with a host of positive developmental outcomes such as better grades, greater engagement in school, and increased resiliency.8,9

      Mentors are often thought to be an important source of support for the development of EABBs through role modeling, guidance, support, and encouragement. Mentors’ “lived experiences” as former (or current) students themselves and their ability to engage with students flexibly in relatively more personalized one-to-one or small group settings may offer them avenues for insight and influence not always available to others such as classroom teachers. Such advantages may be especially well-suited to strengthening what Farrington and colleagues (2012) refer to as “non-cognitive factors” in academic achievement. These include “academic mindsets”, defined as “the psycho-social attitudes or beliefs one has about oneself in relation to academic work” (e.g., growth mindset, sense of school belonging or connectedness, academic motivation, academic self-efficacy beliefs, and perceptions of the value of school), “academic perseverance” (e.g., grit, self-discipline), “learning strategies” (e.g., study skills and self-regulated learning), and “academic behaviors” (e.g., class participation and doing homework).10 Consistent with this possibility, various frameworks have pointed to interpersonal relationships that incorporate features of mentoring as a key mechanism for promoting non-cognitive factors, such as growth mindset (i.e., a belief that one’s ability to learn is malleable, rather than inherent or fixed). Farrington et al., also note that the development of these non-cognitive factors must be considered in the school, classroom, and socio-cultural context, writing that, “any given school and classroom context will reflect a wide variety of variables affecting student motivation and opportunity to learn” (p. 12)

      At times, mentoring has emphasized changing EABBs by targeting the child while neglecting to emphasize systemic issues that also contribute to the development of EABBs.11 Albright et al (2017) argues that “[t]here is likely a risk of mentors from privileged backgrounds to indirectly or directly communicating to youth that they should simply work harder or follow the same path that the mentor followed to obtain their social status. This may increase the likelihood that mentors will blame youth—and consequently, youth will blame themselves—for having difficulty overcome the barriers posed by an unjust system.” (p. 9) If mentors approach mentoring from the perspective that the primary barrier to the mentee is the lack of a role model, or some specific set of skills, they might be perpetuating these unjust systems.

      Collectively, this work points to EABBs as an important set of constructs that that have considerable theoretical potential to be strengthened through mentoring relationships and programs; but the potential for such contributions to be constrained by broader context of educational and other related systems within which they are necessarily embedded. It is worth noting, however, that some research on peer-led mentoring interventions in schools suggests that those types of programs have the potential to disrupt negative school peer ecologies and improve school climate and culture in ways that support the growth of EABBs.12

      All of the preceding considerations could pertain not only to mentoring relationships established through formal programs designed for this purpose, but also those that emerge more organically in the context of the youth’s time spent at school and in other settings. Theoretically, when such “natural” mentoring relationships involve teachers or other adults at school, they may provide opportunities for enhancement of EABB that stem from the mentor’s knowledge of the school environment (e.g., expectations for students, or cultural context), extensive time spent together in learning contexts (e.g., classroom instruction).13


      Mentoring programs have used a variety of methods to support mentees’ development of EABBs. One approach for promoting EABB development is to use mentors as a mechanism for delivering evidence-based practices.14 Researchers, for example, have embedded training on evidence-based cognitive coping strategies (e.g. muscle relaxation; meditation) into youth mentoring programs to support youths’ ability to cope with stressors in- and after-school, leading to small to moderate changes on adolescent life satisfaction, self-esteem, and school engagement.15,16sup> Other studies have focused on brief goal-focused forms of mentoring which has been found to produce small to moderate positive effects on grades, misconduct, and life satisfaction in some studies.17,18,15 McQuillin and Lyons (2016)18 incorporated practices such as having mentors and youth jointly set goals, teaching youth how to cope with stress, and helping youth organize and manage homework. In addition, other programs have used evidence-based study strategies to improve self-efficacy for school and grades, or to provide positive reinforcement for prosocial behavior in specific school contexts (e.g. the lunch table).19,20 Thus, it seems that effects of mentoring programs on EABBs improve when mentoring programs teach youth specific skills (e.g., goal setting, organizational strategies, how to cope with stress) and use mentors to help youth practice these skills over the duration of the mentoring relationship.(for an expanded discussion see McQuillin et al., 2020). Cumulatively, these small to moderate positive effects of mentoring are comparable to other social-emotional interventions,21 but smaller than more focused tutoring interventions that are more focused on academic skill development as opposed to EABBs.22,23

      Despite the small to moderate positive effects observed in some studies (described above), other studies produced no effects across a range of EABBs.24 Several large scale randomized controlled trials of mentoring have found no or small positive effects on youth EABBs.16,25,26 For example, Herrera et al. (2007)25 randomly assigned nearly 1,200 students in grades 4 through 9 to a Big Brothers Big Sisters school-based mentoring program or control and, at the end of the school year, found small, statistically improvement (i.e., more favorable scores for youth assigned to receive mentoring) on self-perceptions of academic ability as well as teacher ratings of quality of class work and number of assignments completed, but no improvement in classroom effort. After 15 months (through fall of the next school year), statistically significant improvements were not found on any of these outcomes, indicating that improvements apparent at the end of the school year were not sustained into the following school year, when only about half of the youth in the treatment group were still paired with a mentor through the program. In general, the effect size of these evidence-based programs on EABB development is positive and ranges from small to moderately positive effects.

      Alternative mentoring models, including group mentoring and cross-age peer mentoring, have sought to capitalize on the potential for positive peer influence on EABBs. For example, Project Arrive, a school-based group mentoring program designed to facilitate adaptation to high school for 9th graders identified as being at high risk for dropout, found significant improvements in participants’ sense of school belonging, perceptions of teacher (and peer) support, and perceptions of meaningful involvement in school, relative to non-mentored comparisons using a quasi-experimental design.27 Better Futures, a group mentoring program for youth transitioning out of the foster care system, was found to produce higher levels of preparation for post-secondary education among program participants randomly assigned to the program as compared to those who did not receive the treatment.28 Take Charge, a similar program that targeted youth ages 14-18 who were involved in the foster care system and received special education services, found small but significant differences related to school engagement favoring program participants who were randomly assigned to receive a mentor as compared to those who did not receive this program.29

      It is also worth noting that studies of programs in which older youth serve as mentors to younger students have also demonstrated benefits to youth mentors that may contribute to their own EABBs. In a review of peer mentoring programs, authors found associations between peer mentoring and perceived gains in school connectedness and social support at school, as well as interest in taking on leadership roles within school.30 Collectively this work shows promise for alternative models of mentoring including group and cross-age peer mentoring.

      Finally, other studies have examined and found significant associations between natural mentoring relationships and EABBs. In a longitudinal study of several hundred high school students, those reporting a school-based natural mentoring relationship (i.e., with a teacher or other adult at school) reported significantly increased feelings of attachment to school at a 1-year follow-up, controlling for their initial feelings of attachment.31 Similarly, in a longitudinal study of African-American adolescents, it was found that having a natural mentor while in high school was associated with stronger beliefs in the importance of school for later success in 12th grade32; another study from the same project found that youth reporting at least one natural mentor was associated with stronger attitudes of school importance.33 Finally, in a cross-sectional study, Latino/a high school students who reported a natural mentor correlated positively with expectancies for future academic success than those without a mentor.34 These studies indicate that institutions beyond formal mentoring programs may benefit from encouraging natural mentoring relationships.

      When viewed collectively, the field of mentoring is somewhat limited in the inferences researchers may make about how youth mentoring affects EABBs. These limitations are caused by inconsistent results across studies, a lack of understanding in the differences between programs tested, and inadequate documentation of what mentors do when they are with mentees.15 Perhaps the best available information is from meta-analyses. A meta-analysis of school-based mentoring (SBM) for adolescents concluded that, on average, the effects of SBM on adolescent attitudes and behaviors about school were near zero.35 In addition, a more recent meta-analysis found that estimated effects of youth mentoring programs on school engagement was small.36 Importantly, there was substantial variability within these effects, clouding conclusions around the general effectiveness of mentoring on EABBs.


      1. Mentoring programs of varying types and formats have a demonstrated potential to strengthen EABB with the size of the effects on outcomes appearing, on average, to be small in magnitude; specific types of EABBs showing evidence of responsiveness to mentoring include school connectedness, school engagement, and attitudes toward school.

      2. Naturally-occurring mentoring relationships, as studied to date, appear to be associated with stronger EABBs relating to attachment to and valuing of school, including prospective improvements in these EABBs in accordance with a possible causal contribution of mentoring to these outcomes.

      3. There is some tentative evidence that more targeted, evidence-based approaches to influencing EABBs will produce stronger results.

  • What factors condition or shape the effects of mentoring on EABBs?


      To better understand positive changes on EABBs, studies have also examined conditions that promote (or inhibit) the development of EABBs. Much of this work is based on current understanding of ecological factors that influence child development. Because mentoring programs are historically based on the notion that mentors support youth development by providing a relationship with another non-familial adult that can support their growth, mentoring programs frequently conceptualize the support they offer through ecological models of human development.37 Ecological models describe the processes and systems in the child’s life that influence their development. Related to the development of EABBs, salient factors that influence the development of EABBs are those in a youth’s environment, such as their family support for school,38 student-teacher relationships,39 and well as their relationships with peers in school.40

      Research indicates that mentoring relationships do not occur in a vacuum, and are influenced by, and influence, other relationships and contexts. For this reason, social relations (e.g., parent-child, mentor-mentee, teacher-student) are thought to be important sources for helping children to develop EABBs.41,42,28 Mentoring programs might benefit from considering how mentoring relationships influence these contexts, how these contexts might affect mentoring outcomes, and how, as programs, they might promote more positive contexts and relationships. For example, Chan et al. (2013)43 found that stronger relationship quality between mentor and mentee predicted stronger relationships between children and parents, children and teachers, and between parents and teachers. In addition, researchers have also identified how social and cultural factors influence a child’s experiences in school and, therefore, influence the development of EABBs. Garcia-Coll et al (1996)44 Integrative Model of Child Development, for example, describes how social position variables (i.e., a person’s race, social class, ethnicity, and gender) interact with sociopolitical variables (e.g., racism, prejudice, discrimination, and oppression) in ways that lead to residential, economic, social, and psychological segregation. The combined effect of the sociopolitical context and subsequent segregation means that, for some youth, some environments are more conducive to positive development (called promoting environments) whereas other environments may not be conducive to positive development (called inhibiting environments). The consequence of this model is that, even among youth who go to the same school or grow up in the same community, may have different experiences based on their social-position and the sociopolitical climate within the community.

      Empirical studies find that youth experiences of racism, prejudice, discrimination, and oppression negatively affect the development of EABBs and that youth-adult relationships can either mitigate or magnify these effects. Researchers, for example, have found that students’ attitudes about how racially fair their school is (i.e., students’ perceptions of race-based discrimination perpetuated by teachers and administrators) influence school engagement, suspensions, and disciplinary referrals,45,46 indicating that schools that feel more racially fair also produce stronger EABBs. Presumably, the same may be true for mentoring programs. In addition, researchers have also found that youth-adult relationships may attenuate the negative effects of race-based discrimination has on youth engagement in school. For example, researchers have found that youth who have strong relationships with adults that promote positive attitudes about racial identity while also supporting resilience tend to also report higher levels of EABBs (e.g., school engagement, school connectedness).47,48 In contrast, other researchers have found that experiences with discrimination and racism may reduce mentoring relationship quality.49 A practical implication of this work is that mentoring programs might achieve stronger effects when they simultaneously provide culturally responsive mentoring and work to improve racial fairness, reduce racist experiences, and promote positive racial identities within the institutions or communities in which they provide services. In summary, mentoring programs that exclusively focus on influencing the individual may miss opportunities to enhance the contexts in which the youth live, thereby reducing the effectiveness of mentoring interactions.

      Mentoring programs focused on EABB development may benefit from considering how programming interacts with youths’ awareness of, and attention to, contextual factors that influence EABB development. Ellis et al. (2018)50 described how mentoring may be able to support Black male development, writing that

      Mentoring also provides an opportunity and place for Black males to discuss the negative and positive narratives around being a Black male in America. Depending on the degree of trust developed in a mentoring relationship, Black males can learn more about how to recognize and manage negative academic racial stereotypes about Black males through the lived experiences and strategies implemented by their mentors. The wisdom shared and learned between mentor and mentee can serve as a protective factor in preventing the internalization of negative academic racial stereotypes for Black male adolescents. (p. 917)

      Others have suggested that a stronger emphasis on the integration of evidence-based practices might influence the effectiveness of mentoring on EABB outcomes.12 For example, intentional efforts to match the context (e.g. school vs. faith-based) of the mentoring program, the structure (e.g. 1-on-1; e-mentoring, etc.) of the relationship, and the goals of the program (e.g. EABB) with evidence-based practices might improve the effectiveness of specific practices on program outcomes. Mentoring practices or programs focused on EABB development alignment of context, structure, and goals means that programs address the mechanisms that are proximal to EABBs. Using Garcia-Coll’s model as a guide, this means that mentoring activities address both individual (e.g., attitudes toward school) and systemic factors (e.g., classroom and school climate, racism, sexism) that influence EABB development.

      Related research in both the areas of mentoring and after-school programs suggests that offering specific guidance or instruction in relevant skills could be important for strengthening EABBs. Durlak et al (2011)51 conducted a meta-analysis of after-school programs seeking to enhance the personal and social skills of school-age youth, for example, found that the presence of four recommended practices for skills training (SAFE: sequenced, active, focused, and explicit) was associated with enhanced program effects on outcomes. In general, these recommendations translate to having structured, step-by-step instructions to actively involving youth on improving specific outcomes. Such structure may be especially important in alternative mentoring approaches, such as group and cross-age peer mentoring programs that tend to hold meetings at a regularly scheduled time and place, given logistical and developmental concerns (e.g., coordinating multiple schedules, supporting teen mentors with emerging skills).22 The use of a curriculum to guide mentoring sessions can support skill development and relationship building, as suggested by qualitative research on the Campus Connections program; however, strict reliance on a curriculum can be a barrier to relational development if it prevents discussion of meaningful topics to mentees.52 In general, research suggests that the value of a balance between structured curriculum-focused activities and the development of a positive, authentic relationship that affords some flexibility in programming.


      In mentoring research focused on promoting EABB, knowledge of models of child development has been used to explain the conditions under which mentors may positively influence EABBs. Historical studies of youth mentoring regarded the mentor as influencing EABBs by providing an additional non-familial adult to act as an older friend who can support youth development by helping them cope with stressful life events, teach them new skills, or help them think about their future goals.1

      Consistent with Garcia-Coll’s previously described Integrative Model of Child Development, other studies of mentoring and EABB suggest that the sociopolitical context is an important condition that influences the development of EABBs. Ellis et al. (2018),50 for example, found evidence that mentors can help shape academic self-efficacy by helping youth navigate experiences related to racial stereotypes. The development of EABBs, from this perspective, suggests that mentors not only have the capacity to influence development through the one-on-one relationship that they develop with youth; but that sociopolitical factors (e.g., racism, prejudice, discrimination, and oppression) create scenarios in which youth’s school environment does not support their positive development (i.e., creates an inhibiting environment). Thus, the sociopolitical climate within youths’ school and community context appears to be a critical contextual factor in how mentors are (un)able to shape the development of EABBs of their mentees.9 This suggests that attempts to simply focus on integrating individual practices designed to increase (or prevent the decline of) EABBs may be misaligned with the causal and maintaining factors of EABBs in the children’s environment. Rather, programs may benefit from incorporating programming directly relevant to the youth’s social context (including race and identity) and community.

      By centering issues of race and identity into mentoring services, programs give agency to youth to address systemic issues present in their school or community context that influence EABB development. Smith and Hope (2020)53 used participatory action research (i.e., research that is led by the youth participating in mentoring programming) to increase mentee’s agency in facilitating systemic change. The authors explain that:

      [Youth participatory action research] allows Black boys the radical space to reimagine and redefine their relationship with their school communities; it offers school communities yet another chance to reimagine and strengthen their commitment to Black boys. (p. 563)

      EABB development is shaped by youth experiences in school, at home, and in their community. As described by Garcia-Coll (1996)44 and others, the development of a youth’s experiences and attitudes about school are shaped by an interaction between the individual and their environment. It is not logically possible to separate the experiences of the individual child from the environment in which they develop. Understanding this developmental context is essential for understanding the conditions by which mentoring programs may facilitate the development of EABBs; yet research on effective practices in this area is very limited.

      Characteristics of the mentors and mentees can also play an important role. For example, whereas meta-analytic studies (e.g., DuBois et al., 201154) have found few overall differences in the effectiveness of 1:1 as compared to group or peer mentoring approaches, it may be that adult mentors play a more influential role in promoting EABBs than peer mentors. An evaluation of Big Brothers Big Sisters’ school-based mentoring programs found that children with teen (high school-age) mentors benefited less than those with adult mentors with regard to classroom effort, difficulty in class, and intentions to go to college.55 This is not to say that teen mentors cannot play an important role in fostering EABBs. For example, a secondary analysis of data from the High School Bigs peer mentoring program found that high school students who expressed more positive attitudes toward youth were particularly effective with more academically disconnected mentees.16 Alternatively, there have been a few studies of peer-led behavioral interventions in schools that suggest peers might be more effective influencers of youth behavior than teachers, on non-academic behaviors (e.g., healthy eating56).

      Some research has also looked at the effects of integrating practices or activities into mentoring programs and relationships that reflect an intentional focus on promoting particular EABBs.57 Perhaps the most evident example of this integrated practice is focused on promoting growth mindsets in mentees. The theory underlying growth mindset posits that peoples’ perceptions of how stable (i.e. “fixed”) their intelligence is influences their effort and subsequent outcomes. This theory hypothesizes that students who believe that the brain functions can be strengthened through practice will typically exert more sustained effort towards difficult tasks, and may be less likely to feel defeated by failure.

      At least three separate studies have investigated potential benefits of the incorporation of intentional activities to promote growth mindset in mentoring programs.58,59,60 In perhaps the most noteworthy of these investigations,60 college students mentors for a sample of largely minority and low-income students entering junior high school were randomly assigned either to encourage their mentees to view intelligence as malleable (incremental) or to attribute academic difficulties in the seventh grade to the novelty of the educational setting (attribution)—both as strategies for reducing effects of stereotype threat on test performance—or to provide general mentoring oriented toward substance use prevention. Illustratively, as described by the authors of this study in the incremental condition, mentees learned from their mentors “that intelligence is not a finite endowment, but rather an expandable capacity that increases with mental work. To reinforce the scientific validity of this perspective, the mentors taught students some facts about the brain and how it works.” Students in this and each of the other conditions also created their own websites in collaboration with their mentors to convey the information they had learned. Results showed that females in both of the stereotype threat reduction conditions earned significantly higher math standardized test scores than their counterparts in the general mentoring condition; similarly, as a group, the students in the threat reduction conditions earned significantly higher reading standardized test scores than students in the general mentoring condition. Although corresponding changes in growth mindset or related EABB are plausible for students in the student stereotype threat reduction conditions given their goals and the activities involved, effects on these outcomes were not reported. Two other more recent studies examined effects of “add ons” to existing mentoring programs that incorporated mentor training and activity resources for promoting growth mindset. In one of these studies, which involved a mix of school- and other site-based mentoring programs, youth in the more intentional mentoring condition showed significantly greater improvement in their growth mindset for their intelligence compared to those in the standard mentoring condition; a corresponding effect was not evident, however, in the other investigation which focused on the Big Brothers Big Sisters community-based mentoring program.59 It is unclear why some studies found positive effects of these practices and others did not.

      Yet, the positive findings on growth mindset are consistent with research on school-based mentoring, suggesting that a focus on goal-setting activities in which mentors provide constructive feedback to mentees in school-based programs may be key to achieving effects on academic outcomes. These findings have also been corroborated by recent work investigating the role of relationship quality and goal-focused activities, wherein researchers found that outcomes for youth are maximized when young people report strong relationships and frequent engagement in goal setting and feedback, whereas the effects are much smaller when one of these pieces are missing.61


      1. Interactive processes between the cultural, sociopolitical, and environmental context in which children live influence how young people experience, and benefit from, youth mentoring programs focused on EABB outcomes.

      2. Research suggests that experiences of racism and discrimination can influence the relations between mentoring and subsequent EABB outcomes; thus this might be one environmental factor of interest for mentoring agencies and advocates.

      3. Research also indicates that mentoring programs can increase effects on EABBs when practices encourage strong relationships and more goal-focused activities.

  • What intervening processes are most important for linking mentoring to beneficial effects on EABB?


      Linking mentoring practices to EABB improvement requires that program activities target EABBs. Although many different models exist that describe this process,62,63 in general, this is a cyclic process by which programs select, implement, and evaluate the effectiveness of activities. (1) First, selecting activities means that programs (or mentors) choose activities that have prior evidence for effecting positive changes on EABBs. Programs or mentors may find additional research-informed activities by accessing publicly available databases (e.g., that summarize available evidence of program practices. (2) Second, implementing activities requires that mentors and mentees use the activities in the way the program or activity was intended. That is, even if programs choose activities that are likely to influence EABBs, these activities are unlikely to impact EABBs if mentors and mentees do not use these activities or if they fundamentally change the activity. (3) Third, programs and mentors must monitor and evaluate the implementation and impact of the selected activities to assess if the activity had the intended effect for youth participating in the program. If, through this process, programs find activities have null or minimal effects on the desired outcomes, this requires programs to adjust practices and start at step one (i.e., selection).

      An additional factor that may be important in mentoring to beneficial effects on EABBs could be the nature and quality of the mentoring that the youth receives. Theory and related research that has focused on mentoring as a contributor to other types of outcomes points to a wide range of considerations that could be important in this regard.64 These include, but are not limited to, the duration of the mentoring relationship, whether EABBs are addressed in the activities that mentors and mentees engage in or discuss, and the degree to which the mentee sees the mentor as a trusted source of guidance. \


      Although the iterative development process described above provides a framework for selecting and maximizing the impact of mentoring on EABBs, this process is not frequently observed in the research on mentoring and EABBs. Central to this process is clearly documenting the desired activities, measuring the implementation of activities, and evaluating the quality or influence of these activities. In our review of 23 studies of mentoring and EABBs, 65% (n = 15) described (usually in brief) the training and support programs offered to mentors. However, only 47% (n =11) included any information on the extent to which programs monitored the implementation of the activities as intended, and only 17% (n=4) reported some formal manual or curriculum designed to communicate the desired activities.

      Although the existing literature is limited because the mentoring activities are frequently underspecified (i.e., researchers do not sufficiently define or describe what occurs in the mentoring relationship), programs focused on EABB development tend to specify program practices in terms of developmentally-focused mentoring or instrumentally-focused mentoring activities.14,65 Developmental models of mentoring focus on relational aspects between the mentor and mentee and emphasize the need to develop a close, long-lasting bond between the youth and adult. These programs are based on developmental research suggesting that close youth-adult relationships are necessary for positive youth development. Others have suggested that the mentoring relationship is more accurately conceptualized as a vehicle for delivering evidence-based practices known to produce positive outcomes,14 wherein mentors incorporate more structured goal-directed activities as opposed to exclusive reliance on relationship development. In practice, the categories of developmental versus instrumental mentoring are a false dichotomy, and most mentoring relationships and programs are rarely exclusively developmental or instrumental, though they may emphasize goal-setting versus long-term relationships differently. It is also likely that some more developmental programs might desire critical periods of more instrumental mentoring activities (e.g. if a youth is failing at school), and that some instrumental programs might relax goal-setting and curricular expectations if they hurt relationship development.

      Researchers have found that the relationship quality that mentees report is more strongly associated with some EABBs (e.g., delinquency and misbehavior) and less strongly associated with other EABBs (e.g., standardized test scores).54 This might indicate that for some outcomes, the relationship between the mentor and mentee is more critically important than for other outcomes. In addition, researchers also find that mentoring impact may be maximized when mentors attend to both relational and instrumental aspects of the relationship.66 In particular, researchers have found that mentors and mentees who report a strong relationship while also setting goals and providing mentees feedback on their school performance tend to have stronger effects on some EABBs (e.g., delinquency) as compared to mentors who have weaker relationships nor engage in more instrumental activities with their mentees.67 Ellis et al. (2018)50 summarized the need to blend developmental and instrumental aspects of the mentoring relationship, writing that:

      Mentoring that offers a consistent space for communication, is structured around goals, and is well-matched according to the needs of the mentee is suggested in an effort to promote school efficacy. (p. 918)

      The inherent flexibility involved in selecting mentoring activities suggests that the manner and extent to which mentor programs and mentors continuously evaluate the effectiveness of the choices of activities they make could be an important intervening process shaping ultimate effects of programs on EABBs. Describing a qualitative approach to evaluation, Mboka (2018)68 writes:

      Another key feature of relational-based mentoring is the freedom to work on thinking and behaviors that the established relationships subjectively deemed to be worthy of attention in assisting the protégé. Inherent in that freedom is the responsibility to self-check or self-validate approaches by keeping account of what behaviors are being worked on, how it is being worked on, and what results are being generated. (p. 2298)

      The extent to which youth experience mentoring as supporting their psychological needs for competence, autonomy, and relatedness may play a critical role in promoting EABBs. Although there were no overall effects of participating in Metodología TUTAL, a school-based group program, youth who experienced support from the mentoring group for those psychological needs reported improvements in academic self-efficacy and academic well-being. In group mentoring models, youths’ experiences both with adults and peers can contribute to growth in EABBs. In the evaluation of Project Arrive, participants’ reports of the quality of their relationship with mentors and of a positive group climate both contributed to improvements in their perceptions of teacher support, whereas only group climate contributed to improvements in meaningful involvement at school.69

      Limited research also has examined the potential implications of the nature and quality of the mentoring relationship. In a study of Latinx high school students, those reporting higher levels of instrumental quality for their natural mentoring relationships (as assessed by a measure asking youth the extent to which they experienced their relationships with their mentors as growth-oriented and involved with learning) reported more perceived benefits of education than those reporting more relational quality (as assessed by a measure of the extent to which youth reported feeling close, happy, and satisfied in their relationships with their mentors), Neither relationship measure, however, was associated with change in these outcomes at a one-year follow-up. Interestingly, though, greater reported instrumental relationship quality was linked to increased perceived benefits of education at the later time point via its intermediary association with greater intrinsic motivation for academic learning. The extent to which the mentoring approach is tailored to program goals may play an important role, as well. For example, an evaluation of CyberMentor, a program designed to increase girls’ exposure to and interest in science, technology, and mathematics (STEM) education, compared 1:1 and group versions, finding that implementation in a group format contributed to greater proportion of STEM communication, STEM-related networking, and in elective intentions.


      1. The variability in mentoring program practices along with the implementation and evaluation of these practices means that it is difficult to generalize about the types of programmatic practices that are most likely to increase EABBs within specific programs and settings.

      2. The impact of mentoring programs on EABBs may be most likely to be maximized if programs follow a development cycle that prioritizes the selection of research-informed practices, consistent monitoring of implementation of these practices, and evaluation of effectiveness of the selected activities.

  • To what extent have efforts to provide mentoring to youth with EABB needs reached and engaged the intended youth, been implemented with high quality, and been adopted and sustained by host organizations and settings?


      In a systematic review of over 500 intervention studies, Durlak and Dupree (2011)70 found that monitoring the implementation of practices increased the effect sizes of interventions by a factor of three. This finding is similar to a youth mentoring specific meta-analysis, also included in the aforementioned synthesis, which reported monitoring implementation was associated with greater program effectiveness.71 Thus it appears that strong implementation, in general, produces larger outcomes, regardless of specific outcome.

      However, a presupposition to the notion of implementation monitoring is that some specific practice, or set of practices or procedures, is in fact available to be monitored. Thus, it could be that more specific practices lend themselves to easier monitoring and larger effect sizes. This hypothesis is consistent with more recent meta-analytic work which indicates that more targeted programs, presumably based on more focused intervention strategies that lend themselves to easier implementation monitoring, produced larger effects, on average72, whereas programs that are more diffuse in their focus, or adjust practices on the basis of match specific needs and desires, might produce smaller effects. Yet some researchers have argued that these smaller effects might not be representative of the actual pragmatic helpfulness of the mentor, rather are a product of a mismatch between the experimental evaluation of mentoring and the wide and varying ways in which mentors help others.15 This is consistent with the original evaluation of Big Brothers Big Sisters of America, wherein over 40 outcomes were considered.

      Importantly, research on mentoring activities that promote EABBs indicates that effective activities prioritize both developmental (i.e., strong trusting relationship between mentor and mentee) and instrumental approaches (i.e., setting goals, providing feedback) and are more likely to be effective in creating the conditions for positive EABB development.65 Given these considerations, and the current literature’s lack of information on implementation monitoring, it is difficult to make strong conclusions to this research question. Yet some recent research highlights general trends in implementation within mentoring (and EABBs targeted mentoring).


      MENTOR: The National Mentoring Partnership’s National Mentoring Survey73 found that roughly 45% of mentoring agencies engaged in implementation evaluation that emphasized assessing or improving fidelity to the program model. It was unclear from this survey what percent of mentoring agencies engage in routine implementation monitoring. However, roughly 50% of responding programs endorsed some type of formal curriculum or mentor manual that was used to guide activities and experiences. This data does not provide information on what specifically the focus of the curriculum was, or how instrumental the guidelines for mentoring were. Other studies on large national mentoring organizations shed some light on this uncertainty. In the most recent evaluation of the Big Brothers Big Sisters school-based model, approximately 11% of mentors endorsed having some instrumental goal, whereas 79% endorsed broad relationship goals. From this literature, it appears that most mentors and mentoring programs do not target specific instrumental goals in their relationships, making it unlikely that they are also monitoring implementation.

      Despite the limited number of studies that sufficiently defined program practices and monitor implementation, there was considerable variability in how programs were defined and how implementation was measured. For example, one experiment described implementation monitoring in brief: “[implementation monitoring] includes monthly meetings and informal communication by phone and email with a designated coordinator.” (p.412). Other studies provided more detail on the amount of implementation monitoring and the specific types of implementation products. For example, Elledge et al., (2010)74 write “Mentors completed a 4-item log sheet after each visit...Mentors met each week as a small group (4-6 mentors) with a graduate research assistant who collected weekly log sheets, monitored the consistency of visits, and addressed any difficulties that arose.” Exceedingly rare were studies that reported any measurement of the implementation beyond the number of visits. One evaluation, however, included semi-structured interviews with mentors following the program, and weekly logs between sessions that documented mentor activities and experiences. Both of these products were then coded and analyzed and variability within logs predicted mentor effectiveness in the program.75 Yet, such measurements are the exception, not the rule. McQuillin et al., (2018)15 conducted a review of how mentoring programs described and defined the mentoring treatment and reached a similar conclusion as we found in this review. The authors concluded that “researchers, on average, are not specifying aspects of the treatment constructs that are presumed to be the essential elements of mentoring relationships.” (p. 219) Because of this reality, it is unlikely that programs are currently monitoring, documenting, or encouraging the specific application of practices or procedures linked to EABBs.


      1. Available research is insufficient for understanding the extent to which programs that target EABBs are successful at reaching strong implementation quality.

      2. Research within and outside of youth mentoring highlights the promise of implementation monitoring as a useful strategy for improving the quality of research to understand how, and for whom, mentoring has the largest effect on EABB development.

  • Implications for Practice

    (Mike Garringer, MENTOR: The National Mentoring Partnership)

    Mentoring programs have long valued academic-related goals and outcomes for their services. A 2016 survey of youth mentoring programs nationally73 found that 36% of all programs reported “academic enrichment” as a core outcome, with another 18% and 15% emphasizing college access and educational attainment outcomes, respectively. Even if those outcomes are not perfectly analogous with the EABBs discussed in this review, chances are that the vast majority of those programs are trying to achieve those matriculation and academic performance outcomes through the strengthening of positive EABBs. These are not tutoring programs, after all—their success on academic issues depends on influencing how youth perceive their own relationship to school and learning as much as anything.

    Thankfully, the review here provides plenty of hints as to things practitioners can do to support the development of positive EABBs in mentees. While the specific strategies deployed by any program will likely vary due to local circumstances and resources, the following principles may set the stage for success:


      One of the clear takeaways from the review is that there are many, many reasons why EABBs may be suffering, both at a large scale (such as in a school that has a pervasive bullying culture) or for individual youth (a specific learning disability, for example). A major pitfall for programs might be thinking too narrowly about the root causes of negative EABBs. Some youth might be getting negative messages about education at home or be influenced by stereotype threat that makes them believe they are not inherent “learners.” Others may be the victim of negative peer relationships while at school, or be in a high-conflict relationship with a teacher. More systemically, as the review notes, youth may be noticing patterns of institutional racism or marginalization, or simply feel that their efforts in the classroom are not supported or valued by a burned-out teaching staff. As noted earlier, school climate issues or widespread misconduct can create an environment where many students feel uncomfortable and not ready to engage or learn. And lastly sometimes negative EABBs are just the byproduct of struggling to understand or master learning content. We’ve all had that one class that was harder than we thought it could be and perhaps made us doubt our abilities or intellect in new ways.

      The point is that the youth in a mentoring program may be experiencing radically different root causes of their EABBs and mentoring relationships are unlikely to offer the right kinds of support if the adults running the program are assuming what those causes are or lazily attributing a lack of youth engagement to simply not caring or misaligned values. Directly asking youth about why they may have negative EABBs can reveal both deeper systemic challenges and personal struggles that are likely at the heart of those feelings, as can data from school climate surveys, teacher report cards, and even disciplinary referrals. So, practitioners should take the time to learn exactly why each youth in the program is struggling with how they think about school and their education and recognize that each of them may need a unique mentoring experience to get back on track. This can be done through surveys at different points in time or through interviews with youth as they enter the program—it could also be positioned as something that mentors are tasked with exploring with the student early in their relationships. One resource that might help with broader surveying, especially for school-based programs, is the great collection of school climate and related surveys made available through the National Center on Safe Supportive Learning Environments. The National Mentoring Resource Center’s Measurement Guidance Toolkit also offers many evaluation tools that could be used to gauge how young people are feeling about their academic abilities and institutions. It’s also worth noting that this toolkit offers measures that could help assess whether the work of mentors in critical areas such as teaching perseverance, building academic identity, and guiding youth on self-advocacy are making a difference.

      Regardless of how a program gets at this information, the main idea is to not guess or assume why mentees might have negative views about education, but to find out what’s really happening to drive those feelings. This might be hyper-specific to each child in the program, but practitioners might also find that there are some very specific issues (or individuals) in the school that are causing many students to disconnect or give up, so don’t pass on the opportunity to tackle pervasive systemic causes too.


      Given that EABBs might be influenced by many factors, it may seem challenging to think about what kinds of mentors would be a good fit for supporting youth on these myriad challenges and concerns. But there are a few traits that, if we read a bit between the lines of the review, seem like they might be helpful to keep in mind:

      • Mentors who are empathetic – Struggling in school, especially if that struggle has made one doubt their own abilities and future, can be a deeply depressing and alienating experience. In order for mentors to make a connection with a young person, to bring that relational focus the review suggests will contribute to improving EABBs, you will need to recruit individuals who have, and can express, empathy for what the mentee is going through. Mentors who struggled with EABBs themselves, mentors of similar backgrounds and identities who may be able to relate to certain causes, and those who have an appetite for examining structural or institutional inequities may be particularly well-positioned to do good work here, although other individuals can certainly be effective provided they have a decent understanding of concepts like fairness and equity and can relate to stories of struggle or alienation. The main idea is to find individuals who will really listen openly to the young person, express empathy, and avoid judgment or negative forms of problem-solving (e.g., unhelpful advice-giving, shaming, victim-blaming). The reviewers here do note that expressing care and forming a strong relationship is only part of the battle—instrumental support and evidence-based strategies are also needed here. But it seems likely that any mentor who wants to support a young person in reframing negative EABBs must be able to make those conversations safe, empowering, and caring above all else. Programs may be able to assess prospective mentors’ ability to express empathy and have supportive conversations on these topics by presenting some scenarios during pre-match interviews or trainings to see how they respond and offer coaching as needed. Some mentors may present serious deficits in expressing empathy and may be screened out, lest they deepen a young person’s disconnection from education through negative conversations and reactions to academic struggles.

      • Mentors who can check their own biases and examine their own frames of reference – Closely related to empathy is the ability of mentors to remember that they often bring a different set of experiences and privileges to mentoring relationships than the youth they are working with. It can be easy, as the review notes, for mentors to project their own struggles onto a youth and recommend solutions that worked for them, but that might be a poor fit for the youth they are mentoring. Adults can often fail to see how their own backgrounds and experiences might shape how they respond to a young person expressing negative EABBs and it can be tempting to rely on abstract solutions such as “grit” or increased “effort” rather than really hearing how the young person is perceiving things and honoring that as a starting point. Mentors who lack cultural understanding, responsiveness, and humility may be especially bad at understanding youth perspectives or seeing institutional problems that need to be addressed to make the learning environment more of a “promoting” one. Trainings and reflection tools, such as the those offered in MENTOR’s Critical Mentoring Toolkit, can help mentors reflect on the biases they bring based on their own backgrounds and offer chances to practice meeting young people where they are at when they view something like educational barriers differently.

      • Mentors who can advocate for institutional and system change – While many of the root causes of EABBs may be tied to the youth’s home or peers, or even the youth’s own thinking and values, many negative EABBs may be directly tied to negative aspects of the learning environment itself: A racist teacher, under-resourced classrooms and equipment, a toxic school climate. An effective mentor may be helpful in helping youth reframe their own internal thinking and values, but the most impactful thing they might do is directly advocate for change within that learning environment. An adult mentor’s voice may carry weight that a student’s does not. All adults involved in a mentoring program can advocate for healthier, better-managed school spaces. They may not be able to directly fix all the issues they encounter, but they should stand up for their mentees and lend their voice to efforts advocating for change where needed. That should be as much of a mentor’s responsibility as anything. Programs should think about how they can empower mentors to advocate around these topics, either through direct communication with their mentees’ teachers and administrators, through dialogue and action with and alongside parents, or through collective action, such as a group of mentors petitioning the school board or other administrative leadership for needed reforms. Training mentors on how to help young advocate for themselves in the school setting can also help. Many disabilities rights organizations, especially those working in the education space, have trainings and other materials that mentoring programs could adapt to teach both advocacy and self-advocacy to program participants. Not every mentor in every program will be able to directly advocate on behalf of their mentee within the school, but they can help empower the young person, their family, and the program to step in as needed, but only if they view advocacy as part of their role and recognize that many students become disengaged from school for perfectly valid reasons that need to be addressed.

      We often think of adult mentors as the primary vehicle for using relationships to turn around negative EABBs. But the review here hints at a few models that may decentralize the role of adults in this work. Group mentoring programs, in which consistent groups of youth meet with adult mentors and engage in meaningful activities, may operate under some different pathways to change than the more typical one adult-one child model. These programs may derive their benefit from the collaboration, mutuality, and strengthening of social-emotional skills and peer relationships as much as from any “wise advice” that comes out of the mouths of mentors. Group models may allow youth to openly discuss their EABBs and root causes, share the solutions some may have found for coping or changing with those issues, and create an empathetic environment where youth may overcome feelings of isolation or desperation around their academic goals or performance. Adults are still there to offer advice or to support more concrete problem-solving, but it is the positive and safe interactions and conversations with fellow students around EABBs that might offer the most value. Discussing challenges can normalize struggles and provide potential solutions, if not outright alleviating challenges related to school climate or negative peer culture.

      Peer mentoring models and relationships also have a track record of improving things like school connectedness, once again by offering youth something different and rewarding as part of the school day, bringing positive relationships into an environment that may be lacking in them, or simply by de-centering the voices of adults (who youth might rightly blame for creating the negative circumstances of their EABBs in the first place). So, for adults running or volunteering in mentoring programs, know that sometimes the best way to promote positive EABBs is by letting the young people lead and be in relationship with one another in confronting their negative attitudes and beliefs around education. If group or peer models are not possible to implement for a given program, adults can still honor youth voice in some creative ways. The Forum for Youth Investment offers a number of tools that can help adults center and honor youth voice in the design of program activities. And MENTOR’s guide on helping youth find a sense of purpose also offers many tools and tips for honoring youth voice (The Mentor’s Guide to Youth Purpose.)


      One of the stronger suggestions in the main review here is to weave concepts from other successful interventions that address EABBs into mentoring interactions. Obviously, mentors need to form good relationships with the youth they are mentoring, as those relationships not only set the stage for more intensive EABB work but likely have some direct benefit on EABBs through the provision of emotional support, encouragement, normalization of struggles, etc. But mentoring scholars such as Jean Rhodes and review co-author Sam McQuillin have been increasingly suggesting that mentors need to do more than just build a positive relationship—they need to build in practices and strategies that have been proven to work in addressing the mentee’s needs. In this case, there are strategies and interventions that can promote positive EABBs. Growth mindset work, in particular, may be something that can help students see their academic struggles in a new light and help them unearth new approaches that can get them feeling more positive about learning. MENTOR’s Growth Mindset for Mentors toolkit is one such resource that can help mentors ensure that students aren’t giving up on themselves or viewing their struggles in the classroom as inevitable. Motivational interviewing is another approach that programs may borrow from the world of counseling and psychology that may be particularly helpful here in addressing ambivalence about learning or issues around motivation (see the work of organizations like MINT for more information on how to apply motivational interviewing to various contexts). If the root causes of negative EABBs are related to school culture or leadership, any number of anti-bullying curricula or programs addressing school climate and behavior might be sources of mentoring activities or critical conversations. As noted in the review, more specified and prescribed practices by mentors might be more effective and easier for program staff to train on and monitor for implementation fidelity.

      It’s also worth noting on this front that a program’s mentors don’t always need to be the deliverers of an evidence-based intervention. If mentees happen to be struggling because of learning disabilities or comprehension issues, the best thing a mentor might do is provide emotional support and encouragement while a highly-skilled tutor tackles the root learning challenges that led to those negative EABBs. Or that mentor might advocate on behalf of their mentee for more professional development for teachers or increases in special education staffing that might better get at the primary causes of disengagement from school and learning. The combination of tutoring and mentoring might be a fantastic pairing here, since one of the main causes of youth becoming disconnected and discouraged around their education is simply struggling with the learning content. Whether by partnering with other service providers or training up your mentors to deliver proven interventions, make sure that you are not relying solely on “caring adults” and friendly relationships to address root causes that might need more than that.


      One of the most important things a mentor can do is to give feedback in ways that the mentee genuinely hears and internalizes. This is certainly not easy when it comes to issues around EABBs, which may really frustrate adult mentors who understand the value of education and the need to persevere through institutional challenges more clearly than a still-developing young person. It can be understandable when confronted with declarations of apathy, self-doubt, or hopelessness to shower the mentee with advice, warnings, punishments, and all manner of prescribed solutions. But those strategies are likely to frustrate the young person and damage a positive relationship, reducing the mentor to just another adult yelling at the mentee to care/improve. Training on some specific skills around offering feedback, active listening, reaffirming youth perspectives, and offering praise in addition to constructive criticism will go a long way in helping mentors push their mentees to change their behaviors or thinking without lapsing into what some call “righting reflexes” that often have no or negative results.

      Knowing how to set quality goals and track progress in affirming ways will also be helpful to mentors. Often, mentors come into a relationship with ambitious ideas about getting a mentee to college or drastically improving grades. But learning to meet the mentee where they are at, to truly hear their concerns and root causes of disengagement, and then come up with realistic and clear goals (together) to start heading in the right direction can be challenging. Mentors may need practice with staff or other mentors to be able to do this well and anticipate scenarios where the youth pushes back or reacts negatively. So, make sure you set aside training time for these skills this review suggest are critical.

      There are a few tools that can help both mentors and practitioners in integrating goal-setting and attainment into their mentoring work. The NMRC’s Measurement Guidance Toolkit recently added a youth-centered goal attainment measurement tool that can be used at the individual match or aggregate level. This might help focus mentor-youth interactions and would allow programs to track how well youth in the program are progressing toward goals at a macro level, offering extra support to matches that seem “stuck.” Online platforms, such as Ascend can also make the experience of supporting youth goals easier, while also teaching youth critical skills they can use long after the mentoring relationship ends.


      The reality for many youth is that their home environment may be the unfortunate source of some of their negative EABBs. A mentor working in the school (or in the community, as it’s important to remember that mentors in community-based programs can tackle many of these issues just as well as those embedded in the school, although some of the direct advocacy may prove more complicated for them) may be able to offset a lot of that. But this work will be much easier if parents and guardians can also reinforce positive EABB messages and strategies that the mentor is attempting to promote. This may require some additional outreach and engagement effort by program staff, but many of the strategies suggested here (e.g., growth mindset) have parent-specific materials and curricula that can ensure that the mentee is hearing affirming, positive educational messages both at school and at home. Other caring adults can also be brought into this mix. In fact, some studies of youth who left school (for example, Center for Promise, 2015)76 because of negative EABBs have found that a web of supportive adults is needed to reengage students and get them back on path to graduate. That web can certainly support direct learning and coursework, but the fight usually begins by turning around those EABBs and getting that young person motivated and confident to learn and engage their educational future. So whatever programs can do to get mentors and parents delivering consistent EABB messages may increase positive outcomes.

  • References

    1. Rhodes, J. E. (2005). A theoretical model of youth mentoring (D. L. DuBois & M. J. Karcher, Eds.).

    2. Proctor, C. L., Linley, P. A., & Maltby, J. (2008). Youth life satisfaction: A review of the literature. Journal of Happiness Studies, 10(5), 583–630.

    3. Larson, R. W. (2000). Toward a psychology of positive youth development. American Psychologist, 55(1), 170–183.

    4. Seligman, M., & Csikszentmihalyi, M. (2000). Positive psychology: An introduction. American Psychologist, 55(1), 5–14.

    5. Pittman, K. (1999). Youth Today: The Power of Engagement. Forum for Youth Investment.

    6. Niehaus, K., Rudasill, K. M., & Rakes, C. R. (2012). A longitudinal study of school connectedness and academic outcomes across sixth grade. Journal of School Psychology,50(4), 443-460.

    7. Thapa, A., Cohen, J., Guffey, S., & Higgins-D’Alessandro, A. (2013). A review of school climate research. Review of Educational Research83(3), 357-385.

    8. Lyons, M. D., Huebner, E. S., & Hills, K. J. (2013). The dual-factor model of mental health: A short-term longitudinal study of school-related outcomes. Social Indicators Research, 114, 549–565.

    9. Lyons, M. D., & Huebner, E. S. (2016). Academic characteristics of early adolescents with higher levels of life satisfaction. Applied Research in Quality of Life, 11, 757–771.

    10. Farrington, C. A., Roderick, M., Allensworth, E., Nagaoka, J., Keyes, T. S., Johnson, D. W., & Beechum, N. O. (2012). Teaching adolescents to become learners: the role of noncognitive factors in shaping school performance–a critical literature review. Chicago: Consortium on Chicago School Research.

    11. Albright, J. N., Hurd, N. M., & Hussain, S. B. (2017). Applying a social justice lens to youth mentoring: A review of the literature and recommendations for practice. American Journal of Community Psychology.

    12. St Vil, C., & Angel, A. (2018). A study of a cross-age peer mentoring program on educationally disconnected young adults. Social Work, 63(4), 327–336.

    13. Sánchez, B., Reyes, O., & Singh, J. (2006). A qualitative examination of the relationships that serve a mentoring function for Mexican American older adolescents. Cultural Diversity and Ethnic Minority Psychology, 12(4), 615–631.

    14. Cavell, T., & Elledge, L. C. (2014). Mentoring and prevention science. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of Youth Mentoring (pp. 29–42). SAGE Publications, Inc.

    15. McQuillin, S. D., Lyons, M. D., Clayton, R. J., & Anderson, J. R. (2020). Assessing the impact of school-based mentoring: Common problems and solutions associated with evaluating nonprescriptive youth development programs. Applied Developmental Science, 24(3), 215–229.

    16. Karcher, M. J., Davidson, A. J., Rhodes, J. E., & Herrera, C. (2010). Pygmalion in the program: The role of teenage peer mentors' attitudes in shaping their mentees' outcomes. Applied Developmental Science14(4), 212-227.

    17. McQuillin, S., Strait, G., Smith, B., & Ingram, A. (2015). Brief instrumental school‐based mentoring for first‐and second‐year middle school students: A randomized evaluation. Journal of Community Psychology43(7), 885-899.

    18. McQuillin, S. D., & Lyons, M. D. (2016). Brief instrumental school-based mentoring for middle school students: theory and impact. Advances in School Mental Health Promotion9(2), 73-89.

    19. Gregus, S. J., Craig, J. T., Rodriguez, J. H., Pastrana, F. A., & Cavell, T. A. (2015). Lunch buddy mentoring for children victimized by peers: Two pilot studies. Journal of Applied School Psychology31(2), 167-197.

    20. Hughes, J. N., Cavell, T. A., & Willson, V. (2001). Further support for the developmental significance of the quality of the teacher–student relationship. Journal of School Psychology39(4), 289-301.

    21. Taylor, R. D., Oberle, E., Durlak, J. A., & Weissberg, R. P. (2017). Promoting positive youth development through school-based social and emotional learning interventions: A meta-analysis of follow-up effects. Child Development, 88(4), 1156–1171.

    22. Dietrichson, J., Bøg, M., Filges, T., & Klint Jørgensen, A.-M. (2017). Academic interventions for elementary and middle school students with low socioeconomic status: A systematic review and meta-analysis. Review of Educational Research, 87(2), 243–282.

    23. Ritter, G. W., Barnett, J. H., Denny, G. S., & Albin, G. R. (2009). The Effectiveness of Volunteer Tutoring Programs for Elementary and Middle School Students: A Meta-Analysis. Review of Educational Research, 79(1), 3–38.

    24. Bernstein, L., Rappaport, C. D., Olsho, L., Hunt, D., & Levin, M. (2009). Impact evaluation of the US department of education’s student mentoring program. Final Report. NCEE 2009-4047. National Center for Education Evaluation and Regional Assistance.

    25. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., & McMaken, J. (2007). Making a difference in schools: The Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures.

    26. Karcher, M. J. (2008). The study of mentoring in the learning environment (SMILE): A randomized evaluation of the effectiveness of school-based mentoring. Prevention Science, 9(2), 99.

    27. Kuperminc, G. P., Chan, W. Y., Hale, K. E., Joseph, H. L., & Delbasso, C. A. (2020). The role of school-based group mentoring in promoting resilience among vulnerable high school students. American Journal of Community Psychology, 65(1–2), 136–148.

    28. Geenen, S., Powers, L. E., Phillips, L. A., Nelson, M., McKenna, J., Winges-Yanez, N., Blanchette, L., Croskey, A., Dalton, L. D., Salazar, A., & Swank, P. (2015). Better futures: A randomized field test of a model for supporting young people in foster care with mental health challenges to participate in higher education. The Journal of Behavioral Health Services & Research, 42(2), 150–171.

    29. Geenen, S., Powers, L. E., Powers, J., Cunningham, M., McMahon, L., Nelson, M., Dalton, L. D., Swank, P., Fullerton, A., & Care, other members of the R. C. to I. the S. of Y. in F. (2013). Experimental study of a self-determination intervention for youth in foster care: Career Development and Transition for Exceptional Individuals.

    30. Kupersmidt, J., Stelter, R., Karcher, M., & Garringer, M. (2020). Peer mentoring supplement to the elements of effective practice for mentoring. (p. 88). The National Mentoring Partnership.

    31. Black, D. S., Grenard, J. L., Sussman, S., & Rohrbach, L. A. (2010). The influence of school-based natural mentoring relationships on school attachment and subsequent adolescent risk behaviors. Health Education Research, 25(5), 892–902.

    32. Hurd, N. M., Sánchez, B., Zimmerman, M. A., & Caldwell, C. H. (2012). Natural mentors, racial identity, and educational attainment among African American adolescents: Exploring pathways to success. Child Development83(4), 1196-1212.

    33. Zimmerman, M. A., Bingenheimer, J. B., & Notaro, P. C. (2002). Natural mentors and adolescent resiliency: A study with urban youth. American Journal of Community Psychology30(2), 221-243.

    34. Sánchez, B., Esparza, P., & Colón, Y. (2008). Natural mentoring under the microscope: An investigation of mentoring relationships and Latino adolescents' academic performance. Journal of Community Psychology36(4), 468-482.

    35. Wood, S., & Mayo-Wilson, E. (2012). School-based mentoring for adolescents: A systematic review and meta-analysis. Research on Social Work Practice, 22(3) 257 – 269.

    36. Raposa, E. B., Rhodes, J., Stams, G. J. J. M., Card, N., Burton, S., Schwartz, S., Sykes, L. A. Y., Kanchewa, S., Kupersmidt, J., & Hussain, S. (2019). The effects of youth mentoring programs: A meta-analysis of outcome studies. Journal of Youth and Adolescence, 48(3), 423–443.

    37. Bronfenbrenner, U., & Morris, P. A. (2007). The Bioecological model of human development. In Handbook of child psychology. John Wiley & Sons, Inc.

    38. Demaray, M. K., Malecki, C. K., Davidson, L. M., Hodgson, K. K., & Rebus, P. J. (2005). The relationship between social support and student adjustment: A longitudinal analysis. Psychology in the Schools, 42(7), 691–706.

    39. Pianta, R. C. (1999). Enhancing relationships between children and teachers. American Psychological Association.

    40. Nelson, R. M., & DeBacker, T. K. (2008). Achievement motivation in adolescents: The role of peer climate and best friends. The Journal of Experimental Education, 76(2), 170–189.

    41. Keller, T. E. (2005). A systemic model of the youth mentoring intervention. Journal of Primary Prevention, 26(2), 169–188.

    42. Larose, S., Boisclair-Châteauvert, G., De Wit, D. J., DuBois, D., Erdem, G., & Lipman, E. L. (2018). How mentor support interacts with mother and teacher support in predicting youth academic adjustment: an investigation among youth exposed to big brothers big sisters of Canada programs. The Journal of Primary Prevention, 39(3), 205–228.

    43. Chan, C. S., Rhodes, J. E., Howard, W. J., Lowe, S. R., Schwartz, S. E., & Herrera, C. (2013). Pathways of influence in school-based mentoring: The mediating role of parent and teacher relationships. Journal of School Psychology51(1), 129-142.

    44. Garcia-Coll, C., Lamberty, G., Jenkins, R., McAdoo, H. P., Crnic, K., Wasik, B. H., & Vázquez García, H. (1996). An integrative model for the study of developmental competencies in minority children. Child Development, 67(5),1891-1914.

    45. Griffin, C. B., Cooper, S. M., Metzger, I. W., Golden, A. R., & White, C. N. (2017). School racial climate and the academic achievement of African American high school students: The mediating role of school engagement. Psychology in the Schools, 54(7), 673–688.

    46. Griffin, C. B., Metzger, I. W., Halliday-Boykins, C. A., & Salazar, C. A. (2020). Racial fairness, school engagement, and discipline outcomes in african american high school students: the important role of gender. School Psychology Review, 0(0), 1–17.

    47. Gray, D. L., Hope, E. C., & Matthews, J. S. (2018). Black and belonging at school: a case for interpersonal, instructional, and institutional opportunity structures. Educational Psychologist, 53(2), 97–113.

    48. Tatum, B. D. (2004). Family life and school experience: Factors in the racial identity development of black youth in white communities. Journal of Social Issues, 60(1), 117–135.

    49. Sánchez, B., Mroczkowski, A. L., Liao, L. C., Cooper, A. C., Rivera, C., & DuBois, D. L. (2017). Mentoring as a mediator or moderator of the association between racial discrimination and coping efficacy in urban, low-income latina/o youth. American Journal of Community Psychology, 59(1–2), 15–24.

    50. Ellis, J. M., Rowley, L. L., Nellum, C. J., & Smith, C. D. (2018). From alienation to efficacy: An examination of racial identity and racial academic stereotypes among Black male adolescents. Urban Education53(7), 899-928.

    51. Durlak, J. A., Weissberg, R. P., Dymnicki, A. B., Taylor, R. D., & Schellinger, K. B. (2011). The impact of enhancing students’ social and emotional learning: A meta-analysis of school-based universal interventions. Child Development, 82(1), 405–432.

    52. Deutsch, N. L., & Melton, T. (n.d.). Preventing girls’ delinquency: A longitudinal evaluation of the young women leaders program. Department of Justice, Office of Juvenile Justice and Delinquency Prevention. Department of Justice, Office of Juvenile Justice and Delinquency Prevention.

    53. Smith, C. D., & Hope, E. C. (2020). “We just want to break the stereotype”: Tensions in Black boys’ critical social analysis of their suburban school experiences. Journal of Educational Psychology, 112(3), 551–566.

    54. DuBois, D. L., Portillo, N., Silverthorn, 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(2), 57–91.

    55. Herrera, C. Kauh, T. J., Cooney, S. M., Grossman, J. B., & McMaken, J. (2008). High school students as mentors: Findings from the Big Brother Big Sister school-based mentoring impact Study. Philadelphia, PA: Public/Private Ventures.

    56. Smith, L., Petosa, R. L., & Shoben, A. (2018). Peer mentor versus teacher delivery of a physical activity program on the effects of BMI and daily activity: Protocol of a school-based group randomized controlled trial in Appalachia. BMC Public Health, 18(1), 633.

    57. Brezina, T., Kuperminc, G. P., & Tekin, E. (2016). Future selves, motivational capital, and mentoring toward college: Assessing the impact of an enhanced mentoring program for at-risk youth. U.S. Department of Justice.

    58. DuBois, D. L., Felner, J., Heubach, J., & Mayer, N. (2014). Mentoring for academic success: A pilot study. Final report submitted to Raikes Foundation.

    59. DuBois, D. L., & Keller, T. E. (2017). Investigation of the Integration of Supports for Youth Thriving into a Community-Based Mentoring Program. Child Development, 88(5), 1480–1491.

    60. Good, C., Aronson, J., & Inzlicht, M. (2003). Improving adolescents' standardized test performance: An intervention to reduce the effects of stereotype threat. Journal of Applied Developmental Psychology, 24(6), 645-662.

    61. Lyons, M. D., & McQuillin, S. D. (2019). Risks and rewards of school-based mentoring relationships: A reanalysis of the student mentoring program evaluation. School Psychology Quarterly, 34(1), 76–85.

    62. Gottfredson, D. C., Cook, T. D., Gardner, F. E., Gorman-Smith, D., Howe, G. W., Sandler, I. N., & Zafft, K. M. (2015). Standards of evidence for efficacy, effectiveness, and scale-up research in prevention science: Next generation. Prevention Science, 16(7), 893–926.

    63. Komosa-Hawkins, K. (2010). Best practices in school-based mentoring programs for adolescents. Child & Youth Services, 31(3–4), 121–137.

    64. Rhodes, J. E., & DuBois, D. L. (2006). Understanding and facilitating the youth mentoring movement. Social Policy Report, 20(3), 1–20.

    65. Hamilton, S. F., & Hamilton, M. A. (1990). Linking Up: Final Report on a Mentoring Program for Youth.

    66. Karcher, M. J., & Berger, J. R. M. (2017). One-to-one cross-age peer mentoring (p. 30). MENTOR: The National Mentoring Partnership.

    67. Lyons, M. D., McQuillin, S. D., & Henderson, L. J. (2018). Finding the sweet spot: Investigating the effects of relationship closeness and instrumental activities in school-based mentoring. American Journal of Community Psychology.

    68. Mboka, A. K. (2018). University students’ relationship-based mentoring in school settings. International Journal of Offender Therapy and Comparative Criminology, 62(8), 2271–2291.

    69. Kuperminc, G. P., Chan, W., & Hale, K. (2018). Group mentoring for resilience: Increasing positive development and reducing involvement in the juvenile justice system [(Technical Report No. 252131; Award #2013-JU-FX-0006).]. U.S. Department of Justice, Office of Juvenile Justice and Delinquency Prevention.

    70. Durlak, J. A., & DuPre, E. P. (2008). Implementation matters: A review of research on the influence of implementation on program outcomes and the factors affecting implementation. American Journal of Community Psychology41(3), 327-350.

    71. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta‐analytic review. American Journal of Community Psychology, 30(2), 157-197.

    72. Christensen, K. M., Hagler, M. A., Stams, G. J., Raposa, E. B., Burton, S., & Rhodes, J. E. (2020). Non-specific versus targeted approaches to youth mentoring: A follow-up meta analysis. Journal of Youth and Adolescence, 49, 959–972.

    73. Garringer, M., McQuillin, S., & McDaniel, H. (2017). Examining youth mentoring services across America: Findings from the 2016 Mentoring Program Survey. Boston, MA: MENTOR: The National Mentoring Partnership. Retrieved from

    74. Elledge, L. C., Cavell, T. A., Ogle, N. T., & Newgent, R. A. (2010). School-based mentoring as selective prevention for bullied children: A preliminary test. The Journal of Primary Prevention31(3), 171-187.

    75. Converse, N., & Lignugaris, B. (2009). Evaluation of a school-based mentoring program for at-risk middle school youth. Remedial and Special Education, 30(1), 33–46.

    76.Center for Promise (2015). Don’t quit on me: What young people who left school say about the power of relationships. Washington, DC: America’s Promise Alliance.

You can also find several tools and activity guides that can support group 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 group mentoring program.

Mentoring for Preventing and Reducing Delinquent Behavior Among Youth Research Review
PDF button Facebook button Twitter button

February 2020

This review examines research as it relates to mentoring as a prevention strategy for delinquent behavior. The appeal of mentoring as a delinquent behavior prevention strategy is understandable given its relatively low cost and ability to capitalize on the resources of local communities and caring individuals. The review is organized around four topics:

  1. The overall contributions of mentoring to reducing or preventing delinquent behavior.

  2. Factors that may condition or shape the extent to which mentoring has effects on delinquent behavior.

  3. Processes that may be involved in accounting for the effects of mentoring on delinquent behavior.

  4. The extent to which approaches to mentoring focused on preventing or reducing delinquent behavior have reached intended youth, been implemented with high quality, and been adopted and sustained by settings.

Research on the effectiveness of mentoring for preventing or reducing delinquent behavior found mentoring relationships, both those that are provided through programs and those that are naturally occurring, appear more likely than not to contribute, on net, to lower levels of delinquent behavior. Research suggests, but does not definitively identify, several factors that may influence the effectiveness of mentoring for preventing or reducing delinquent behavior. One possibility is that prior involvement in the courts may lead youth to resist rather than to receive mentors whom they may view as an extension of an unpleasant justice system. However, both “mattering” (defined as being noticed, needed, and an object of concern, as well as the perception of being acknowledged and relevant to others), and strengthening of core indicators of positive development, and “thriving” (e.g., skills for setting and pursuing goals) appear to be processes through which mentoring may be able to help prevent or reduce delinquent behavior. When assessing the reach, quality, and sustainability of mentoring initiatives to prevent or reduce delinquent behavior, there remains work to be done. Mentoring services directed toward preventing or reducing future delinquent behavior and engaging youth in these services have proved only partially successful and, as such, there is a substantial unmet need for mentoring directed toward these goals. Limitations pertaining to the organizational capacity of juvenile justice settings and mentoring programs and their degree of coordination with one another appear to be important barriers.

Insights for practice based on currently available knowledge are appended to this review.

  • Introduction

    Historical and Contemporary Context

    Matching young people with adult mentors is one of the oldest strategies employed in community-based interventions designed to prevent youth problem behaviors and promote positive youth development.1 Early in the twentieth century, concern for the welfare of youth whose behavior brought them into contact with the court system provided much of the impetus for the emergence of what would become the Big Brothers Big Sisters program.2 Roots of youth mentoring also are evident in the juvenile justice system, with the nation’s first juvenile court in Chicago established in 1899 where probation officers were assigned to provide guidance and support to youth who were detained.2 The appeal of mentoring for the prevention of delinquent behavior is understandable given its low cost and ability to capitalize on the resources of local communities and caring individuals. With this background in mind, it is useful to briefly consider the contemporary context of delinquent behavior, particularly as relates to juvenile justice system involvement, and utilization of mentoring as a prevention strategy for these outcomes.

    Since reaching a peak in the mid-1990s, the juvenile arrest rate in the United States has declined fairly steadily to a level in 2017 — 2.4 arrests per 100,000 youth ages 10–17 — that is only one-third of its earlier high watermark.i, 3 From 2008 through 2017, the number of delinquency cases handled by juvenile courts in the United States similarly declined 59 percent. Yet delinquent behavior continues to be a source of concern for a number of reasons. These include the reality that negative encounters with the justice system are not equally distributed in the United States, with disproportionately high rates of incarceration for African-Americans and those with the fewest resources (e.g., the poor).4 Such patterns are especially concerning in view of evidence that involvement in the juvenile justice system itself can have iatrogenic (i.e., unintended harmful) effects, such as increased likelihood of involvement in the penal system in adulthood.5 Engaging in delinquent behavior, furthermore, whether brought to the attention of the juvenile justice system or not, appears to increase youth risk for negative outcomes (e.g., depression), thus potentially increasing likelihood of continued or intensified conduct problems in a vicious cycle.6 It is the recognition that involvement in the juvenile justice system can have a negative impact on a young person’s life that led to the proliferation of diversion programs in recent years. Some diversion programs are pretrial or predisposition, which means that youths are diverted away from juvenile justice system processing from the outset, whereas others begin only after formal adjudication. In these latter cases, youths are diverted away from detention or incarceration, but are still formally processed.7 Mentoring is among the main community-based strategies that have been used in diversion programs.8

    The historical connection between mentoring and juvenile delinquency remains prominent today. Founded in 1974, the Office of Juvenile Justice and Delinquency Prevention (OJJDP) is the chief component of the federal government in the United States charged with enacting programs and other initiatives aimed at decreasing levels of delinquent behavior. Two years after its founding, OJJDP’s Special Emphasis branch provided $10 million in funding for the development of diversion programs.7 In fiscal year 2017, youth mentoring was one of OJJDP’s primary strategies for decreasing delinquency, with $60.7 million awarded in FY 2017 through its Mentoring Opportunities for Youth initiative to support and strengthen youth mentoring programs nationwide. Interestingly, though, only a minority of programs in a recent national survey identified top outcome goals directly related to delinquency prevention (juvenile justice/re-entry 3.65 percent; violence prevention 5.24 percent).9 Many programs may see reductions in delinquent behavior as a more distal and indirect objective rather than as a more immediate or primary point of emphasis. Programs, for example, commonly target factors protecting against delinquent behavior, such as academic enrichment (36.46 percent), life skills/social skills 53.89 percent), and providing a caring adult relationship (44.25 percent).9 Also important to note is that there are large numbers of adults both within the justice system, such as law enforcement officers and staff of juvenile detention facilities, and outside of it in varied contexts, such as youth sports, schools, and after-school programs, who regularly assume more informal mentoring roles in the lives of youth who have already exhibited or are at risk for involvement in delinquent behavior.10,11

    Overview of this Review

    This review takes stock of the research that addresses the potential for mentoring to serve as a strategy for preventing and reducing delinquent behavior (as defined below). The review focuses on the following four questions:

    1. What are the effects of mentoring on delinquent behavior among youth?

    2. What factors condition or shape the effects of mentoring on delinquent behavior?

    3. What intervening processes are most important for linking mentoring to beneficial effects on delinquent behavior?

    4. To what extent have efforts to provide mentoring to youth with preventing or reducing delinquent behavior as a priority outcome reached and engaged the intended youth, been implemented with high quality, and been adopted and sustained by host organizations and settings?

    The scope of the review was limited to mentoring as defined by the National Mentoring Resource Center (i.e., relationships and activities that take place between youth [i.e., mentees] and older or more experienced persons [i.e., mentors] who are acting in a nonprofessional helping capacity — whether through a program or more informally — to provide support that has its aim or realistic potential benefitting one or more areas of the young person’s development; for further detail, see What is Mentoring?). This definition excludes services and supports that are offered in formal professional roles by those with advanced education or training (e.g., social work, counseling) as well as those that are exclusively or predominantly didactic in orientation (e.g., structured curriculum).

    The review’s scope was further limited to studies examining mentoring in relation to delinquent or other clearly serious antisocial behaviorii and/or juvenile justice system involvement (e.g., arrest). Studies that examined effects of mentoring received during childhood or adolescence on relevant adult outcomes such as arrests or incarceration also were included. Studies were excluded, however, if the behaviors assessed would not typically be considered serious (e.g., minor forms of classroom misbehavior). Studies focused on mentoring in relation to substance use or misuse as well as mentoring for youth during reentry from juvenile justice-related confinement also were excluded as separate NMRC reviews address these topics.12,13

    Using a prevention framework,14 the review focuses on studies of mentoring as a means to primary prevention (in this context, preventing delinquent behavior and juvenile justice system involvement before it ever occurs), secondary prevention (in this context, keeping initial, infrequent, and/or relatively less serious delinquent behavior and/or juvenile justice system involvement from becoming recurrent, more extensive, and/or problematic), and tertiary prevention (in this context, reducing the longer-term complications of delinquent behavior and/or juvenile justice system involvement for outcomes later in development, such as adult arrest). Using the above-described criteria, a systematic literature search was conducted to identify journal articles, book chapters, and other types of documents that reported findings pertinent to one or more of the review’s four organizing questions. Search strategies included (a) using a set of relevant keywords to search PubMed, Proquest Dissertations and Theses, PsycINFO, and Google Scholar; (b) outreach to a listserv on youth mentoring research and practice as well as members of the NMRC Research Board; and c) examination of research referenced in broader treatments of mentoring and delinquent behavior, such as chapters on this topic in different editions of the Handbook of Youth Mentoring and a recent synthesis of OJJDP-sponsored research on mentoring. The search was inclusive of both quantitative and qualitative research.

    As a final point of introduction, readers of this review are encouraged to keep in mind that behavior labeled as delinquent or antisocial behavior is best viewed within context. A contextual perspective calls attention to the reality that rates of delinquent behavior, the extent to which they are detected by different measures (e.g., official arrest records), the consequences with which they are associated (e.g., diversion vs. confinement), and the seriousness with which they are viewed within society all can be influenced by a broad range of current and historical influences that are outside a youth's control.

    i The data being referenced are for the period of 1980 through 2017.
    ii OJJDP's definition of delinquent behavior as a performance measure is similarly limited to criminal acts, such as offenses against persons or property. In view of the potential for mentoring to relate differentially to youth involvement in delinquent (i.e., unlawful) and other antisocial behavior, the distinction between the two types of behavior is taken into account throughout the review to the extent possible.

  • What Is the Effectiveness of Mentoring for Preventing or Reducing Delinquent Behavior?


      Sociological theories of delinquent behavior call attention to the reality that some youth, especially those from socioeconomically disadvantaged backgrounds, perceive conventional pathways for achieving widely valued outcomes (e.g., educational attainment as a means of securing monetary success) to not be viable for them.15,16 Such youth then may be more disposed to engage in delinquent behavior both as an alternative means to securing valued life outcomes and for other reasons, such as rebellion against cultural norms or expectations for behavior. From this perspective, mentoring relationships may be useful for both preventing and curbing existing delinquent behavior because they provide youth with a basis for greater hope and optimism that they can achieve conventional goals, such as success in school, and that they matter to society as individuals. Other theories and frameworks suggest additional reasons that mentoring could help youth avoid involvement in delinquent behavior. Blechman and Bopp,17 for example, draw on Host Provocation theory to identify three ways in which mentoring may prevent re-offending: “(a) increase external controls by helping parents and teachers with supervision; (b) strengthen internal controls by promoting attachment bonds, self-regulation, and prosocial values; (c) reduce exposure to antisocial provocations, such as deviant peers, drugs and alcohol, violent mass media, through immersion in prosocial activities” (p. 457).

      Considerable research is in line with the preceding possibilities. There is evidence, for example, that mentoring program participation can stimulate both improved educational expectations18 and hopeful expectations more generally19 as well as gains in direct contributors to educational attainment, such as school attendance and grades.20 Mentoring also has a demonstrated potential to strengthen a range of other well-established protective factors against delinquent behavior. These include both individual and environmental assets, such as self-control,21,22 social competence,23 and stronger relationships with parents and other adults (e.g., teachers).24,25 Related research links mentoring to improvements in the “5 Cs” of positive youth development (i.e., confidence, caring, competence, connectedness, and character)26,27 as well as various indicators of youth thriving, such as having a motivating passion or interest (“spark”) and the skills for effectively setting and pursuing goals,28,29 all of which to varying degrees have been linked to less involvement in problem or delinquent behavior. Taken as a whole, these varied lines of theory and research point toward a potential for mentoring to serve as a source of social capital that increases youths’ perceived and actual prospects for achieving valued life outcomes through conventional pathways, thereby reducing prospects for delinquent behavior.

      The same theory and research, however, also point toward potential limits in the capacity of mentoring, received either through a program or under more naturally occurring circumstances, to prevent or reduce delinquent behavior. Time constraints, boundaries on relationships set by programs, and a range of contextual risk factors (e.g., household poverty, community violence), for example, may hinder the ability of mentors to prevent or ameliorate youth involvement in delinquent behavior. One strategy suggested for countering these types of potential limitations is to design mentoring programs to be more directly responsive to various types of challenges that may be encountered by participating youth in their communities, such as through specialized mentor training and support for collective involvement of mentors and mentees in social-political activism.30,31 It should be noted, however, that only one of the programs evaluated in studies identified for this section of the review32 appears to be a strong exemplar of this type of approach (this study’s findings are most relevant to processes through which mentoring may affect the likelihood of delinquent behavior and thus are discussed in the later section of this review addressing that topic).


      Primary prevention.

      Tolan and colleagues33 used the technique of meta-analysis to synthesize findings from randomized control trials (experimental) as well as quasi-experimentaliii evaluations of the effects of mentoring programs on delinquent behavior that were available through 2011. This review is considered here under primary prevention because it focused on “studies that involved youth who were involved in the mentoring program being evaluated because they were ‘at risk’ for juvenile delinquency” due to either individual (e.g., school failure) or environmental (e.g., living in a neighborhood with a high crime rate) characteristics; youth in the priority populations for programs thus had not necessarily already exhibited delinquent behavior or come into contact with the juvenile justice system (although this clearly is the case by design for some of the programs). Results indicated that, on average, mentoring program participation was of significant benefit for reducing delinquent behavior (25 studies) as assessed by measures which included standardized self- and teacher-report scales as well as arrest and court records. The size of this benefit (referred to as an “effect size” in research studies) was small in magnitude; furthermore, there was substantial variability in findings across studies, a point that will be returned to when considering factors that may condition (moderate) effects of mentoring on delinquent behavior. The review’s findings were sufficiently strong, however, for an independent assessment conducted by to rate mentoring as “effective” for “reducing delinquency outcomes.”

      Other individual studies not included in the review by Tolan and colleagues (e.g., because they have been published only recently) also are relevant to primary prevention of delinquent behavior through mentoring (see Table 1 for information regarding selected examples of these studies). Illustratively, a randomized control trial (RCT) of the Big Brothers Big Sisters (BBBS) community-based mentoring (CBM) program involving approximately 600 youth did not find evidence of an effect on the likelihood of youth report of arrest or juvenile detention over a 13-month period, although benefits in the form of reduction of less serious conduct problems were evident based on parent reports.34 Several studies also have reported associations between naturally occurring mentoring relationships (not considered in the Tolan and colleagues review) and lower levels of delinquent behavior. Although these findings are correlational in nature and thus not definitive evidence of a causal benefit of natural mentoring for delinquent behavior, their relative consistency is nonetheless notable. Directly in line with this research, the National Guard Youth Challenge program includes a youth-nominated mentoring component in which the participating older adolescents (who are designated at risk due to factors such as school dropout, but without any requisite involvement with the justice system) recruit mentors from their existing social networks (excluding immediate family members). As summarized in Table 1, Schwartz and colleagues found that youth-initiated mentoring relationships established and supported through this program that lasted at least 21 months were associated with reduced likelihood of conviction for an offense at a 36-month follow-up.35

      Secondary prevention.

      Another systematic review, published in 2007,36 focused on mentoring in relation to youth re-offending after an initial arrest and thus is most germane to considerations of secondary prevention. The authors concluded that the most methodologically rigorous studies did not suggest that mentoring was responsible for a statistically significant reduction in re-offending. There are, however, a number of rigorous studies published since the time of this review that point to a potential for mentoring to be of value in secondary prevention. These include a RCT evaluation of the Reading for Life diversion program, which found statistically significant declines in rates of rearrest and number of arrests for a two-year period following participation in the program.62 These program impacts were most evident for relatively serious felony offenses compared to misdemeanors. Similarly, an evaluation of the Campus Corps program, which utilizes undergraduate students as mentors for adolescents who for the most part have already come into contact with the juvenile justice system, found that program participants reported significantly less delinquent behavior at a post-test relative to a comparison group of nonparticipants.63 An evaluation of the Arches Transformative Mentoring program provides a final and most recent example.64 This program uses a group mentoring format, interactive journaling curriculum, and mentors who are known as “credible messengers” because they share similar backgrounds of justice system involvement with the youth who are served by the program. The evaluation found evidence that involvement in the program led to significantly reduced felony reconvictions at 12- and 24-month assessments for participating young persons, although comparable benefits were not apparent for other outcomes (e.g., overall arrests).

      Long-term follow-up into adulthood (tertiary prevention).

      As with outcomes in other areas, there appears to be a potential for effects of mentoring programs on delinquent behavior to erode relatively quickly after program participation has ended. In a large-scale RCT of the BBBS school-based mentoring program, for example, the program group experienced reduced likelihood of teacher-reported serious misconduct in school (e.g., suspension) at the end of the school year, although this benefit was no longer evident at a follow-up assessment in the fall of the subsequent school year when the mentoring relationships of most youth in the program group were no longer active.18 This does not augur well for finding rigorous evidence of effects of mentoring on outcomes related to delinquency extending into adulthood. Yet the findings of the small number of studies that have looked at this question suggest that this possibility should not be ruled out (see Table 1). Most notably, a follow-up of the RCT evaluation of the Communities in Schools school-based mentoring program found that those assigned to receive the program were significantly less likely to have been arrested by age 21 based on court records.67 In comparison, a follow-up of the landmark Public/Private Ventures RCT of the BBBS CBM program showed that arrest rates did not differ significantly between those assigned to receive the program and members of the control group.37 It should be noted that a portion of those in the control group did end up receiving BBBS mentoring after the 18-month time frame of the original study, a reality which speaks to the complexities of this type of research.

      iii In quasi-experimental studies, there are groups of individuals who do and do not participate in the intervention being studied (e.g. mentoring program), but the individuals are not assigned randomly (by chance) to one or the other group as in a randomized control trial.


      1. Mentoring relationships, both as provided through programs and naturally occurring, have a well-established potential to contribute to reduced delinquent behavior and juvenile justice system involvement among youth; estimated benefits are typically small in magnitude, however, and are not evident consistently across studies (potentially, as discussed in the next sections of this review, due to factors such as differences in programs and in the background characteristics of participating youth).

      2. Evidence is consistent with potential benefits of mentoring program participation both (a) among those who are at-risk for, but not yet demonstrating, delinquent behavior or having juvenile justice system involvement (primary prevention); and (b) for youth who have already exhibited initial delinquent behavior or having justice system involvement (e.g., reduced likelihood of rearrest; secondary prevention). Again, though, findings for both types of prevention are variable and most consistent with a demonstrated potential for impact rather benefits that are realized on a consistent basis.

      3. The available evidence is not sufficient in scope or consistency to support even a preliminary conclusion regarding typical benefits of mentoring for longer-term outcomes, such as arrest during adulthood (tertiary prevention); it does appear, however, that these may be possible.

  • What Factors Influence the Effectiveness of Mentoring for Preventing or Reducing Delinquent Behavior?


      Understanding the conditions under which mentoring to prevent or reduce delinquency may be more or less effective, or how program effects may vary as a function of unique youth or mentor characteristics, is critical to effectively reaching youth at risk for starting or continuing delinquent activities. It is also useful to consider the way that program goals, curricular content, records of mentoring interactions and activities, as well as the general degree of adult-imposed structure and focus all factor into outcomes, as was revealed in multiple recent studies. The research literature has consistently found youth engaged in or at-risk for later delinquent activity to show less favorable attitudes toward authority and greater general resistance to compliance with adult directives (including, potentially, those of mentors) compared to youth at lower risk; thus, examining these program and contextual factors is important. The degree of imposed structure or centrality of adult-selected goals for programs (e.g., mentoring that explicitly includes social skills training) may be reacted to differently by youth with fewer negative prior experiences with adults, and may pose a challenge to program implementation fidelity, thereby affecting program effectiveness and long-term outcomes.


      Beyond the overall main estimated benefits of youth mentoring program participation on delinquency and aggression for youth at risk for such behaviors reported by Tolan and colleagues,33 their meta-analytic study also identified several moderating factors associated with differences in estimated benefits of mentoring on delinquency influencing how mentoring program participation reduces or prevents later delinquency, misconduct, or criminality. The study did not find benefits to differ according to levels of youth risk, either individual (e.g., behavioral, academic) or environmental (e.g., living in poorly resourced communities). Rather, it found more evidence that both mentors’ characteristics and the program’s interaction focus moderated outcomes. Specifically, greater benefits accrued to youth when mentors’ primary motivation to mentor was for the purpose of their own professional development, and when programs prioritized mentors providing advocacy and emotional support. However, a significant concern of the authors was the absence of information on the actual interactions that took place, noting that what programs espouse is sometimes confounded with youth characteristics and also may not correspond with what actually happens in relationships. Therefore, this section of the review considers how select youth and adult characteristics may moderate program impacts and how each also may be shaped by the nature of program activities.

      Characteristics of mentees.

      In the meta-analysis of Tolan and colleagues neither gender nor mentee risk characteristics moderated program effectiveness; thus, this review does not discuss sex differences in outcomes, even though sex differences appear in multiple studies. For example, when rates of arrest and conviction are used as the outcome variables, the benefits of mentoring typically favor boys in the mentoring condition. It is important to note that boys engage in higher rates of crime than girls overall, making variation in crime between mentees and control groups among boys more likely to be found. In none of these studies were the benefits of mentoring program participation significantly different in nature or direction by sex — the size of the benefits was simply somewhat larger.

      Prior meta-analyses23,20 that did not restrict their focus to programs serving only youth at risk for delinquent behavior (as done by Tolan and colleagues) have found that those at greater individual or environmental risk appeared to benefit more from mentoring. Even though some studies, like the 35-year follow-up of the Buddy System,54 have found those at greater risk (i.e., prior arrests) initially benefitted more, Tolan and colleagues did not find risk to consistently moderate outcomes for this population of youth. However, looking at outcomes in secondary prevention mentoring programs, that target populations of youth involved in the court system or at risk for delinquency, reveals some evidence for the opposite pattern (i.e., that youth at greater risk benefit less from mentoring); namely, when skills-focused programs are delivered through mentoring relationships in order to prevent problem behavior, those entering programs with more developmental assets and who have prior support from adults may benefit the most. It may be that the specific environmental risk of experiencing insufficiently trusting, reliable, and supportive adult relationships at home, in school, or in one’s community prior to program enrollment can make it harder for youth to forge the close bond with mentors needed to collaboratively engage in goal-directed mentoring. Adverse childhood experiences can make youth more resistant to adult authority and, understandably, more reluctant to take the interpersonal risks necessary to engage in trusting relationships with adults that could divert or disrupt a pattern of misbehavior and prevent criminal activity.38

      Youth who present to programs with more indicators of positive youth development (PYD) and of youth thriving, may be best prepared to take advantage of mentors specifically trained to further coach specific developmental skills as a way to reduce susceptibility to later delinquent behavior. Illustratively, one study of the Step-It-Up-2-Thrive approach29 found partial support for a model in which mentor support for youth thriving was linked to reduced problem behavior (a composite measure of delinquent behavior and less serious conduct problems) over a 15-month period, but only for those who entered the program reporting higher levels of prior support from adults. The indirect effect of the goal-focused mentor training and activities operated through increased overall adult support for thriving, which was linked to increased indicators of thriving (e.g., growth mindset, goal pursuit skills) that predicted lower problem behaviors in the Step-It-Up-2-Thrive condition. Although there were no main effects differentiating the Step-It-Up-2-Thrive and the standard BBBS approach at the 15-month follow-up, tests of indirect effect revealed a greater reduction in delinquency and conduct problems among youth in the Step-It-Up-2-Thrive condition, but only among a subsample (38% of the youth) who viewed the Step-It-Up-2-Thrive activities as either helpful or fun. The authors point out that “higher initial youth reported levels of support from adults . . . predicted greater likelihood of positive engagement” (p. 1487), which means only those youth familiar with receipt of adult support were able to take advantage of it from mentors.

      This finding supports a view called the Matthew Effect, which holds that in prevention programming often it is the rich who get richer.39 Those youth who were unfamiliar with the practice of utilizing adults’ support were less able to take advantage of support from mentors and the Step-It-Up-2-Thrive activities. The readiness of mentees to receive and build on informational encounters may be shaped by their prior experiences of support from adults. The finding is similar to that of the landmark RCT of the BBBS CBM study introduced earlier in that it was those with the least supportive and most challenging interpersonal histories who benefited the least from the program,40 it seems, because they were more likely to quit or were less willing to trust the mentor. Another example of this is that mentees with no arrest or prior court involvement benefited more than mentees with prior arrests in the My Life Mentoring program,68 which also used “a purposeful advocacy and teaching approach” (p. 9).

      It is important to note the opposite phenomenon as well, wherein youth who are least at-risk for later delinquency may be negatively influenced by participating in mentoring programs that are intended to benefit those already engaged in delinquent behavior. One of the earliest lessons to be learned about the use of mentoring to prevent delinquent behaviors, comes from the Cambridge-Somerville Youth Study, and specifically the longitudinal research, which was summarized by McCord.41 Long-term negative effects were discovered, which McCord and colleagues linked to participation in one element of the program: repeated participation by some youth in relatively unsupervised summer camps. Despite whatever good might have come from time spent with the caring counselor or mentor-type adult, for those who ended up at the summer camp, their program participation became a breeding ground for deviant behavior (i.e., “deviancy training”), which seemed to secure their position on a trajectory toward later crime.

      The effects of deviancy training in mentoring programs also are found in the 35-year follow-up evaluation of the Buddy System mentoring program.54 For girls not previously engaged in the legal system, the program introduced them to peer groups that seemed to profoundly increase their likelihood of selecting a later partner who was or would become court involved. The researcher expected and found that among those in the study who had prior arrests, the rates of arrest in adulthood were much larger (75 percent) for controls than program participants (55 percent). For women who entered the study without prior arrests, the rates of arrest in adulthood for women in the Buddy System (29 percent) were much higher than those in the control group (10 percent). For them, the authors suspected, the association with more deviant peers in the program altered their peer affiliation practices with negative long-term consequences for who were to later marry or cohabitate with a partner who had an arrest record. It is important to point out, however, that in other studies of mentoring programs that did not target youth at designated risk for later criminality, the group setting did not seem to yield negative long-term effects.

      Mentor characteristics.

      Both the Tolan et al.33 and DuBois et al.20 meta-analyses reveal evidence of the value of teaching and advocacy as program priorities or program approaches. But it is important to keep in mind that meta-analyses of differences in outcomes across programs do not tell us whether being matched with an individual mentor who has teaching experience or who tends to use advocacy as a mentoring approach is more or less valuable to youth. Nor does Tolan et al.’s finding that programs which emphasize providing emotional support reveal whether mentors who are emotionally supportive are more effective than those mentors who are not. Those meta-analyses reveal the value of overall program goals, training, or hiring priorities related to mentors. Knowing whether mentors engaging in advocacy, teaching, or being emotionally supportive is helpful requires match-level analyses of actual processes within rather than between programs.

      One study of mentor characteristics and mentoring interactions within the Youth Advocate Program (YAP), revealed how the same mentor characteristics can contribute in different ways (directly or indirectly; positively or negatively) to a program’s effectiveness in reducing delinquency.66 This study focused on similarity between mentor and mentee in terms of educational experience. YAP as an organization has long held that part of their effectiveness comes from employing advocates as mentors who are racially, culturally, and socioeconomically similar to the youth they serve and their families. Therefore, when working with youth from families with adults tending to have very limited educational attainment or success, the value of hiring advocates with prior teaching experience or advocates with higher levels of educational achievement than those of the families of the youth they serve may come into conflict with the goal of cultural similarity. Yet the program also holds that a primary goal of the mentor is advocating for the youth’s success in school, work, and neighborhood settings, and assisting them in learning how to be successful in these settings. This approach reflects an increasingly argued view that mentoring programs need to help position youth to better navigate important systems, develop skills to overcome challenging structural and systemic conditions, and even become participants in systems to transform these systems. Thus, this study examined the value of hiring advocates who themselves were skilled in these systems. Analyses revealed that both the education level of the advocates and the advocates having prior teaching experience were moderately strong predictors of declines in mentees’ delinquency. But before extrapolating this finding to mean that teacher-type mentors are effective by taking an academic and problem-solving approach with court involved mentees, consider that individual mentor characteristics are not synonymous with actual mentoring interpersonal interactions, which are discussed as mediating processes in the following section. In conclusion, there are some basic takeaways from the consideration of unique characteristics of mentors and mentees that may influence program outcomes, but none stand as an independent factor that is divorced from the nature of the program, which type of youth it recruits, and what its staff hold as its primary mechanism of change.


      1. Increased association with court-involved peers through mentoring program participation can pose barriers to the effective use of mentoring programs to prevent or reduce later problem behaviors, especially among those least at risk to begin with, who otherwise seem most likely to benefit from such programs.

      2. Prior involvement with the court system by youth, and their having negative prior experiences with adults or not having previously felt supported by adults, may lead such youth to resist rather than welcome the support of mentors, thereby lessening mentors’ effectiveness, especially when using activities targeting developmental assets and skills.

      3. 3. Mentors with teaching experience or advanced levels of education may provide unique benefits, but it seems this occurs not through acting like teachers or eschewing play over educational activities, which further underscores the complex interplay that can occur between background characteristics of both mentors and mentees on program outcomes.

  • What Pathways Are Important in Linking Mentoring to the Reductions in or the Prevention of Delinquent Behavior?


      An important question remains in the mentoring field as it relates to mentoring and its relationship to delinquent behavior: To the extent that a mentoring program or relationship prevents or reduces delinquent behavior, what are the processes through which this occurs? Mentoring programs and relationships, in theory, may help set in motion a myriad of psychological and/or behavioral processes that in turn impact youths’ level of involvement in delinquent behavior. Prevention research supports seven key benefits to focusing on the intervening processes that link prevention efforts to outcomes.42 In the case of this research review, intervening processes are the pathways that connect the characteristics or behaviors of the mentor to the intervening process and then the pathway from that intervening process to lower levels of delinquent behavior.

      1. It provides a check on whether mentoring efforts changed any of the intervening processes it was thought to impact.

      2. It can identify successful and unsuccessful mentoring program components.

      3. It can highlight whether or not the measurement used to assess process change is reliable or valid enough to detect changes.

      4. It provides findings on proximal outcomes when the impact of mentoring on delinquent behavior may not be realized within the timeframe of the study.

      5. It increases understanding of the mechanisms underlying changes in the outcome.

      6. It facilitates the test of the theories upon which mentoring programs are based.

      7. It contributes to knowledge of best practices if effective and ineffective components can be identified and disseminated with the mentoring field.

      The delinquency literature is replete with examples of the important role of examining intervening processes (e.g., mediators). A recent study concluded that the relationship between gang affiliation and delinquency in the Gang Resistance Education and Training (G.R.E.A.T.) sample was mediated by particular types of antisocial forms of cognition.43 Similarly, Sampson and Wilson offer the theory that a cognitive landscape that accepts offending is a mediator between neighborhood disadvantage and delinquency.44 Last, there is evidence that perceived personal discrimination may be an important mediator of the effects of concentrated disadvantage and racial isolation on delinquency.45 This possibility suggests supporting youth in coping with experienced discrimination as a potential mechanism through which mentoring could contribute to reduced delinquent behavior.46


      Research with natural mentors have provided the most insight into the intervening processes that link mentoring to the beneficial effects on delinquent behavior. One study found support for improvements in school attachment as a pathway through which school-based natural mentoring relationships may reduce adolescent reports of violence-related behavior.

      Mentoring interactions.

      The Youth Advocate Program (YAP) study that examined the relative benefit of the timing of advocates and youth working together on problem-solving versus friendship-building and playful interactions (using structural models controlling for mentee age, sex, and starting levels of misconduct) found that time spent playing, but only later in the match (months three through four) predicted reductions in delinquency, while problem-focused conversations later in the match predicted increases in misconduct over time.66 Considering that the YAP advocates need to form a connection or bond with their mentees, and that the match will spend up to 20 hours a week together, it is easy to see why an “all work and no play” approach could easily backfire. Indeed, in this study, time spent in play, particularly later in the relationship, was a stronger predictor of declines in misconduct than were teaching, advocacy, and problem-focused conversations.

      But these benefits of playful interactions, however, varied as a function of advocates’ educational background characteristics noted earlier. Later in the match, advocates with teaching experience were less likely to focus on academics and problems which further benefitted youth, whereas advocates with more education were less likely to engage in play and thereby acted in ways that dampened program benefits. So, there were direct benefits of having mentors with these characteristics, but these mentor characteristics also predisposed them to engage interpersonally in both helpful and unhelpful ways, thereby also indirectly affecting outcomes. Although the more educated advocates brought something that was helpful to their youth, it was not a tendency to value playing in their matches. Similarly, mentors with teaching experience also brought something helpful, but it was not an increased tendency to teach (i.e., to take on a problem-focused or teaching role). In this study with court involved youth, after getting to know the youth, it was important to be willing to play.

      The importance of a balance between work and play in preventing misconduct (i.e., later arrest) also was found with youth not explicitly at-risk at the start of the program or court involved, wherein this balance contributed to the experience of feeling known by the mentor. In the 10-year follow-up of the Communities in Schools RCT of school-based mentoring, the effect of the mentoring program on likelihood of later arrest was moderated by specific mentoring activities.67 Data from weekly activity logs revealed that the more time mentors spent learning about the mentee the lower the likelihood of a later arrest. In contrast, having a mentor with a problem-focus was linked to higher likelihood of arrest. In effect, the mentors’ focus on academic problems was itself problematic (controlling for starting levels of misbehavior, grades, attendance, sex, and age), but more so when there had been minimal time spent learning about the mentee and when, perhaps as a result of this, the mentee felt like they didn’t matter to the mentor. Indeed, in matches that spent little to no time getting to know each other, high levels of problem-focused conversations predicted a 15 percent greater likelihood of arrest when mentees felt they mattered little to their mentors. When youth reported feeling they did matter to their mentors, time spent in problem-focused conversations was unrelated to later arrest.

      Changes in youth attributes.

      In line with the research just described, a study of natural mentoring relationships using data from the Add Health Study50 also found support for “mattering”— defined as being noticed, needed, and an object of concern, as well as the perception of being acknowledged and relevant to others — as a process through which natural mentors may help lessen delinquent behavior during the transition to adulthood. There is also research support for a role of the “5 C” indicators of positive youth development (PYD) and indicators of youth thriving, both referenced earlier in this review, as mechanisms of change linking program-supported mentoring through the BBBS CBM program to reduced susceptibility to delinquent behavior.27,29 One of these studies,29 described previously, found support for a model in which mentor support for youth thriving was linked to reduced problem behavior (one of two indicators of which was a measure of delinquent behavior) over a 15-month period via links with increased overall adult support for thriving and, in turn, enhanced thriving (e.g., growth mindset, goal pursuit).

      Social justice–oriented mentoring.

      In a notable example of a program adopting a social justice lens that included a mentoring component, Zimmerman and colleagues reported on the evaluation of the Youth Empowerment Solutions (YES) program.32 As described by the authors, “YES applies empowerment theory to an after-school program for middle school students. YES is an active learning curriculum designed to help youth gain confidence in themselves, think critically about their community, and work with adults to create positive community change” (p. 20). Findings indicated no significant benefits of assignment to participate in YES on a measure of delinquent behavior (or other outcomes). However, the “dose” of YES activities received by youth was linked to lower youth-reported delinquent behavior at post-test; this association involved improvement in psychological empowerment as a go-between (mediator) between YES dosage and delinquent behavior. The concept of psychological empowerment included measures for leadership efficacy, civic efficacy, and self-esteem.


      1. Regardless of the youth’s background, mentoring time spent in playful interactions may help buttress the efforts by mentors to establish the trust needed to help their mentees address problems in their lives.

      2. The ideal timing of mentors’ efforts to get to know the mentee or engage in play may vary depending on the youth’s readiness to receive support from their adult mentors, with some needing it early to communicate that they matter to the mentor whereas other mentees may start ready to see the mentor as someone offering specific important life lessons.

      3. Evidence suggests that mentoring relationships can contribute to lower levels of delinquent behavior among youth, in part, by mentors taking time to get to know their mentees and developing authentic relationships whereby mentees believe their importance in the lives of their mentors.

      4. Although limited in scope, available evidence is consistent with a potential for program-established mentoring relationships with volunteers to contribute to lower levels of delinquent behavior by enhancing personal skills and assets that are important for positive development and thriving and by taking a social justice approach to mentoring.

  • To What Extent Have Mentoring Initiatives to Prevent or Reduce Delinquent Behavior Reached and Engaged Youth, Been Implemented with High Quality, and Been Adopted and Sustained?


      Reach and engagement.

      The prevention framework utilized in this review points to the importance of reaching and engaging both youth who may be at risk for, but not already engaged in, delinquent behavior or have come into contact with the juvenile justice system (primary prevention) and those already exhibiting some of this type of behavior and/or having had some juvenile justice system involvement (secondary prevention). A tertiary prevention perspective also suggests the value of reaching and engaging older youth given the increased potential this may offer for utilizing mentoring to reduce the potential for outcomes such as arrests during the transition to adulthood. Each of these aims could potentially require different strategies, making multipronged efforts at reach and engagement most useful in supporting overall aims of reducing and preventing delinquent behavior. Illustratively, whereas approaches such as diversion may be instrumental as an avenue for reaching youth who come into contact with the juvenile justice system, other strategies oriented toward outreach through youths’ natural ecologies (e.g., schools) may be critical for engaging youth prior to any system involvement. The degree to which youth, once reached, are successfully engaged in receiving mentoring programs and relationships is a further key consideration. Underscoring the importance of this concern, several of the evaluations referenced in this review were only able to uncover evidence of benefits for delinquent behavior when analyses took into account actual levels of mentoring (or related program activities) received29,32,53,47 and, in one instance, also whether youth were positively engaged by the activities involved.29

      Fidelity and quality of implementation.

      A robust body of research underscores the importance of the fidelity and quality of program implementation for the effectiveness of prevention programs,48 including youth mentoring programs.23 In line with these findings, the extent to which mentoring programs focused on reducing or preventing delinquent behavior are able to be implemented successfully is likely to be significant in determining the level of benefits that they confer to participating youth. Improvements in implementation, furthermore, are dependent on a sound understanding of factors that account for variation in levels and quality of program implementation.46 Many of these facilitators and barriers may be tied to challenges and opportunities that are especially likely to be encountered in efforts to use mentoring for purposes of delinquent behavior prevention or reduction. Consider, for example, that community resources often are lower at schools, communities, and neighborhoods where delinquent behavior is more prevalent, and with this come barriers to systematic implementation of programs. Clearly, too, the juvenile justice system may present its own opportunities and challenges as a setting for implementation. Potential opportunities include leveraging existing activities, such as detention and probation, for purposes of mentoring through strategies such as staff training. Possible challenges include limitations in the capacity of juvenile justice entities to effectively implement their own mentoring initiatives as well as a relative lack of coordination and referral linkages with established mentoring programs in the broader community.

      Adopting and sustaining mentoring programs to reduce or prevent delinquent behavior.

      The prospect of achieving a measurable overall positive impact on rates of delinquent behavior and related outcomes (e.g., juvenile justice system involvement) through mentoring programs and strategies logically hinges, in significant part, on the extent to which they are adopted and put into place on a relatively broad scale. This is likely to be the case not only with respect to geography (e.g., urban and rural settings), but also different types of host organizations that are apt to be best positioned for contributing to different types of aims (e.g., primary prevention versus recidivism prevention) as well as reach and engagement of different types of youth (e.g., younger and older, male and female, varying racial and ethnic minority backgrounds). Also important may be the extent to which such programs and initiatives are able to be sustained over time. In line with this possibility, a national survey of mentoring programs49 found that the number of years a program had been in operation was predictive of a greater reported percentage of mentees referred to the program from juvenile justice settings meeting or exceeding the goals set for them. Potential explanations for this finding could relate to the degree to which more established programs are successful with reaching and engaging youth and/or maintaining a high quality of implementation. Thus, although these varying sets of factors have been discussed separately here, the potential for important interactions and interdependencies within and across these areas should be kept in mind.


      Available research includes attention to factors involved with the provision of mentoring to populations of youth who for the most part demonstrate risk factors for delinquent behavior, but have not yet begun to engage to a notable degree in this type of behavior. Other findings address similar considerations of adoption, reach, and implementation for mentoring programs and strategies directed toward youth who have come into contact with the juvenile justice system. Some research, too, has looked at the provision of mentoring for older youth, including those with histories of engaging in delinquent behavior and/or having been involved in the juvenile justice system, with an eye at least in part toward forestalling similar behavior and justice system involvement during adulthood. These results map roughly onto the aims of primary, secondary, and tertiary prevention as used in this review and highlights from each are summarized below. It bears noting, as well, that the long-standing support of OJJDP for mentoring programs directed toward youth involved in or at-risk for delinquent behavior and juvenile justice system involvement (summarized earlier in this review) has quite arguably resulted in greater reach of mentoring to these populations, not to mention potentially enhanced quality of implementation. However, the authors of this review are not aware of any research that has examined these contributions.

      Youth with risk for delinquent behavior and juvenile justice system involvement.

      A national survey of 18- to 21-year-olds undertaken on behalf of MENTOR50 found that among those designated as “at-risk” (several indicators of which are established risk factors for delinquent behavior and/or juvenile justice system involvementv), 37 percent reported that they never had a mentor of any kind while growing up. Notably, the greater the number of risk factors reported, the more likely young persons were to recall a time when they did not have, but wished they had had an adult mentor (43 percent of respondents with two or more risk factors compared to 22 percent with no risk factors).48

      Along with these indications of limited reach of mentoring efforts for youth susceptible to delinquent behavior, available findings also point to challenges with implementing and engaging such youth in program practices oriented toward delinquent behavior prevention. The previously referenced study of the Step-It-Up-2-Thrive model,29 for example, nearly 4 in 10 of the youth assigned to receive the model (38.8 percent) reported positive engagement (doing the activity and finding it fun or helpful) with none of its six core components; likewise, only about one in four (22.4 percent) reported positive engagement with all six. Positive engagement in at least three of the six components, which as noted previously was predictive of less problem behavior at post-test, was found to vary considerably across the 10 BBBS agencies involved in this research, thus suggesting a program or organizational level influence. In line with this possibility, the Mentoring Enhancement Demonstration Program (see Table 1) evaluation reported that implementation for the teaching and advocacy practices studies in this research appeared to be hampered by organizational factors such as staff turnover and lack of adequate opportunities for staff professional development and training. Organizational resource constraints, furthermore, were highlighted as a barrier to sustaining the new practices beyond the time frame of the research. In line with the foregoing findings, a recent synthesis of OJJDP-funded research on youth mentoring (most of which has been focused on youth at-risk for delinquent behavior and juvenile justice system involvement) found “abundant evidence of challenges involved with program implementation.” Salient difficulties included challenges of attempting to introduce (and evaluate) potential enhancements into existing mentoring programs as well as limited success of programs in providing youth with mentoring relationships that were sustained over intended minimum periods of time.

      Youth who have come into contact with the juvenile justice system.

      A fairly recent national survey51 found that about 6 in 10 juvenile justice settings reported provided mentoring to youth internally through their own “embedded” programs or services and/or referred youth to external mentoring programs. Approximately equal numbers reported each of these approaches. Approximately 40 percent of mentoring programs surveyed in the same research reported that at least 10 percent of the youth they served were referred by the juvenile justice system. It is not possible from these numbers to glean the overall extent to which mentoring programs and services are reaching youth who come into contact with the juvenile justice system.

      Further findings from the same research point to considerable challenges with adoption, reach, and implementation of mentoring services for youth with juvenile justice system involvement. Among the justice settings not using or referring youth to mentoring, the most common barrier cited (50.6 percent) was lack of access to mentoring programs. More than one-third (38.6 percent) of juvenile justice settings, furthermore, reported that only between 0 and 25 percent of the youth they referred to outside programs were ultimately matched with a mentor. For their part, mentoring programs most commonly cited lack of mentor availability as a barrier to providing services to referred youth (49.8 percent), although a substantial portion (26.7 percent) also reported that refusal/lack of acceptance of the referral on the part of the youth or family was an issue. At the same time, there is also some evidence that when youth with juvenile justice system involvement are able to be engaged with mentoring, this can be beneficial for their overall engagement with recommended supports. In this research, court-referred adolescent males (77 percent Black) were ten times more likely to remain in the targeted community-based intervention if they utilized the mentoring component during the initial six months.69

      v At-risk youth were defined as those who were disconnected (out of school and out of work) and/or one or more of seven risk factors. The latter factors included engaging in delinquent behavior (as indexed by a report of getting into trouble with the law; 13 percent of the overall sample) as well as other factors associated with greater delinquent behavior (e.g., incarcerated parent, poor school performance).


      1. Mentoring has considerable reach as an intervention strategy both for youth susceptible to delinquent behavior and those who have already come into contact with the justice system for offenses; yet there is also clearly a substantial unmet need for mentoring within both of these populations that remains to be addressed, part of which appears to be attributable to limited coordination or collaboration between juvenile justice settings and mentoring programs.

      2. Implementing mentoring services directed toward preventing or reducing future delinquent behavior and engaging youth in these services has proved only partially successful; limitations in organizational capacity appear to be among the most influential barriers to implementation and engagement both within and outside of the juvenile justice system.

      3. There is limited but intriguing evidence that efforts to engage justice system–involved youth in mentoring, when successful, can help to sustain their overall engagement with court-recommended services.

  • Implications for Practice

    (Mike Garringer, MENTOR: The National Mentoring Partnership)

    It seems clear from the research evidence reviewed in the previous pages that volunteer mentors can be effective in both the initial prevention of delinquent and criminal behavior by youth and in efforts to prevent recidivism and deeper involvement for those who may already have some involvement. But this body of research also hints at some factors that programs doing work in this space may want to consider to maximize their results.


      There are several hints in the research described previously that serving delinquent youth can be challenging and that these youth can often bring mindsets, trauma histories, and other personal circumstances to the mentoring relationship that can make it hard for them to engage effectively with a mentor. As noted in the previous pages, these youth may have a history of abuse or exploitation by adults that leaves them not only questioning (or rejecting) adult authority, but also adult offers of support (or even claims of supportive intent). These are young people who may have been previously let down by many adults in their lives, including adults who promised to keep them safe, and may reject new offers of support due to such offers going unfulfilled in the past. It is also noted that these youth may often see mentors referred to them by a court as being representative of a justice system that, in their estimation, is questionable in moral intent if not exploitative of or adversarial to their community. And given the co-incidence of delinquency with other concerns, such as mental health challenges, which can inhibit the development of a good mentoring relationship, programs may wish to take some time at the referral stage to learn more about the attitudes, mindsets, and emotional barriers that may get in the way of some youth benefitting from the program. Those youth can be referred to more appropriate services and can always come back to mentoring at a future point where they would be more amenable to the experience. Trying to serve youth who simply don’t want to or can’t be served by a volunteer mentoring may explain, in part, some of the variance in findings noted in the main review.

      The review also hints, however, at some strengths that these same youth can bring to the table that might also influence who is accepted into the program. Several of the studies mentioned note that youth who had access to other caring adults or more indicators of positive youth development in their personal histories and current circumstances seemed more likely to benefit from mentoring. The review even notes the “Matthew Effect,” in which services like mentoring mostly help those who are in a position to be helped by way of some notable strengths lacking in other potential recipients of the service. And while it’s true that this can result in “the rich getting richer” and leaving lower-asset peers behind, the reality is that not every program is for every young person. If high percentages of a programs’ participants are a poor fit for what a mentor offers — either because of negative attitudes that will be a barrier or a lack of other supportive adults that can reinforce the work with the mentor — then programs are more than justified in applying more rigorous criteria to who they serve. It certainly beats failing in the mission. Thus, programs working in the delinquency space are encouraged to prioritize serving youth who seem open to a relationship and have some level of additional adult support and life stability to build on. Youth who don’t fit this profile can be referred to other services that can be effective with their needs and what they will respond to. Which begs a similar, yet related point


      In 2018, the National Mentoring Resource Center’s Reflections on Research podcast featured an interview with noted criminologist Dr. Edward Latessa. Latessa had just finished a large evaluation of mentoring for youth on parole and probation in the state of Ohio that had referred youth to local mentoring programs with the aim of reducing their recidivism and ongoing engagement in the juvenile justice system. Unfortunately, the evaluation showed no positive impact on recidivism rates at any of the sites compared to control groups, and even though these youth were deeper in their delinquent and criminal involvement than the youth discussed here in this review, it did beg the question as to whether community-based mentoring was an effective approach for preventing recidivism in general and cast some doubt on its ability to prevent worsening of behavior and consequences.

      But Latessa noted that mentoring programs do something very well that almost all other juvenile justice intervention struggle with mightily: the caring relationship. What they didn’t do so well, in his opinion, was integrate their services into an established offender rehabilitation framework — in particular, the Risk-Need-Responsivity (RNR) model. This model, which has substantial research evidencevi supporting its effectiveness in guiding appropriate interventions that reduce recidivism, is built around three core principles:

      1. Risk principle: Match the level of service to the offender’s risk to re-offend (the more likely to re-offend, the more the service is needed and with more intensity).

      2. Need principle: Assess criminogenic needs and target them in treatment (in other words, what are the real-life reasons that might lead to criminal behavior).

      3. Responsivity principle: Maximize the offender’s ability to learn from a rehabilitative intervention by providing cognitive behavioral treatment and tailoring the intervention to the learning style, motivation, abilities, and strengths of the offender (essentially meeting each person’s needs through targeted change talk and in accordance with their unique learning styles).

      Latessa pointed out that the mentoring programs in his study did not match the level of service (e.g., the amount or frequency of mentoring) to the likelihood of re-offending, nor did the mentoring tend to incorporate cognitive behavioral principles by which trained mentors would use specific activities and conversations to try and influence thinking patterns and decision-making of mentees. And probably most discouraging is that the mentoring was often provided in isolation, with not nearly enough attention paid to those criminogenic factors that might lead someone to reoffend even if they were getting great mentoring. These programs, to Latessa’s point, put all their eggs in the relationship basket and ignored much of what we know about preventing worsening delinquent and criminal behavior through programmatic services.

      This is not to say that mentoring programs looking to prevent or reduce delinquency and criminality in young people need to start viewing mentees as serious offenders and responding accordingly — remember, the youth in the Ohio study were much further along in their behaviors and system involvement than the youth being discussed in this review. But it does mean that if mentoring programs, especially those working with youth who are already minimally engaged in criminal behavior, want to influence recidivism rates they may want to pay attention to frameworks like the RNR and ask how much their services align with what we know about working with youth in the juvenile justice system. If neither their mentors nor the other organizations working with the youth are attending to the things outlined in that framework, success may be limited.

      For what it’s worth, the RNR model has its critics, and alternative approaches, such as the “Good Lives Model,” or GLM, have grown in prominence.vii The GLM model is more strengths-based than the deficit-driven RNR and argues that the best way of supporting individuals in avoiding criminality is to build up strengths and start with the positives that all people possess. So there are several frameworks that may influence the mentoring provided delinquent youth and some are likely to be a better fit than others. What is clear is that ignoring these research-supported frameworks to offer only a generic “wise friend” type mentoring relationship might ignore critical components of helping that young person.

      Programs are encouraged, especially in cases where youth are being referred to their services in lieu of deeper court or justice system involvement, to ask if there has been any formal assessment of the youth’s needs or likelihood for re-offending. Models like the RNR start with a very comprehensive assessment that tries to determine the youth’s risk, needs, and responsiveness keys. Those types of assessments can be a gold mine of information for programs and service providers in how to best meet the needs of a young person.

      vi Andrews, D. A., & Bonta, J. (2006). The psychology of criminal conduct (4th ed.). Newark, NJ: LexisNexis.
      vii Ward, T., Mann, R., & Gannon, T. (2007). The good lives model of offender rehabilitation: Clinical implications. Aggression and Violent Behavior, 12, 87–107.


      As noted above, one of the big factors in how well mentoring works for delinquent youth is their openness to the experience and how the relationship feels from their perspective. If the mentor is viewed as a liaison of an unpleasant justice system, or as someone who will break promises, it can be almost impossible to reach a reticent youth. The research reviewed here pointed to several examples of how youth can reject such offers of support. But there are certain approaches mentors can take that might be more effective than others. The YAP study noted in the evidence review offers an interesting example of this. That study analyzed the amount of time matches spent on either problem-focused, “serious” conversations (e.g., adherence to the court-mandated treatment plan) or on more relational and playful activities (e.g., shooting hoops, cooking a meal, talking about fun topics, etc.). Matches that engaged in more fun, playful activities produced better results with youth misconduct, but only when that play was most prominent toward the end of the match. Those who engaged in play early on, likely in an attempt to build rapport and trust, and saved those tougher conversations for later were associated with increased misconduct.

      Mentors are almost always told to spend some time learning about their mentee and building trust through more playful interactions up front, and those are still important steps in most mentoring relationships. And the mentors here who saved play for later in the relationship still did spend some time getting to know their mentees and having a laugh at the beginning of the relationship. But what they didn’t do was ignore the reason for their time together: helping that youth adhere to their court-mandated treatment plan and getting them on a positive track which would end their engagement in misbehavior. They had serious work to do together and they made sure that their mentee knew it. These mentors dove into the task at hand (and one can imagine them explicitly looking to improve those criminogenic factors), using the hard work of the program, rather than play, as the context for getting to know each other. Over time, these mentors sprinkled in more and more “play” activities to the point that by the end of their relationships, they were maintaining what they had built by checking in, but were no longer doing the messy hard work of figuring out paths forward for the mentee. This allowed them to focus on fun and play in a way that solidified their bond and kept them from rehashing the same areas for improvement. Unfortunately, some mentors never increased that play ratio and the evaluation results suggest that this led to some negative perceptions of the relationship. One can see how talking about the same old challenges and problems after so many months together would make mentees perhaps feel worse about themselves and the relationship. These mentors were less effective in keeping their mentees from further trouble.

      One doesn’t automatically think of “fun activities” when conceptualizing a program for delinquent youth. But studies like YAP are a good reminder that mentoring relationships, even ones with a serious criminal justice reason for being and related goals, are still human relationships. And the second they become more of a drag than a value to the recipients, the more likely they are to start failing. Mentors in these programs may overcome those rejection tendencies and earn mentee trust by getting down to business and saving the bulk of the fun times together for the later stages of the relationship.


      As noted in the review and earlier in this section, juvenile justice-focused mentoring programs can often struggle with getting youth buy-in. One of the main barriers in this regard is that youth from seriously disadvantaged communities may feel a sense of hopelessness about their situations and that their communities critically lack resources that can help them avoid delinquency (e.g., adequate employment opportunities, safe public spaces, strong educational institutions, stable housing, etc.). In fact, it’s hard to argue that youth in many communities in this country are incorrect in noticing the uphill battle in front of them due to a lack of resources and supports.

      But a new strain of thought within the mentoring movement has started to shift this dynamic and gives young people a reason to embrace the role of a mentor in their lives. Critical mentoring, a term coined by Dr. Torie Weiston-Serdan, is a form of mentoring that emphasizes youth empowerment and the development of critical consciousness — a deep understanding about the history of a community in relation to one’s self, and ongoing analysis of the many influences, both positive and negative, that have led to the current circumstances a person finds themselves in. In this form of mentoring, youth are encouraged to not only grow as individuals and strive for personal achievement, but are also taught to view the world through a critical lens and are, in turn, empowered to make meaningful changes in their community through the help of mentors and other caring adults. Rather than teaching youth to succeed in spite of their circumstances, this critical mentoring approach empowers them to identify community issues and find solutions that change those circumstances for future generations. In this way, youth are empowered to not only improve themselves through mentoring, but also improve their communities and start addressing root causes in service of future generations.

      This critical mentoring concept exists within a larger general movement within mentoring toward embracing diversity and inclusion and ensuring that mentoring is working in culturally relevant and respectful ways with communities of color and communities of poverty. Unfortunately, in many of these communities youth mentoring programs are often seen, fairly or unfairly, as “helicoptering” in to “save” youth, with an assumption that the community itself lacks assets or motivation. While it’s true that mentoring programs bring new and different social capital into the lives of youth who might never have gained it otherwise, the programs these days are paying much more attention to how they build on strengths and how they build up the whole community, not just individual youth.

      One can see how this approach might be appealing to a young person who has assumed that they have limited opportunities to make meaningful, lasting change in the world. Critical approaches to mentoring have the ability to inspire young people to not only work on their own issues, but to then translate their personal growth into growth for their community. Programs looking to disrupt cycles of delinquency might do well to incorporate these principles and make the case to youth that they have the ability to not only change themselves but also change the reality of their community and leave a better space for future young people to thrive.


      As noted in the background of this review, programs working in this space have a history of being all over the place with regards to the measures and metrics they use, with little consistency in terms of how they define and measure delinquency or its consequences. Even within a simple idea like “criminal involvement” there are myriad ways of measuring youth behavior and program impact. Is the program focused on primary prevention of keeping youth from ever having criminal involvement? Or is the emphasis keeping that involvement from getting worse? And how is “involvement” defined? Being arrested? Being tried? Convicted? For that matter, how is “worse” defined? Is the goal no re-offending at all? Re-offending less? Not committing more serious offenses (e.g., moving from misdemeanors to felonies)?

      The answers to some of these questions have implications for how services are designed and who is recruited as mentors. But they also have implications for what data is collected and how success is measured. One NMRC Program Review highlighted an evaluation of a program focused on recidivism prevention (Reading for Life, noted in the evidence review here) that did a great job of talking about how the program settled on the delinquency outcomes that made sense for them. This was especially tricky given the many, many statuses that youth could be tagged with in the justice system. Practitioners may want to look at the data collection and measures section of the evaluation report as well as the discussion of this in the Insights for Practitioners developed by the NMRC. Asking and discussing questions such as these can also help programs promise the right delinquency and recidivism outcomes:

      • How do we define delinquency? What behaviors are contained within that definition? What about juvenile justice “involvement” specifically? What is contained within that involvement?

      • Are our mentors and services better positioned to focus on preventing the start of delinquent behaviors or stepping in after those behaviors have started?

      • How would we define recidivism for our program? Would we want to keep youth from...
        • Any ongoing law enforcement contact?

        • Being detained but not arrested? Arrested? Arrested but not charged? Charged? Charged but not convicted? Convicted only? What is the specific status we want to prevent?

      • What about status offences or a warrant being issued for failure to check in with probation or following a court order? Do those count?

      • Does it matter if the reoffending is for lesser crimes? For example, if our youth have some criminal involvement, but it’s for minor things and misdemeanors rather than felonies or violent crimes, is that “success”? Or is our goal the elimination of all delinquent behavior? What’s realistic for our youth and our services?

      • What kind of time frame would we consider “success”? How long after the mentoring services start or end would we hope to see ongoing reductions in misbehavior?

      • Do we have access to the data we need through courts or other systems or would we need to rely on self-reports? If not, how can we get it? Does it come in formats that align with the decisions made in the questions above?

      These questions will have profound influence on what data programs collect and how they define success. There are no right answers, but practitioners should simply note that evaluating delinquency outcomes is often complicated and involves nuances such as these that can influence whether a program can claim to have “worked” or not.

  • References

    1. Blakeslee, J. E., & Keller, T. E. (2012). Building the youth mentoring knowledge base: Publishing trends and co-authorship networks. Journal of Community Psychology, 40, 845–859. DOI: 10.1002/jcop.21494

    2. Baker, D., & Maguire, C. P. (2005). Mentoring in historical perspective. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 14–29). Thousand Oaks, CA: SAGE.

    3. Office of Juvenile Justice and Delinquency Prevention (2018, October). OJJDP Statistical Briefing Book. Online. Retrieved from

    4. Western, B., & Pettit, B. (2010). Incarceration & social inequality. Daedalus, 139 (3), 8–19.

    5. Gatti, U., Tremblay, R. E., & Vitaro, F. (2009). Iatrogenic effect of juvenile justice. The Journal of Child Psychology and Psychiatry, 50, 991–998.

    6. Beyers, J. M., & Loeber, R. (2003). Untangling developmental relations between depressed mood and delinquency in male adolescents. Journal of Abnormal Child Psychology, 31, 247–266. DOI: 0091-0627/03/0600-0247/0

    7. Models for Change Juvenile Diversion Workgroup. (2011). Juvenile diversion guidebook. Models for Change: Systems Reform in Juvenile Justice. Retrieved from

    8. Cocozza, J. J., Veysey, B. M., Chapin, D. A., Dembo, R., Walters, W., & Farina, S. (2005). Diversion from the juvenile justice system: The Miami-Dade Juvenile Assessment Center Post-Arrest Diversion Program. Substance Use and Misuse 40, 935–51.

    9. Garringer, M., McQuillin, S., McDaniel, H. (2017) Examining youth mentoring services across America: Findings from the 2016 National Mentoring Program Survey. Technical Report produced by MENTOR: The National Mentoring Partnership. Retrieved from

    10. Chan W. Y., & Henry D. B. (2014). Juvenile offenders. In D. L. DuBois and M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp. 315–324). Thousand Oaks, CA: SAGE.

    11. Hirsch, B., Deutsch, N. L., & DuBois, D. (2011). After-school centers and youth development: Case studies of success and failure. New York: Cambridge University Press.

    12. Eddy, J. M., & Schumer, J. (2016). Mentoring for youth and young adults during reentry from confinement. National Mentoring Resource Center Research Review. Accessible at:

    13. Kaufman, M., & Erdem, G. (forthcoming). Mentoring for preventing and reducing substance use. National Mentoring Resource Center Research Review.

    14. Robinson, W. L., Brown, M., Beasley, C. R., & Jason, L. A. (2017). Advancing prevention intervention from theory to application: Challenges and contributions of community psychology. In M. A. Bond, I. Serrano-García, & C. B. Keys (Eds.), APA handbook of community psychology (Vol. 2, pp. 193–213). Washington, DC: American Psychological Association. doi:10.1037/ 14954-012

    15. Hagan. J. (2001). Theories of juvenile delinquency. In E. F. Borgatta (Ed.-in-Chief) and R. J. V. Montgomery (Managing Ed.), Encyclopedia of sociology (Vol. 3, 2nd ed., pp. 1493–1499).

    16. Jessor, R., & Jessor, S. L. (1977). Problem behavior and psychological development: A longitudinal study of youth. New York: Academic Press.

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

    18. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Making a difference in schools: The Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures.

    19. Bowers, E. P., Geldhof, G. J., Schmid, K. L., Napolitano, C. M., Minor, K., & Lerner, J. V. (2012). Relationships with important nonparental adults and positive youth development: An examination of youth self-regulatory strengths as mediators. Research in Human Development, 9, 298–316.

    20. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, 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. doi:10.1177/1529100611414806

    21. Aseltine, R. H., Dupre, M., & Lamlein, P. (2000). Mentoring as a drug prevention strategy: An evaluation of Across Ages. Adolescent and Family Health, 1, 11–20.

    22. Kogan, S. M., Brody, G. H., & Chen, Y. F. (2011). Natural mentoring processes deter externalizing problems among rural African American emerging adults: A prospective analysis. American Journal of Community Psychology, 48, 272–283.

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

    24. Grossman, J. B., & Tierney, J. P. (1998). Does mentoring work? An impact study of the Big Brothers Big Sisters program. Evaluation Review, 22, 403–426.

    25. DuBois, D. L., Neville, H. A., Parra, G. R., & Pugh-Lilly, A. O. (2002). Testing a new model of mentoring. New Directions for Youth Development, 93, 21–57.

    26. Lerner, R. M., Napolitano, C. M., Boyd, M. J., Mueller, M. K., & Callina, K. S. (2014).  Mentoring and positive youth development. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp. 17–27). Los Angeles, CA: SAGE.

    27. Erdem, G., DuBois, D. L., Larose, S., DeWit, D. J., & Lipman, E. L. (2016). Mentoring relationships, positive development, and emotional and behavioral problems among youth: Investigation of a mediational model. Journal of Community Psychology, 44, 464–483. doi:10.1002/jcop.2178

    28. Bowers, E. P., Wang, J., Tirrell, J. M., & Lerner, R. M. (2016). A cross-lagged model of the development of mentor-mentee relationships and intentional self regulation in adolescence. Journal of Community Psychology, 44, 118–138.

    29. DuBois, D. L., & Keller, T. E. (2017). Investigation of the integration of supports for youth thriving into a community-based mentoring program. Child Development, 88, 1480–1491. doi:10.1111/cdev.12887

    30. Albright, J. N., Hurd, N. M., & Hussain, S. B. (2017). Applying a social justice lens to youth mentoring: A review of the literature and recommendations for practice. American Journal of Community Psychology, 59, 363–381. doi:10.1002/ajcp.1214

    31. Weiston-Serdan, T. (2017). Critical mentoring: A practical guide. Sterling, VA: Stylus.

    32. Zimmerman, M. A., Eisman, A. B., Reischl, T. M., Morrel-Samuels, S., Stoddard, S., Miller, A. L., . . . Rupp, L. (2018). Youth Empowerment Solutions: Evaluation of an after-school program to engage middle school students in community change. Health Education & Behavior, 45, 20–31. doi:10.1177/10901981177104

    33. 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. doi:10.1007/s11292-013-9181-4

    34. Herrera, C., DuBois, D. L., & Grossman, J. B. (2013). The role of risk: Mentoring experiences and outcomes for youth with varying risk profiles. New York, NY: A Public/Private Ventures project published by MDRC. Retrieved from

    35. Schwartz, S. E., Rhodes, J. E., Spencer, R., & Grossman, J. B. (2013). Youth initiated mentoring: Investigating a new approach to working with vulnerable adolescents. American Journal of Community Psychology, 52, 155–169. doi:10.1007/s10464-013-9585-3

    36. Jolliffe, D., & Farrington, D. P. (2007). A rapid evidence assessment of the impact of mentoring on re-offending: A summary. Cambridge, UK: Cambridge University (Home Office Online Report 11/07). Retrieved from

    37. DuBois, D. L., Herrera, C., & Rivera, J. (2018). Investigation of long-term effects of the Big Brothers Big Sisters community-based mentoring program. Final Technical Report for Office of Juvenile Justice and Delinquency Prevention (NCJ 251521). Retrieved from

    38. Baglivio, M. T., Wolff, K. T., Piquero, A. R., & Epps, N. (2015). The relationship between adverse childhood experiences (ACE) and juvenile offending trajectories in a juvenile offender sample. Journal of Criminal Justice, 43(3), 229–241. doi:10.1016/j.jcrimjus.2015.04.012

    39. Ceci, S. J., & Papierno, P. B. (2005). The rhetoric and reality of gap closing: When the "have-nots" gain but the "haves" gain even more. American Psychologist, 60, 149–160.

    40. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring programs. American Journal of Community Psychology, 30, 199–219.

    41. McCord, J. (2003). Cures that harm: Unanticipated outcomes of crime prevention programs. The Annals of the American Academy of Political and Social Science, 587, 16–30.

    42. MacKinnon D.P. (1994). Analysis of mediating variables in prevention intervention studies. In A. Cazares & L. A. Beatty (Eds.), Scientific methods for prevention intervention research (pp. 127–153). NIDA Research Monograph 139. Washington, DC: U.S. Dept. Health Human Services.

    43. Walters, G. D. (2019). Mediating the gang–delinquency relationship with proactive criminal thinking. Criminal Justice and Behavior, 46, 1044–1062. doi:10.1177/0093854819831741

    44. Sampson, R. J, & Wilson, W. J. (1995. Toward a theory of race, crime, and urban inequality. In J. Hagan & R. D. Peterson (Eds.), Crime and inequality (pp. 37­–56). Stanford, CA: Stanford University Press.

    45. Martin, M. J., McCarthy, B., Conger, R. D., Gibbons, F. X., Simons, R. L., Cutrona, C. E., & Brody, G. H. The enduring significance of racism: Discrimination and delinquency among Black American youth. Journal of Research on Adolescence, 21, 662–676. doi:10.1111/j.1532-7795.2010.00699.x

    46. Sánchez, B., Colón-Torres, Y., Feuer, R., Roundfield, K. E., & Berardi, L. (2014). Race, ethnicity, and culture in mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp.145–158). Thousand Oaks, CA: SAGE.

    47. Jarjoura, G. R., Tanyu, M., Altschuler, D., Forbush, J., Herrera, C., & Keller, T. E. (2018). Evaluation of the Mentoring Enhancement Demonstration Program: Technical Report. Washington, DC: American Institutes for Research.

    48. Durlak, J. A., & DuPre, E. P. (2008). Implementation matters: A review of research on the influence of implementation on program outcomes and the factors affecting implementation. American Journal of Community Psychology, 41, 327–350. DOI: 10.1007/s10464-008-9165-0

    49. Miller, J. M., Barnes, J. C., Miller, H. V., & McKinnon, L. (2013). Exploring the link between mentoring program structure and success rates: Results from a national survey. American Journal of Criminal Justice, 38, 439–456.

    50. Bruce, M., & Bridgeland, J. (2014). The mentoring effect: Young people's perspectives on the outcomes and availability of mentoring. A report for MENTOR: The National Mentoring Partnership. Retrieved from

    51. Miller, J. M., Miller, H. V., Barnes, J. C.,  Clark, P. A., Jones, M. A., Quiros, R. J., Peterson, S. B. (2012). Researching the referral stage of youth mentoring in six juvenile justice settings An exploratory analysis. Final report to Office of Juvenile Justice and Delinquency Prevention. Retrieved from

    52. Kelley, M. S., & Lee, M. J. (2018). When natural mentors matter: Unraveling the relationship with delinquency. Children and Youth Services Review91, 319–328.

    53. Black, D. S., Grenard, J. L., Sussman, S., & Rohrbach, L. A. (2010). The influence of school-based natural mentoring relationships on school attachment and subsequent adolescent risk behaviors. Health Education Research, 25, 892–902.

    54. Bodin, M., & Leifman, H. (2011). A randomized effectiveness trial of an adult-to-youth mentoring program in Sweden. Addiction Research & Theory19, 438–447.

    55. Deutsch, N. L., & Melton, T. N. (2013). Preventing girls’ delinquency: A longitudinal evaluation of the Young Women Leaders Program. Technical report submitted to Office of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice.

    56. O’Donnell, C. R., & Williams, I. L. (2013). The Buddy System: A 35-year follow-up of criminal offenses. Clinical Psychological Science, 1, 54–66.

    57. Weiler, L. M., Haddock, S. A., Zimmerman, T. S., Henry, K. L., Krafchick, J. L., & Youngblade, L. M. (2015). Time-limited, structured youth mentoring and adolescent problem behaviors. Applied Developmental Science, 19, 196–205.

    58. Kuperminc, G. P., Chan, W. Y., & Hale, K. E. (2018). Group mentoring for resilience: Increasing positive development and reducing involvement in the juvenile justice system. (Report No. 252131). Retrieved from the U.S. Department of Justice website:

    59. Theodos, B., Pergamit, M. R., Derian, A., Edelstein, S. & Stolte, A. (2016). Solutions for youth: An evaluation of the Latin American Youth Center’s Promotor Pathway program. Washington, DC: Urban Institute.

    60. Berry, V., Little, M., Axford, N., & Cusick, G. R. (2009). An evaluation of Youth at Risk's coaching for communities programme. The Howard Journal of Criminal Justice48(1), 60–75.

    61. Braga, A. A. (2008). Pulling levers focused deterrence strategies and the prevention of gun homicide. Journal of Criminal Justice36, 332–343.

    62. White, M. D., Fyfe, J. J., Campbell, S. P., & Goldkamp, J. S. (2003). The police role in preventing homicide: Considering the impact of problem-oriented policing on the prevalence of murder. Journal of Research in Crime and Delinquency40, 194–225.

    63. Gies, S. V., Cohen, M. I., Edberg, M., Bobnis, A., Spinney, E., & Berger, E. (2015). The Girls Circle: An evaluation of a structured support group program for girls. Retrieved from the U.S. Department of Justice website:

    64. Seroczynski, A. D., Evans, W. N., Jobst, A. D., Horvath, L., & Carozza, G. (2016). Reading for Life and adolescent rearrest: Evaluating a unique juvenile diversion program. Journal of Policy Analysis and Management35, 662–682.

    65. Haddock, S. A., Zimmerman, T. S., Thomas, A. G., Weiler, L. M., Krafchick, J., & Fredrickson, G. J. (2017). A qualitative analysis of mentee experiences in a campus-based mentoring program. Journal of Youth Development, 12(4), 61–80.

    66. Lynch, M., Astone, N. M., Collazos, J., Lipman, M., & Esthappan, S. (2018). Arches Transformative Mentoring Program. Washington, DC: Urban Institute.

    67. Duriez, S. A., Sullivan, C., Sullivan, C. J., Manchak, S. M., & Latessa, E. J. (2017). Mentoring best practices research: Effectiveness of juvenile mentoring programs on recidivism. Cincinnati: University of Cincinnati.

    68. Karcher, M. J. & Johnson, D.J. (2016). Final Technical Report: An evaluation of advocacy-based mentoring as a treatment intervention for chronic delinquency (OJJDP award 2011-JU-FX-0001). Office of Justice Programs’ National Criminal Justice Reference Service, Report #250454.

    69. Brezina, T., Kuperminc, G., & Tekin, E. (2016). Future Selves, Motivational Capital, and Mentoring Toward College: Assessing the Impact of an Enhanced Mentoring Program for At-Risk Youth. Final Technical Report.(NCJRS# 250499). Atlanta, GA: Georgia State University.

    70. Karcher, M. J. (2019). Ten-year follow-up on the RCT Study of Mentoring in the Learning Environment (SMILE): Effects of the Communities in Schools mentoring program on crime and educational persistence. Technical report submitted to the Office of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice.

    71. Blakeslee, J. E., & Keller, T. E. (2018). Extending a randomized trial of the My Life Mentoring Model for youth in foster care to evaluate long-term effects on offending in young adulthood. U.S. Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs’ National Criminal Justice Reference Service, Report #250454.

    72. 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 Journal21, 573–592.

    73. Weinrath, M. & Donatelli, G. & Murchison, M. J. (2016). Mentorship: A missing piece to manage juvenile intensive supervision programs and youth gangs? Canadian Journal of Criminology and Criminal Justice 58, 291–321.

  • 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.

Mentoring for Preventing and Reducing Substance Use and Associated Risks Among Youth Outcome Review
PDF button Facebook button Twitter button

January 2020

This review examines research on youth mentoring as a strategy for preventing and reducing adolescent substance use, including opioids. The review is organized around four questions:

  1. What are the effects of mentoring on substance use and associated risks to personal health and well-being among youth?

  2. What factors condition or shape the effects of mentoring on substance use and associated risks for youth?

  3. What intervening processes are most important for linking mentoring to beneficial effects on substance use and associated risks for youth?

  4. To what extent have efforts to provide mentoring to youth with substance use prevention and intervention as priority outcomes reached and engaged the intended youth, been implemented with high quality, and been adopted and sustained by host organizations and settings?

Overall, there were few studies that focused primarily on the impact of youth mentoring on adolescent substance use prevention, and the studies mostly followed either a primary prevention (addressing problems before they occur; targeted to a broad population of youth) or a secondary prevention framework (focusing efforts on at-risk youth). The review found that the studies assessed more commonly used substances (e.g., alcohol and marijuana), with less attention paid to the impact of mentoring on preventing the initiation of hard drug use, including opiates. Therefore, the limited evidence that is available shows tentative promise for mentoring to have a positive effect on the prevention and reduction of substance use among youth. The review identified only two studies that utilized mentoring as an add-on intervention to an evidence-based substance abuse treatment. This is possibly due to the limited focus on the role of mentoring in tertiary prevention efforts (intervention or treatment to prevent harm among youth already abusing substances). The ways in which mentors can promote the recovery process of substance-misusing youth remains unexplored. Of note, studies of natural mentoring showed significant positive effects more frequently than did studies of the impact of programmatic mentoring. It seems that building on youth’s existing social resources and the presence of adult role models may play a meaningful role in preventing adolescent substance use.

The review suggests several take-home messages for mentoring researchers and practitioners. At intake, practitioners can incorporate a more thorough assessment of youth alcohol and illicit drug use and/or exposure to substance use in peer, school, and family contexts. Identifying youth’s risks and needs at early stages of the intervention will prepare mentors to be better equipped to support youth. Given the promising findings on the role of natural mentors, substance use preventative programs can also employ the Youth Initiated Mentoring practice where youth nominate mentors from their own social network. Such an approach may help to promote resilience among youth to resist substance use and associated problems. Finally, future research could explore how mentoring programs can be implemented from secondary and tertiary prevention frameworks. Outreach programs, in particular, can utilize mentoring components to engage and sustain youth in treatment programs.

  • Introduction

    Adolescence is a time for experimentation, sorting through problems with school and peers, and growing independence. Adolescents are neurologically wired to seek new experiences and take risks while they work to form their individual identities. These factors often put young people at risk for engaging in unhealthy behaviors, including substance use, that can have serious long-term consequences.1

    People are most likely to try and may begin abusing various substances—including tobacco, alcohol, marijuana, prescription, and illegal drugs—during adolescence. By the senior year of high school, almost 70 percent of students in the United States will have tasted alcohol, about half will have taken an illegal drug, and more than 20 percent will have used a prescription drug for a nonprescribed purpose.2 Opioid use is also of concern for youth, as 3.6 percent of adolescents ages 12–17 in the United States reported misusing opioids (primarily in the form of prescriptions) over the past year.3 While rates of opioid misuse have decreased in recent decades among youth because such pills are harder to obtain, deaths from overdose among adolescents is increasing.4

    Individual factors such as genetic vulnerability; a lack of impulse control; susceptibility to peer pressure; high sensation seeking; the presence of mental health conditions including depression, anxiety, or ADHD; or positive attitudes toward drugs influence whether a young person will initiate substance use. Social factors also heavily influence whether an adolescent uses substances, including the availability of drugs within their school or community, whether they have friends who use drugs, a history of being exposed to violence, experience of physical or emotional abuse, the presence of mental illness in the family, drug use in the household, and level of parental monitoring.5

    Those who work with youth often speculate about the positive impact a mentor can have on preventing the initiation of substance use, reducing use, or assisting a young person with a substance use disorder in getting treatment. This review takes stock of the research that addresses the potential for mentoring to serve as a strategy for preventing and reducing substance use (as defined below) and the negative effects on personal health and well-being that may stem from this behavior. We were particularly interested in the prevention and intervention of use of opioids among youth, as this is a priority for the U.S. government, including the Office of Juvenile Justice and Delinquency Prevention (OJJDP) and the National Institutes of Health (NIH). The review focuses on the following four questions:

    1. What are the effects of mentoring on substance use and associated risks to personal health and well-being among youth?

    2. What factors condition or shape the effects of mentoring on substance use and associated risks for youth?

    3. What intervening processes are most important for linking mentoring to beneficial effects on substance use and associated risks for youth?

    4. To what extent have efforts to provide mentoring to youth with substance use prevention and intervention as priority outcomes reached and engaged the intended youth, been implemented with high quality, and been adopted and sustained by host organizations and settings?

    The scope of the present review was limited to mentoring as defined by the National Mentoring Resource Center (i.e., relationships and activities that take place between youth [i.e., mentees] and older or more experienced persons [i.e., mentors] who are acting in a nonprofessional helping capacity — whether through a program or more informally — to provide support that has its aim or realistic potential benefitting one or more areas of the young person’s development; for further details, see What is Mentoring?). This definition excludes services and supports that are offered in formal professional roles by those with advanced education or training (e.g., social work, counseling) as well as those that are exclusively or predominantly didactic in orientation (e.g., structured curriculum).

    The review’s scope was further limited to studies examining mentoring in relation to substance use, defined as the use of alcohol and/or drugs (licit and illicit) that have the potential to lead to dependence, overuse/abuse, and associated psychological, social, academic, relational, and physical health problems. In considering prevention and reduction, the review follows a public health framework of levels of prevention: primary prevention (prevention of problems before they occur; targeted to a broad audience), secondary prevention (early intervention to prevent development of more serious substance abuse problems; targeted to a more focused audience with identified risk factors or showing early signs of substance abuse), and tertiary prevention (intervention or treatment to prevent significant risks to health and well-being for those already abusing substances). Likewise, some efforts may be focused on reducing more significant levels of substance use and thus fall outside of a prevention framework.

    Using these definitions, a systematic literature search was conducted to identify journal articles, book chapters, and other types of documents that reported findings pertinent to one or more of the review’s four organizing questions. Search strategies included (a) using a set of relevant keywords to search PubMed, Proquest Dissertations and Theses, PsycINFO, and Google Scholar literature databases; (b) outreach to a listserv on youth mentoring research and practice as well as members of the National Mentoring Resource Center (NMRC) Research Board; and c) examination of research referenced in broader treatments of mentoring and delinquent behavior, such as chapters on this topic in multiple editions of the Handbook of Youth Mentoring, a recent synthesis of OJJDP-sponsored research on mentoring, and meta-analyses and systematic reviews of youth mentoring programs. Of the 50 youth mentoring studies with substance use as an outcome, a total of 19 studies met criteria for inclusion in the review; 13 studies were primarily or exclusively quantitative, and the remaining 6 were mixed methods (i.e., combination of quantitative and qualitative data). Below we summarize these studies in regard to the four questions. It should be noted that this review, while focused on mentoring and substance use research, is intended for a broad audience—both researchers and research-minded practitioners. As such, we attempt to describe the studies without getting too deep into the details of their methodologies, statistical analyses, or design limitations. Many of these details are summarized in Table 1 at the end of this review.

  • What Is the Effectiveness of Mentoring for Preventing or Reducing Substance Use and Associated Risks to Personal Health and Well-being among Youth?


      Adolescent Substance Use and Associated Risks

      With adolescence being a time of experimentation, it is also the time when initiation of substance use typically occurs, and this can happen as early as 10 or 12 years old.6,7 Many adolescents occasionally use substances, and a smaller number do so regularly. But by twelfth grade, about half of adolescents have misused an illicit drug at least once.8 There are several factors that contribute to a greater likelihood that adolescents will stay drug free: strong, positive connections with parents, other family members, their school environment, and religious institutions; having parents that monitor them consistently and with clear limits and consequences; and reduced access to illegal substances in the home, as well as legal substances that may have the potential for abuse, such as prescription medications, glues, and aerosols.9,10 Children in foster care,11,12 those with mental health concerns,13 those from low socioeconomic situations,14,15 and homeless youth16 are at particular risk for substance use. Specifically for opioids, risk factors for adolescent misuse include acute and chronic pain, physical health problems, a history of mental illness, and other substance use or misuse. Youth who have witnessed someone in their family overdose or whose peers misuse prescription painkillers are also at increased risk for opioid use.17

      There are many potential additional negative outcomes that have been shown to be predicted by substance use during adolescence. Adolescents who experiment with substances early in life are more likely to develop a substance use disorder as an adult.18,19 Adolescents who use substances, including alcohol, marijuana, and prescription drugs, have shown significantly higher rates of unprotected sex, multiple sexual partners, sex with intravenous drug users,20 inconsistent condom use, and sex in conjunction with use.21 Among male adolescents, use has been associated with aggression perpetration and victimization, including physical, psychological, and electronic aggression, and sexual coercion.22 Alcohol and marijuana use have been shown to predict physical aggression among high school students across racial backgrounds,23 while hard drug use has been found to be a predictor of sexual aggression in adolescents ages 14 to 17.24 Early substance use also predicts future perpetration of teen dating violence, even when controlling for other factors.25

      Mentoring to Prevent and Reduce Substance Use

      Mentoring has been cited by the Institute of Medicine as an example of a secondary preventive intervention, whereby prevention efforts focus on adolescents who are deemed “at-risk” for substance use but do not yet present behavioral health issues.26 However, the science of mentoring as prevention and its effects on adolescent health outcomes, including substance use, is still developing.27,28 Some mentoring programs have assessed their impact on substance use,29,30,31 but they are rarely empirically tested, leaving a gap in the literature as to the processes by which mentoring’s interpersonal nature can serve as an effective prevention tool. Big Brothers Big Sisters (BBBS) is a widely cited example of a successful large-scale mentoring program with evidence of effectiveness. Child participation was shown in one of the most widely reported studies of BBBS to be associated with a 46 percent lower likelihood of illicit drug use initiation, suggesting the potential value of mentoring as a primary prevention strategy.32 Furthermore, findings revealed in several controlled evaluations of other mentoring programs allow for cautious optimism about potential viability of mentoring interventions, but more rigorous evaluations with a focus on effect sizes are needed.33 In addition, little research specifically mentions the impact of mentoring on opioid misuse.

      In general, evaluations of mentoring programs and their impact on child and adolescent substance use vary in their designs and how substance use is measured: the studies define substance use in inconsistent ways, tend to use psychometrically weak instruments, have high attrition, often do not include a control or comparison group, have inconsistent sampling procedures, and data collection is often cross-sectional.33 With approximately 3 million youth in formal mentoring relationships in the United States,33 and the potential number of additional mentors available, the conceivable impact of evidence-based mentoring on substance use is significant. While the research to date may be inconclusive, it is in the best interest of researchers, practitioners, and youth themselves to continue to improve on the rigor of the research evaluating the effectiveness of mentoring on youth substance use and to particularly focus on the impact of mentoring on opioid misuse. While OJJDP has focused its attention on opioid affected youth and their families,34 and the National Institutes of Health are interested in research that addresses curbing use in youth through the HEAL (Helping to End Addiction Long-term) Initiative,35 results from these initiatives are still forthcoming.


      Nineteen articles reported findings that address the effectiveness of mentoring for preventing or reducing substance use among youth, with mixed results. Only four studies focused on substance use as a primary outcome;36,37,38,39 it was much more common for substance use to be included as one of a set of “risk behavior” outcomes relevant to young people.

      The studies varied in how they defined “substance use”—some used a composite measure where they asked youth about their use of alcohol, tobacco, and illicit drugs; others looked at illicit drug use alone; while others differentiated between ever using a substance and using a substance within a given time frame (such as the past two weeks or one month). This makes it nearly impossible to compare the effectiveness of various mentoring models on substance use, as the outcomes, while all substance-focused, are so varied. For those that did assess substance use as a primary outcome, they tended to be divided into studies that looked at reduction in alcohol use, marijuana use, illicit substance use, and substance use defined more broadly (often lumping together alcohol, tobacco, and other substances). There was also a fairly even split between studies that looked at the effect of natural mentoring and those that looked at programmatic mentoring. However, no studies compared the two types of mentoring.

      In summarizing these studies, we chose to divide them by the type of outcome they assessed, specifically the type of substance. Some of these studies evaluated the impact on more than one substance, so they appear in more than one section below. Of note, no studies to date specifically evaluated the effectiveness of mentoring on opioid use.

      Effects of Mentoring on Alcohol Use

      Seven studies assessed the impact of mentoring on alcohol initiation or use. Four32,36,38,40 of these studies found a significant effect, whereby mentoring was associated with delayed initiation or reduction in use. The other three studies found no significant effect of mentoring on youth drinking.41,42,43

      One study of note36 assessed alcohol and marijuana use (along with other outcomes) in a sample of Massachusetts public-school students in sixth grade assigned to one of three conditions in the Across Ages program: a mentor, a curriculum (life-skills training and community service activities), or a control group. Those students in the mentor and curriculum groups showed significantly lower levels of alcohol use compared to those in the control group. This study did not find significant differences by group for marijuana use.

      Effects of Mentoring on Marijuana Use

      A total of four studies looked at the impact of mentoring on marijuana use.36,38,40,43 Two of these studies40,43 found that having a natural mentor (as opposed to a formal mentor through a program) was associated with lower levels of marijuana use. The first natural mentoring study40 was a secondary analysis of data from a convenience sample of 65 high schools from 8 states who were part of a larger trial of Project Towards No Drug Abuse (TND), an evidence-based substance abuse prevention program.44 The authors of this secondary analysis reported the impact of having a natural mentoring relationship on substance use prevention, but all children were part of this larger prevention program. They found that having a natural mentor was associated with a lower likelihood of marijuana use compared to youth who did not report having a natural mentor.

      The second natural mentoring study43 also assessed the impact on marijuana use. This study included 770 adolescents from a longitudinal study of school dropout and drug use rates. The authors of a subset analysis concluded that youth who reported having a supportive nonrelative adult in their life (i.e., a natural mentor) tended to have lower levels of marijuana use. This study also showed that having a natural mentor was predictive of less “problem behavior,” which was a composite measure of alcohol and marijuana use and both violent and nonviolent delinquency behavior.

      The other two studies36,38 focused on the impact of mentoring on marijuana use assessed youth in formal programs but saw no significant effects of the program on this outcome. It is notable that both studies of natural mentoring that assessed marijuana use did show positive effects, but neither of the two programmatic mentoring studies reported positive effects; however, because none of these studies compared the effects of natural mentoring to those of formal mentoring programs, it is unknown whether natural mentoring is a better way of preventing marijuana use than formal mentoring programs or if the positive effects of mentoring in the two studies described above are unique findings.

      Effects of Mentoring on General Substance Use

      Most studies (10 of the 19 reporting on the effectiveness of mentoring on substance use) looked at the effect of mentoring on substance use more generally, either by asking youth broadly whether they had used drugs or by compiling responses to several individual substance use items into one substance use score. Of the 10 studies assessing general substance use, 4 reported a positive effect of mentoring, such as a decrease in the number of days of substance use,45 a protective effect regarding substance use concerns,37 less use in the past month,42 or general reduction in use.46 The remaining six studies did not show a relationship between mentoring and substance use generally defined.

      One study45 looked at a sample of 90 homeless adolescents randomized to either a community reinforcement approach (CRA) plus mentoring or treatment as usual at a drop-in center. For those who received CRA and a mentor, the percentage of substance use days in the past three months decreased over time. However, the number of mentoring sessions did not predict the variance of change in substance use, so it is unknown how many mentoring sessions were needed to see this meaningful decrease. The number of mentoring sessions did predict a decrease in problem consequences of substance use, such that attending more mentoring sessions was associated with greater decreases in problem consequences.

      Effects of Mentoring on Illicit Substance Use

      Four studies looked at the effect of mentoring on what they defined as “hard drugs” or “illicit substance use.”32,40,41,47 Three of the four studies reported a significant effect of mentoring, such that having a mentor predicted a reduction in illicit substance use47 or a lower likelihood of initiating use.32,40 One study of natural mentoring47 looked at youth recruited from a health clinic who identified as having a “significant adult” in their lives with those who did not, or the effect of natural mentoring. Results showed a significant effect of having a mentor on reported illicit drug use in the past 30 days. It should be noted that none of these four studies specifically looked at the effect of mentoring on opioid use, although it is possible youth who reported using “hard” or “illicit” drugs may have been reporting opioid painkiller or heroin use.


      1. There is promising evidence that natural mentoring may contribute to reductions in marijuana and illicit drug use, although without studies comparing the effects of natural to programmatic mentoring, it is difficult to draw firm conclusions.

      2. It is possible that programmatic mentoring may be more useful for primary prevention, where the goal is to delay substance use initiation or to reduce the use of substances until a young person is of legal age. Natural mentoring may be more useful for youth engaged in illicit drug use, although more research is needed to test these hypotheses.

      3. Due to the varied definitions of “substance use” in research to date and variations in the timeframe for when it is measured, it is difficult to compare the results of mentoring on substance use across multiple studies.

      4. Studies that ask youth more generally about substance use, particularly “illicit substance use” or hard drugs as one overarching variable, tend to show a positive impact from mentoring more consistently than do studies that ask about individual substances.

      5. While there is some support for the direct effects of formal and informal mentoring on adolescent substance use prevention, no research was found specifically reporting the effect of mentoring on opioid misuse.

  • What Factors Influence the Effectiveness of Mentoring for Preventing or Reducing Substance Use among Youth?


      Conceptually, mentee characteristics and mentoring components or practices may have an impact on the extent to which formal or informal mentoring programs are effective in preventing adolescent substance use. Meta-analytical studies suggest some evidence that youth mentoring may be more effective for boys48,49 and youth who have been involved in problem behaviors.48 Other demographic characteristics, such as age and ethnicity/race, have also been widely researched, but findings vary by special populations of youth as well as program setting and goals. Programs that serve youth with greater levels of individual and environmental risk and match youth and mentors based on similarity of interests have also shown more beneficial outcomes for youth.48

      The existing literature also points to higher estimated effects of mentoring programs when mentors endorse an advocacy role and have a helping profession background.49 Another key process that attenuates the effects of mentoring programs on adolescent delinquency, substance use, and related problem behavior is the program’s emphasis on providing emotional support (rather than teaching, modeling, and identification) for at-risk youth.50


      There were 11 studies that examined factors attenuating the effects of youth mentoring on adolescent substance use. The majority of the studies were focused on the evaluation of formal mentoring programs at community and school settings, while two studies37,43 utilized large adolescent samples to examine the role of natural mentors in preventing or reducing substance use among adolescents. The samples tended to include adolescents who are deemed “high-risk,” such as homeless adolescents,45 foster youth with ongoing child maltreatment investigations,37 those with a history of early drug use and delinquency,38 and adolescents who dropped out of high school or were suspended.51 There were, however, some studies that took a primary prevention approach, such that they targeted low-risk urban youth who lacked adult role models41 or had some form of economic or social disadvantage.32,43,52,53 The moderators primarily examined included mentee characteristics (mainly demographics and risk factors) as well as program components and processes pertaining to match duration, program attendance, and intensity of the intervention.

      Mentee Demographics

      A study exploring the impact of the BBBS community-based mentoring program on prevention of alcohol and illicit drug use found that estimated effects of the program varied by mentee gender and ethnicity.32 While the overall program evaluation revealed longer delay of initiation of illicit drug use for the mentored youth at the 18-month follow-up, further evaluation of subgroups revealed that boys and youth of color benefited from the program more than girls or Caucasian youth. For instance, youth of color were 70 percent less likely to have started using drugs. The results were mixed for Caucasian mentees; the overall impact of the program was relatively low for Caucasian boys (37.6 percent less likelihood to initiate drug use), but the risk of initiation of illicit drug use increased for Caucasian mentored girls. The effects, however, were marginally significant for prevention of alcohol use.

      The moderating effects of ethnicity and gender were also examined in a sample of high school students in an urban setting.43 The study found benefits of natural mentoring relationships to reduce only marijuana use, but not alcohol. The results were consistent across subgroups (boys and girls as well as youth of color and Caucasian youth) with no moderating effects. Similarly, an outcome evaluation of BBBS’s school-based mentoring54 found no differences in substance use outcomes in regards to mentee demographic characteristics (gender, race/ethnicity, and age).

      Mixed effects of age were found in an evaluation of a multicomponent program with group mentoring, parenting sessions, and individual counseling components targeting African-American adolescents at risk of deviance.38 The findings indicated that estimated program effects in reducing alcohol use were larger for younger mentees as compared to the older mentees, but there was no interaction of treatment group and mentee age for marijuana use. In a study with homeless youth,45 the estimated effect of a multicomponent program (community reinforcement approach + mentoring through a drop-in center) on substance use in the past three months was not associated with youth age; however, the mentored subsample was too small to capture variability among youth.

      Environmental and Individual Risks for Adolescents

      Differences in program effects on substance use were examined for youth exposed to varying levels of risk. Mentoring programs that enrolled youth who lacked natural mentors had lower estimated favorable effects on substance use outcomes than those that did not.53 In an evaluation of seven separate trials of youth mentoring programs,53 the authors noted there is no strong evidence that mentoring benefited youth differently based on their risk profile or other background characteristics; however, mental health screening was recommended to identify any unique needs of mentees during the enrollment phase.

      Other characteristics of the samples of participating youth considered whether the protective role of adult relationships compensate for the risks to which adolescents are exposed and result in reduced substance use. Foster youth with adverse childhood experiences self-reported less frequent substance use when they endorsed more protective adult relationships.37 Of note, protective adults were a composite of both parents and natural mentors, making it difficult to tease out the unique protective role of natural mentors for foster youth in that study. The benefits of natural mentors were noted in a study with an urban sample of low-risk adolescents;43 having a natural mentor alleviated the negative effects of friends’ problem behaviors as well as adolescents’ problem behavior norms and reduced adolescent substance use. Yet the link between natural mentoring and peer deviance and norms with adolescent marijuana or alcohol use were unclear, as the study utilized composite measures for substance use.

      Match Duration in Formal Mentoring Programs

      In research on the BBBS mentoring program with 10- to 16-year-old mentees from low-income backgrounds, findings indicated stronger favorable effects of the program on substance use among youth matched with mentors for 12 or more months.52 Another study examining the impact of the National Guard Youth Challenge Program did not find differences in binge drinking or marijuana outcomes as a function of match length.51 There were no differences between youth who were matched for less than 21 months, 21 to 38 months, or more than 38 months.

      Program Dose, Attendance, and Mentoring Quality

      Mixed effects were identified regarding the program intensity and youth outcomes in two different studies. An evaluation of the Across Ages intergenerational mentoring program found no differences between middle school students assigned to the mentoring and parenting intervention in their overall frequency of substance use in the past two months, as compared to the students in the parenting intervention only — and the results did not vary across groups by session attendance or intensity.39 However, further subgroup analysis revealed that although mentor relationship quality was not related to substance use on its own, adolescents whose mentors were assessed by program staff as more involved with youth and showed high fidelity to the program components attended more sessions in the program and reported significantly more knowledge on substance use.39

      Finally, the aforementioned study with homeless youth45 found no interaction between program intensity and attendance for estimated effects on overall frequency of substance use. However, attending more sessions of CRA and mentoring moderated the effects of the intervention and intensity of the treatment was associated with higher reductions in problems related to substance use.


      1. The review indicates some support for moderating effects of gender, age, and ethnicity/race in mentoring program outcomes. In particular, boys, youth of color, and younger youth were more likely to benefit from mentoring programs, but research is too limited to be able to draw firm conclusions.

      2. Natural mentoring relationships seem to be protective against substance use among a variety of subgroups (boys and girls as well as youth of color and Caucasian youth). These findings, although preliminary, illustrate that the presence of positive adult role models may compensate for the individual and environmental risks associated with adolescent substance use.

      3. There are some promising findings associated with a mentor’s level of involvement and adherence to the program. When program staff rated mentors as highly involved and loyal to the program implementation, there were some effects toward substance use knowledge, but not necessarily a link to substance use behavior.

  • What Pathways Are Important in Linking Mentoring to Prevention or Reduction of Substance Use among Youth?


      A variety of theoretical approaches have focused on the processes through which informal and formal mentoring relationships promote positive youth developmental outcomes and prevent problem behavior. Rhodes’ Developmental Model30 and Keller’s Systemic Model55 point to the role of mentors to foster youth’s connection to parents, teachers, and prosocial peers. Consistent with those conceptual models, studies exploring risk and protective factors related to adolescent substance use also emphasize the role of social context with particular attention to adolescents’ interactions with their peers and parents. For instance, a large body of research has focused on adolescents’ deviant peer associations and perceptions of peer substance use and found links to an increase in substance use.56,57,58 Indeed, peer substance use is considered to be one of the strongest predictors of adolescent substance use, and longitudinal studies suggest that such negative effects of peers are stable through early adulthood.58 These findings hold true across different adolescent substance use patterns, including alcohol59 and marijuana use.60,61 It appears that affiliation with substance-using peers is a context of socialization for adolescents who are reinforced to use alcohol or drugs and/or endorse related delinquent norms and activities to “fit in” with the peer group.62,63

      Adolescents’ interactions with their parents also play a role in their substance use initiation or use patterns. Exposure to adverse family environments including parental drug use, poor and inconsistent parenting, and negative communication patterns have been linked to adolescent drug use.64,65 Prospective studies have shown that adolescents who grew up in highly conflictual families had a greater risk for developing substance use disorders in late adolescence and emerging adulthood.66 Childhood maltreatment, including physical abuse, violence, and neglect, has been linked to increased risk for adolescent substance use.67,68 Family risk factors are also associated with illicit drug use; youth whose parents have an attitude that favors substance use, have a history of addiction, or youth who witnessed a family member’s overdose are at higher risk of misusing opioids and other illicit drugs.17,69

      Parents may be also protective agents for adolescent substance use. Studies document direct and indirect links of parental warmth, parental support, and family cohesion in reducing the risk of substance use. Presence of close and trusting relationships with parents may mitigate the risks of affiliation with substance-using peers and decrease the risk for alcohol and drug use.70 When adolescents have quality relations with their parents, they may be more open to hearing parental advice on peer relationships.52 The Centers for Disease Control and Prevention (CDC) reports that aforementioned factors are also protective against illicit drug miuse, including opioids; parental engagement, monitoring, and support buffer the effects of risks associated with adolescent substance misuse.69

      In sum, interventions that aim to improve parent-adolescent relationships and overall family functioning may be promising to reduce the risk of substance use among adolescents. Rhodes and colleagues52 argue that mentoring can provide a safe context for at-risk adolescents to step out of those relationships, develop new skills to navigate peer pressure, and negotiate problems that arise from conflicts with parents. An evaluation of the BBBS mentoring program has supported that claim, showing that mentoring was associated with improvements in parent-child relationships which, in turn, promote better youth developmental outcomes.71 Taken together, connections to positive adult models, outside of one’s peer group and family, may function as a corrective experience for adolescents and reduce the risk of substance use.


      There is limited research that tests the processes that link youth mentoring to adolescent alcohol and drug use. An analysis of data from 3,320 high school students40 found that school attachment mediated the longitudinal relationship between school-based natural mentoring and a wide range of substance use outcomes (any alcohol, marijuana, or hard-drug use; ever getting drunk in the past 30 days; having a binge-drinking episode in the past two weeks). However, the aforementioned indirect effect of having natural mentors was small. In another study, there were also indirect longitudinal effects of natural mentors for 770 high school students.43 The reported presence of a natural mentor predicted lower adolescent and peer problem behavior and norms, which, in turn, were associated with lower overall adolescent substance use (composite of alcohol and marijuana use).

      The review found only one study that examined the mechanisms of change in formal youth mentoring programs to prevent substance use.52 The researchers found empirical support for the benefits of parent-child relationship quality in a BBBS program. Specifically, mentee reports of their communication, level of trust, and (reverse coded) alienation to primary caregiver mediated the effects of mentoring on alcohol and drug use frequency, but only for those matched for more than 12 months. Interestingly, in that study positive and negative peer relationships and perceived self-worth were not significant paths to reduce substance use among mentees.

      Of note, some studies reported secondary effects of formal mentoring programs to promote resiliency and socioemotional competence among at-risk youth. Unfortunately, those secondary outcomes, each of which has been found in other research to be related to lower levels of substance use, were not investigated further as potential pathways to reduce substance use. For example, a randomized controlled trial examining the effects of the Across Ages Mentoring program36 over three years in a sample of ethnically diverse, low-income adolescents found improvements in self-control, self-confidence, and family bonding among mentored adolescents, but did not link any of these outcomes to substance use.


      1. Research has demonstrated limited but promising positive effects of mentoring on school, peer, and parent attachment. Because these factors have been found in previous research to be related to reductions in substance use, mentoring programs that are specifically designed to prevent adolescent substance use may incorporate components to target those contextual and relational processes.

      2. The review lends some support for the benefits of natural mentoring relationships to alleviate peer and adolescent problem behaviors which, in turn, may reduce adolescent substance use.

  • To What Extent Have Mentoring Initiatives with Potential to Prevent or Reduce Substance Use Reached Youth Most Likely to Benefit, Been Implemented with High Quality, and Been Adopted and Sustained?


      Successful implementation of mentoring programs requires a careful evaluation of the targeted youth’s characteristics and risk profiles.53 Youth who have already experimented with alcohol and drugs (as compared to those who report no use), and those who are affiliated with substance-using peers and have ongoing conflictual family relationships may be more vulnerable to having substance abuse issues.58 Those youth may present additional needs for further services and could be more challenging to recruit and retain in mentoring programs. Other challenges may include difficulties in securing primary caregiver approval for the participation of their children in the mentoring programs and facilitating regular and consistent contact with the mentee. Meta-analytical studies have pointed to the critical importance of training and supporting mentors to increase the likelihood that the relationship lasts for at least one year.48,49


      The review identified one large scale feasibility study of Project Amp,46 in which trained staff across six sites (three school-based and three health clinics) and mentors screened 1,192 adolescents in their substance use patterns and referred those who were at low to moderate risk of substance use to a preventative mentoring program. Qualitative interviews with the participating staff and mentors suggested high levels of staff buy-in to the utility of the program to reach adolescents at risk as well as high satisfaction with the training and program objectives. Mentors reported good adherence to the program curriculum and met the majority of session goals set out in the intervention, but they recommended adding more mentoring sessions to the program. There were, however, several issues related to sustainability of the screening, referral, and overall mentoring program implementation procedures. Mentors found it challenging to obtain parental consent to recruit adolescents, commute to the mentoring sites, and handle scheduling conflicts. Staff raised concerns over a lack of funding and resources to continue the program. Taken together, the findings were promising that the Project Amp was feasible and acceptable for mentors and staff to screen, refer, and intervene with adolescents at moderate risk of substance use.

      Mentee recruitment and engagement were common issues raised in mentoring programs and were also linked to program fidelity and adherence.41,51,72,73 Difficulties in engagement were reflected in treatment completion rates, which were as low as 51 percent in one study41 where authors mentioned the program was not implemented as intended. However, mentoring relationship quality among program completers was high, suggesting some promising evidence that building strong connections with the mentees is one of the first steps to sustain mentoring relationships throughout the program. Additionally, one strategy to address aforementioned recruitment issues was to utilize after-school programs as potential contexts of engaging at-risk youth in mentoring programs.72 This strategy was discussed in a mentoring program targeting girls of color and appears to be important to explore further in future research. We did not identify any studies that empirically tested the role of recruitment setting in participant engagement and program completion.

      Adherence to the program and links to sustainability were further discussed in an evaluation of the National Guard Youth Challenge Program.51,73 While fidelity to the program core elements was relatively high across sites, the last phase (which included a Youth Initiated Mentoring phase) had lower fidelity to the program.73 Sites differed from one another in terms of training, support, and other resources they offered the mentors. Qualitative interviews with participating youth showed that mentors’ encouragement and support to continue the program motivated youth to persevere despite the challenges they faced during the residential phase. Authors also underscored the utility of engaging and training natural mentors in formal mentoring programs as a means of catalyzing their ongoing connection with youth. For instance, the Youth Initiated Mentoring practice was associated with longer match length and higher likelihood of maintaining the relationship at the completion of the program, even after controlling for mentee demographic characteristics and risk factors (school suspension, conviction, or drug/alcohol use).


      1. Integrating a substance use screening and referral process to the mentee recruitment phase of mentoring programs could be promising to reach substance misusing adolescents.

      2. Ongoing support of mentors throughout the program and providing needed resources are necessary for sustainability of mentoring programs for adolescent substance use.

      3. While the empirical research is still growing, Youth Initiated Mentoring appears to be a promising practice to promote long-lasting mentoring relationships.

  • Implications for Practice

    (Mike Garringer, MENTOR: The National Mentoring Partnership)

    While one certainly wishes that there was more direct research from which to draw regarding mentoring and substance use prevention and recovery (particularly around opioids given the critical importance of that topic to the nation’s health at this time), there is some empirical evidence available to suggest mentors can play a meaningful role in preventing the onset of use in adolescents and reducing the harm once they have started on a path toward misuse. Indeed, there is a long history of using mentor-like roles in recovery programs (e.g., the “sponsor” in Alcoholics Anonymous and its variants) and one might expect that the emotional and moral support of a mentor might really boost resiliency and keep youth from turning to substances to deal with the stressors of the world. Given that mentors can also play a role in supporting youth alongside a more clinical provider’s role (for example, see the NMRC review of mentoring for youth with mental health needs) it is similarly easy to see roles for mentors in helping youth stick to treatment plans or in processing trauma and negative experiences from their youth.

    But what exactly is the right role for mentors on this topic? Unfortunately, there is little evidence in the preceding pages about whether mentors are better deployed purely in service of primary prevention or whether their role is better positioned in reducing harm once use has begun. We certainly need more research about the specific actions and types of support mentors might offer at these various stages. It is also unclear whether mentors can be effective against specific substance misuse issues, such as opioids, which may operate via unique processes as compared to other substances and respond better to different treatment paths.

    But the reality is that the nation’s mentors are being asked to support youth in both prevention and interventions for misuse. A 2017 MENTOR study74 found that about 3 percent of the nation’s mentoring programs listed this as a top four focus for their matches, but it is clear that substance misuse can potentially impact the youth in 100 percent of programs, either directly or through misuse in the home.

    So what should mentoring programs keep in mind, since they are diving into these waters anyway? There are several hints at good practice recommendations in the evidence review offered here:


      It can be challenging to deploy mentors in a prevention or intervention capacity if it is unclear how many youth in the program are struggling with misuse issues (either themselves, in the home, or among their peer group). There are many valid and reliable scales available to programs that assess youth exposure to and consumption of drugs and alcohol, and a simple Google search will identify hundreds of options of varying value. One shortcut for mentoring programs is to consult the National Mentoring Resource Center’s Measurement Guidance Toolkit, which offers several tools that might be helpful in this assessment: a general assessment of drug and alcohol use and potential for use, as well as two opioid-specific scales just added this year—one focused on youths’ direct use of opioids and one assessing their risk for misuse based on a number of factors research suggests facilitate access to and misuse of opioids. Of course, youth may not always be willing disclosers of this information, but these tools can help identify youth who are likely to need substance misuse support from their mentor, especially if included as part of a broader assessment that looks at individual and environmental risk. The characteristics discussed in relation to Question 2 in the review can help practitioners clarify what to look for in terms of elevated risk factors.


      If there is a clear thread in much of the research covered in the main review here, it is that natural mentors can have as much impact, if not more so, on youth substance use than programmatic mentors. This might be for a number of reasons: informal mentors might be more available during critical moments, might be able to keep a closer eye on a youth’s behavior and activities, and might be able to relate better to stressors in the child’s life. It might also be that programmatic mentors are often asked to focus on other things (remember, only 3 percent of programs had substance use as a top four area of emphasis). But whatever the reasons, it seems that natural mentors are key players in supporting youth around this issue.

      That means mentoring programs should consider ways of involving natural mentors in their work. One obvious way of doing this is tasking program-provided mentors with doing some additional relationship mapping with youth so that they can identify other supportive adults that they can turn to if they feel like they are struggling with issues related to substance misuse. These individuals can enhance the mentoring provided through the program and provide support in situations where the formal mentor is unavailable. Some programs may even arrange for their mentors to meet with this broader support team so that they can share information and ideas about the young person’s challenges and progress.

      The other way of bringing natural mentors into programs is to explore Youth-Initiated Mentoring models where youth nominate someone from their existing circles to step up into a more formal mentor role through a program. This can strengthen an existing relationship and may provide the youth with a mentor who is more familiar, available, and knowledgeable about the challenges they are facing.

      But really, when it comes to addressing issues of substance misuse and chemical dependency, especially with substances that are as immediately deadly as newer variants of opioids, it is “all hands on deck,” and programs should maximize the mentoring youth receive both through their program services and by rallying other community adults who can support the young person.


      Unless they are volunteering in a program that is recovery-focused for youth who are trying to break cycles of misuse, the reality is that most mentors hope they never have to deal with issues related to their mentee’s use or misuse of drugs and alcohol. While all mentors expect some ups and downs and challenges their mentee will have to overcome, it is safe to say that most mentors would rather avoid something as complicated and scary as substance abuse. But, this is a challenge for today’s young people, and the best way for mentors to be helpful around this issue is to be knowledgeable and know how to respond to warning signs. There are hundreds of websites offering guidance for parents and other caring adults on how to recognize substance misuse in adolescents, but this guide from the National Institute on Drug Abuse offers an excellent starting point for programs and mentors. Mentor training should include role plays and scenarios where mentors respond to a number of scenarios about youth and substance use, ranging from a mentee expressing interest in drugs or recent peer pressure all the way to disclosures of serious misuse.

      It is up to programs to determine how they want mentors to respond to these scenarios based on the goals and support offered by the program. But programs should not leave mentors to figure out these things on their own in stressful situations. In fact, one of the most critical aspects of this training is getting mentors to recognize where their role ends with regards to substance abuse and when the services of more experienced professionals and evidence-based interventions are needed. Mentors may struggle to understand the limitations of their role and those subjective “handoff points,” perhaps even downplaying youth admissions of substance use as a way of avoiding breaking confidentiality with the youth or not wanting to overreact to what can be normal adolescent experimentation. But programs must ensure that mentors are prepared for whatever their mentee may bring forward around this critical topic and know when the potential for harm is great enough that they should seek support from the program or the youth’s parents/guardians. This preparation is critical — no one has ever been helped with a substance misuse issue by pretending it did not exist.


      Noted mentoring researcher Jean Rhodes once said, “Mentoring is like a vitamin, not an inoculation . . . it works while you are taking it. . . .” That sentiment applies here to the resiliency needed over the long journey through childhood and adolescence if youth are to avoid imitation of substance use or the descent into misuse. It is not like having a great mentor in fifth grade will inherently pay off when offered drugs at a party in tenth grade. The reality is a really caring mentor might not be able to make much difference to a youth who is deep into opioid misuse by the time they enter their life. This means that we need to keep youth’s lives full of mentors the whole way through if we are to win this long-term struggle. Mentoring programs should think carefully about how they transition youth out of their programs and into their next mentoring relationship. Ensuring that there are handoffs — to natural mentors, to other programs, to extended family or religious leaders or coaches — can help extend the impact of programmatic mentors when those relationships inevitably come to an end. Practitioners should think about who can help keep resiliency high once their mentors have built it up.


  • References

    1. National Institute on Drug Abuse (n.d.). Principles of Adolescent Substance Use Disorder Treatment: A Research-Based Guide. Retrieved from

    2. Johnston, L. D., O’Malley, P. M., Miech, R. A., Bachman, J. G., & Schulenberg, J. E. (2014). Monitoring the future national results on adolescent drug use: Overview of key findings, 2013. Ann Arbor, MI: Institute for Social Research, The University of Michigan.

    3. Substance Abuse and Mental Health Services Administration. (2017). Key substance use and mental health indicators in the United States: Results from the 2016 National Survey on Drug Use and Health (HHS Publication No. SMA 17-5044). Rockville, MD: Center for Behavioral Health Statistics and Quality, Substance Abuse and Mental Health Services Administration. Retrieved from

    4. Curtin S. C., Tejada-Vera B., Warner M. Drug overdose deaths among adolescents aged 15–19 in the United States: 1999–2015. NCHS data brief, no 282. Hyattsville, MD: National Center for Health Statistics. 2017. Retrieved from

    5. Sussman, S., Skara, S., &Ames, S. L. (2008). Substance abuse among adolescents. Substance Use & Misuse, 43, 1802–1828.

    6.  Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. (July 17, 2014). The TEDS Report: Age of Substance Use Initiation among Treatment Admissions Aged 18 to 30. Rockville, MD. Retrieved from

    7.  Jordan, C. J., & Andersen, S. L. (2017). Sensitive periods of substance abuse: Early risk for the transition to dependence. Developmental Cognitive Neuroscience, 25, 29–44.

    8. Johnston, L. D., Miech, R. A., O'Malley, P. M., Bachman, J. G., Schulenberg, J. E. & Patrick, M.E. (2018). Monitoring the future national survey results on drug use, 1975–2017: Overview, key findings on adolescent drug use. Ann Arbor, MI: Institute for Social Research, The University of Michigan. Retrieved from - PDF

    9. U.S. Department of Health & Human Services (n.d.). Drug Use in Adolescence. Retrieved from

    10. Robertson, E. B., David, S. L., & Rao, S. A. (2003). Preventing Drug Use Among Children and Adolescents: A Research-Based Guide for Parents, Educators, and Community Leaders (2nd ed.). Washington, DC: National Institutes of Health, National Institute on Drug Abuse. NIH publication 04-4212(A). Retrieved from

    11. Braciszewski, J. M., Moore, R. S., & Stout, R. L. (2014). Rationale for a new direction in foster youth substance use disorder prevention. Journal of Substance Use, 19, 108–111.

    12. Keller, T. E., Salazar, A. M., & Courtney, M. E. (2010). Prevalence and timing of diagnosable mental health, alcohol, and substance use problems among older adolescents in the child welfare system. Children and Youth Services Review, 32, 626–634.

    13. Schwinn, T. M., Schinke, S. P., & Trent, D. N. (2010). Substance use among late adolescent urban youths: Mental health and gender influences. Addictive Behaviors, 35, 30–34.

    14. Finch, K. A., Ramo, D. E., Delucchi, K. L., Liu, H., & Prochaska, J. J. (2013). Subjective social status and substance use severity in a young adult sample. Psychology of Addictive Behaviors, 27, 901–908.

    15. Kendler, K. S., Gardner, C. O., Hickman, M., Heron, J., Macleod, J., Lewis, G., & Dick, D. M. (2014). Socioeconomic status and alcohol-related behaviors in mid to late adolescence in the Avon longitudinal study of parents and children. Journal of Studies on Alcohol and Drugs, 75, 541–545.

    16. Johnson, K. D., Whitbeck, L. B., & Hoyt, D. R. (2005). Substance abuse disorders among homeless and runaway adolescents. Journal of Drug Issues, 35, 799–816.

    17. U.S. Department of Health and Human Services, Office of Adolescent Health (2019). Opioids and Adolescents. Retrieved from

    18. Arteaga, I., Chen, C. C., & Reynolds, A. J. (2010). Childhood Predictors of Adult Substance Abuse. Children and Youth Services Review, 32(8), 1108–1120.

    19. Merline, A. C., O’Malley, P. M., Schulenberg, J. E., Bachman, J. G., & Johnston, L. D. (2004). Substance use among adults 35 years of age: prevalence, adulthood predictors, and impact of adolescent substance use. American Journal of Public Health, 94, 96–102.

    20. Ritchwood, T. D., Ford, H., DeCoster, J., Lochman, J. E., Sutton, M., & Lochman, J. E. (2015). Risky sexual behavior and substance use among adolescents: A meta-analysis. Children and Youth Services Review, 52, 74–88.

    21. Bonar, E. E., Cunningham, R. M., Chermack, S. T., Blow, F. C., Barry, K. L., Booth, B. M., & Walton, M. A. (2014). Prescription drug misuse and sexual risk behaviors among adolescents and emerging adults. Journal of Studies on Alcohol and Drugs, 75, 259–268.

    22. Margolin, G., Ramos, M. C., Baucom, B. R., Bennett, D. C., & Guran, E. L. (2013). Substance use, aggression perpetration, and victimization: temporal co-occurrence in college males and females. Journal of Interpersonal Violence, 28, 2849–2872.

    23. Mercado-Crespo, M. C., & Mbah, A. K. (2013). Race and ethnicity, substance use, and physical aggression among U.S. high school students. Journal of Interpersonal Violence, 28, 1367-1384.

    24. Yeater, E. A., Lenberg, K. L., & Bryan, A. D. (2012). Predictors of sexual aggression among male juvenile offenders. Journal of Interpersonal Violence, 27, 1242–1258.

    25. Temple, J. R., Shorey, R. C., Fite, P., Stuart, G. L., & Le, V. D. (2013). Substance use as a longitudinal predictor of the perpetration of teen dating violence. Journal of Youth and Adolescence, 42, 596–606.

    26. Hawkins, J. D., Jenson, J. M., Catalano, R., Fraser, M. W., Botvin, G. J., Shapiro, V., . . . Stone, S. (2015). Unleashing the Power of Prevention. NAM Perspectives. Discussion Paper, National Academy of Medicine, Washington, DC.

    27. Cavell, T. A., & Elledge, C. (2014). Mentoring and prevention science. In D. L. Dubois & M. J. Karcher (Eds.), Handbook of Youth Mentoring (2nd ed., pp. 29–42). Los Angeles, CA: SAGE.

    28. DuBois, D. L., Doolittle, F., Yates, B. T., Silverthorn, N., & Tebes, J. K. (2006). Research methodology and youth mentoring. Journal of Community Psychology, 34, 657–676.

    29. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today's youth. Cambridge, MA: Harvard University Press; 2002.

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

    31. DuBois, D. L., Neville, H. A., Parra, G. R., & Pugh-Lilly, A. O. (2002). Testing a new model of mentoring. New Directions in Youth Development, 93, 21–57.

    32. Grossman, J. B., & Tierney, J. P. (1998). Does mentoring work? An impact study of the Big Brothers Big Sisters Program. Evaluation Review, 22, 403–426.

    33. Rhodes J. E., & Lowe, S. R. (2009). Mentoring in adolescence. In R. M. Lerner & L. Steinberg (Eds.), Handbook of Adolescent Psychology (pp. 152–190). Hoboken, NJ: John Wiley & Sons.

    34. U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention (2019). OJJDP FY 2019 Opioid Affected Youth Initiative FY 2019 Competitive Grant Solicitation. Retrieved from

    35. U. S. Department of Health and Human Services, National Institutes of Health (n.d.). About the HEAL Initiative. Retrieved from

    36. Aseltine, R. H., Dupre, M., & Lamlein, P. (2000). Mentoring as a drug prevention strategy: An evaluation of Across Ages. Adolescent & Family Health, 1, 11–20.

    37. Brown, S. M., & Shillington, A. M. (2017). Childhood adversity and the risk of substance use and delinquency: The role of protective adult relationships. Child Abuse & Neglect, 63, 211–221.

    38. Hanlon, T. E., Bateman, R. W., Simon, B. D., O’Grady, K. E., & Carswell, S. B. (2002). An early community-based intervention for the prevention of substance abuse and other delinquent behavior. Journal of Youth and Adolescence, 31, 459–471.

    39. LoSciuto, L., Rajala, A. K., Townswend, T. N., & Taylor, A. S. (1996). An outcome evaluation of Across Ages: An intergenerational mentoring approach to drug prevention. Journal of Adolescent Research, 11, 116–129.

    40. Black, D. S., Grenard, J. L., Sussman, S., & Rohrbach, L. A. (2010). The influence of school-based natural mentoring relationships on school attachment and subsequent adolescent risk behaviors. Health Education Research, 25, 892–902.

    41. Bodin, M., & Leifman, H. (2011). A randomized effectiveness trial of an adult-to-youth mentoring program in Sweden. Addiction Research and Theory, 19, 438–447.

    42. DuBois, D. L., & Silverthorn, N. (2005b). Natural mentoring relationships and adolescent health: Evidence from a national study. American Journal of Public Health, 95, 518–524.

    43. Zimmerman, M. A., Bingenheimer, J. B., & Notaro, P. C. (2002). Natural mentors and adolescent resiliency: A study with urban youth. American Journal of Community Psychology, 30, 221–243.

    44. Rohrbach, L. A., Gunning, M., Sun, P., & Sussman, S. (2010). The Project Towards No Drug Abuse (TND) dissemination trial: Implementation fidelity and immediate outcomes. Prevention Science, 11, 77–88.

    45. Bartle-Haring, S., Slesnick, N., Collins, J., Erdem, G., & Buettner, C. (2012) The Utility of Mentoring Homeless Adolescents: A Pilot Study. The American Journal of Drug and Alcohol Abuse, 38, 350–358.

    46. Winn, L. A. P., Paquette, K. L., Donegan, L. R. W., Wilkey, C. M., & Ferreira, K. N. (2019). Enhancing adolescent SBIRT with a peer-delivered intervention: An implementation study. Journal of Substance Abuse Treatment, 103, 14–22.

    47. Beier, S. R., Rosenfeld, W. D., Spitalny, K. C., Zansky, S. M., & Bontempo, A. N. (2000). The potential role of an adult mentor in influencing high-risk behavior in adolescents. Archives of Pediatrics and Adolescent Medicine, 154, 328–331.

    48. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, 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.

    49. Raposa, E. B., Rhodes, J., Stams, G. J. J., Card, N., Burton, S., Schwartz, S., . . . & Hussain, S. (2019). The effects of youth mentoring programs: A meta-analysis of outcome studies. Journal of Youth and Adolescence, 48, 423–443.

    50. 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.

    51. Schwartz, S. E., Rhodes, J. E., Spencer, R., & Grossman, J. B. (2013). Youth Initiated Mentoring: Investigating a new approach to working with vulnerable adolescents. American Journal of Community Psychology, 52, 155–169.

    52. Rhodes, J. E., Reddy, R., & Grossman, J. B. (2005). The protective influence of mentoring on adolescents’ substance use: Direct and indirect pathways. Applied Developmental Science, 9, 31–47.

    53. Herrera, C., DuBois, D. L., & Grossman, J. B. (2013). The Role of Risk: Mentoring Experiences and Outcomes for Youth with Varying Risk Profiles. New York, NY: A Public/Private Ventures project distributed by MDRC.

    54. Herrera, C., Grossman, J. B., Kauh, T. J., & McMaken, J. (2011). Mentoring in schools: an impact study of Big Brothers Big Sisters school-based mentoring. Child Development, 82, 346–361.

    55. Keller, T. E. (2005). A systemic model of the youth mentoring intervention. The Journal of Primary Prevention, 26, 169–188.

    56. Henry, K. L. (2008). Low prosocial attachment, involvement with drug-using peers, and adolescent drug use: A longitudinal examination of mediational mechanisms. Psychology of Addictive Behaviors, 22(2), 302–308.

    57. Sussman, S., Pokhrel, P., Ashmore, R. D., & Brown, B. B. (2007). Adolescent peer group identification and characteristics: a review of the literature. Addictive behaviors, 32(8), 1602–1627.

    58. Van Ryzin, M. J., Fosco, G. M., & Dishion, T. J. (2012). Family and peer predictors of substance use from early adolescence to early adulthood: An 11-year prospective analysis. Addictive Behaviors, 37, 1314–1324.

    59. Jamison, J., & Myers, L. B. (2008). Peer-group and price influence students drinking along with planned behaviour. Alcohol and Alcoholism,43, 492–497.

    60. Aseltine, R. H. (1995). A reconsideration of parental and peer influences on adolescent deviance. Journal of Health and Social Behavior, 36, 103–121.

    61. Rose, R. L., Bearden, W. O., & Teel, J. E. (1992). An attributional analysis of resistance to group pressure regarding illicit drug and alcohol consumption. Journal of Consumer Research, 19, 1–13.

    62. Dishion, T. J., & Owen, L. D. (2002). A longitudinal analysis of friendships and substance use: Bidirectional influence from adolescence to adulthood. Developmental Psychology, 38, 480–491.

    63. Patterson, G. R., Dishion, T. J., & Yoerger, K. (2000). Adolescent growth in new forms of problem behavior: Macro- and micro-peer dynamics. Prevention Science, 1, 3–13.

    64. Vakalahi, H. F. (2001). Adolescent Substance Use and Family-Based Risk and Protective Factors: A Literature Review. Journal of Drug Education, 31, 29–46.

    65. Madu, S. N., & Matla, M. P. (2003). Illicit drug use, cigarette smoking and alcohol drinking behaviour among a sample of high school adolescents in the Pietersburg area of the Northern Province, South Africa. Journal of Adolescence, 26, 121–136.

    66. Skeer, M., McCormick, M. C., Normand, S. L. T., Buka, S. L., & Gilman, S. E. (2009). A prospective study of familial conflict, psychological stress, and the development of substance use disorders in adolescence. Drug and Alcohol Dependence, 104, 65–72.

    67. Wall, A. E., & Kohl, P. L. (2007). Substance use in maltreated youth: Findings from the National Survey of Child and Adolescent Well-Being. Child Maltreatment, 12, 20–30.

    68. Singh, VA. S., Thornton, T., & Tonmyr, L. (2011). Determinants of substance abuse in a population of children and adolescents involved with the child welfare system. International Journal of Mental Health and Addiction, 9, 382–397.

    69. Centers for Disease Control and Prevention. (2019). High-Risk Substance Use Among Youth. Atlanta, GA: US Department of Health and Human ServicesRetrieved from

    70. Dishion, T. J., Nelson, S. E., & Bullock, B. M. (2004). Premature adolescent autonomy: Parent disengagement and deviant peer process in the amplification of problem behavior. Journal of Adolescence, 27, 515–530.

    71. Rhodes, J. E., Grossman, J. B., & Resch, N. L. (2000). Agents of change: pathways through which mentoring relationships influence adolescents’ academic adjustment. Child Development, 71, 1662–71.

    72. 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.

    73. Millenky, M. Bloom, D., Muller-Ravett, S., & Broadus, J. (2011). Staying on Course: Three-Year Results of the National Guard Youth Challenge Evaluation. MDRC Paper.

    74. Garringer, M., McQuillin, S., & McDaniel, H. (2017). Examining youth mentoring services across America: Findings from the 2016 Mentoring Program Survey. Boston, MA: MENTOR: The National Mentoring Partnership.

  • 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.

Mentoring for AI/AN Youth
PDF button Facebook button Twitter button

May 2019

This review examines research on mentoring American Indian and Alaska Native (AI/AN) youth. The review is organized around four questions:

  1. What is the effectiveness of mentoring for American Indian and Alaska Native youth?

  2. What factors influence the effectiveness of mentoring for American Indian and Alaska Native youth?

  3. What pathways are most important in linking mentoring to outcomes for American Indian and Alaska Native youth?

  4. To what extent have mentoring initiatives for American Indian and Alaska Native youth reached and engaged the youth, been implemented with high quality, and been adopted and sustained?

The mentoring literature focused on American Indian and Alaska Native youthi is growing, and it is further informed by empirical work from other countries, such as Canada and New Zealand. Nevertheless, rigorous research remains limited in understanding the strengths and needs of this group. Studies to date demonstrate some promise of mentoring as an intervention for Native youth. However, due to the risks, barriers, and challenges, often including geographic isolation, experienced by this population, it is important for researchers and program developers to carefully consider benefits and risks of formal and natural mentoring within this broader cultural context and experience. With this consideration as context, the existing evidence points toward several preliminary conclusions:

  • Mentoring for American Indian and Alaska Native youth appears to have the potential to improve academic outcomes (e.g., academic performance, school attendance), health outcomes (e.g., mental health, obesity), interpersonal strengths (e.g., confidence, leadership), and social relationships among this group, although the experimental or well-controlled quasi-experimental designs needed to most rigously assess possible benefits are lacking.

  • Research supports the role of culture in all types of mentoring for Native youth, including traditional mentoring programs (such as Big Brothers Big Sisters) and informal or natural supports, such as those that may be provided through tribal elders and extended family and community members. Examples of ways of emphasizing culture in the programs studied can possibly include traditional storytelling and activities, giving back to the community, cultural dances and ceremonies, as well as reflecting on historical trauma and current hardships.

This review describes information to consider when developing and implementing mentoring programs for Native youth. Historically, many Native communities have been affected by assimilationist policies and programs that separated children from their families (e.g., boarding schools and foster care/adoption programs). As a result, programs should be aware of the potential for cultural mistrust, which could pose challenges to engaging in relationships with mentors who come from outside the community. Therefore, mentoring programs should carefully consider recruitment, engagement, and retention efforts that include incorporating the cultural perspective and family and community support, as well as collaborative relationships with existing and trusted agencies. Preparing mentors to understand the importance of building and maintaining trust, and operating from a strengths-based and culturally attuned perspective, may be critical elements of mentoring programs for this population. Native communities often place a strong value on inclusion of family and community members , and programs may capitalize on this value as a means of engaging family and community members to serve as supportive role models and advisers for the youth, and/or to support mentor-mentee relationship of mentors drawn from outside the community. Other elements to consider include incorporating culturally related activities, such as traditional games and historical storytelling, which can advance the cultural identity of the youth and help teach them tools used to thrive within, and outside, their communities. In mentoring work with Native youth, it is critical to provide healthy, supportive role models, keeping in mind that the program context may be situated in an environment with overlapping social problems influenced by experience of historical trauma.

i From this point forward, when discussing this population as a group, they will be referred to as “Native” or AI/AN youth.

  • Introduction

    In 2017, there were 370 million indigenous people worldwide, equaling less than 5 percent of the world’s population.2 In 2015, there were 5.4 million people in the United States who identified as American Indian and Alaska Native (AI/AN) composing 2 percent of the total population.3 Individuals identify as AI/AN at varying levels in states across the United States. Of the five states with the greatest proportion of AI/AN residents, 19.4 percent of the total population of Alaska identified as AI/AN with 13.5 percent in Oklahoma, 10.4 percent in New Mexico, 10.1 percent in South Dakota, and 8.0 percent in Montana.3 According to the National Congress of American Indians (2018)4 “there are 573 federally recognized Indian Nations (variously called tribes, nations, bands, pueblos, communities, and native villages) in the United States.”

    Diversity of language, traditions, values, and strengths characterizes the different groups of people who identify as American Indian and Alaska Native. Generally, AI/AN people bring a variety of important strengths to their communities and to the broader society. The prominence of the role of elders as natural mentors is located within a larger set of cultural strengths that are characteristic across many, if not most, Native groups.5 Other strengths often cited across multiple Native groups include a historical view, a collectivist value of community, the importance of spirituality,6 and a holistic approach to health and well-being7 as common themes. Native American worldviews attach sovereignty to “the people and the land, a world of stories rather than facts, a world in which the community rather than the person is central.”8 Many of these strengths coincide well with mentoring as a positive youth development model.

    Alongside these strengths, Native youth also are more likely than other youth in the United States to face risks to their development. These include poverty,3 suicide,9 alcohol and drug abuse,9 teen pregnancy,10 and high school dropout11. This is not an exhaustive list, but reflects many of the challenges encountered by today’s Native youth. For example, exposure to poverty and unemployment can negatively influence mental health, including substance use and suicide attempts.12

    In considering mentoring as an intervention for Native youth, it is necessary to consider the term “historical trauma,” which has been and continues to be experienced by this population.13, 14 Historical trauma has been exacerbated by assimilation programs practiced just a generation or two ago, including residential, military-style boarding schools and foster care/adoption programs where the Native child(ren) were placed in the homes of non-Native families. In these programs, many Native children experienced verbal, emotional, spiritual, physical, and sexual abuse15, 16 and were often removed from their culture, including their people, language, religion, and customs.17 Based on research by Mooradian and colleagues,18 such experiences are likely to lead to long-term, negative effects on future relationships, including relationships with intimate partners, friends, parents, and tribal community members. A common theme in this study is that the Native participants expressed fear of social services and mistrust of mainstream society as a result of these practices.18 Consistent with these findings, other research has noted that assimilationist practices have negatively impacted the development of Native people, including education, employment and health insurance, homelessness, homeownership, household income, and need for public assistance.19

    Responding to these intersecting challenges is complex, with mentoring playing one potential role. Providing access to trained mentors via a mentoring program can help reduce challenges, such as drug and alcohol abuse,20 risk of suicide,21 teen pregnancy,10 geographic isolation, limited health care,20 and high school dropout rates.22, 23, 11 However, given the negative perception that some social service agencies have had with historical trauma, there may be a tendency for youth and their families in Native communities to be wary of entrusting their children to agency-sponsored mentors in the absence of a strong and trusted community-agency partnership.24 This distrust is of particular importance to consider in developing and evaluating mentoring programs.

    The National Mentoring Resource Center defines mentoring as “relationships and activities that take place between youth (i.e., mentees) and older or more experienced persons (i.e., mentors) who are acting in a nonprofessional helping capacity, whether through a program or more informally, to provide support that benefits one or more areas of the young person’s development” (for further details, see What is Mentoring?).25 Like other indigenous populations in the United States and abroad, there is a lengthy history of American Indian and Alaska Native communities emphasizing the centrality of family, which may include elders and community members, as a source of motivation and support. Community members include neighbors and members of tribal schools and community organizations. Extended family and community members are of particular interest to this mentoring review, given the importance that is customarily placed on these relationships within indigenous cultures.

    Recognizing the potential for mistrust of social service agencies within Native communities, this review takes the position that, as an intervention built on relationships between adults and children, mentoring may represent a critical linkage between people who have historically been separated by assimilation practices. In many Native communities, elders play an important role in providing guidance to families and youth.5 Although not synonymous with mentoring, elders play a role within Native communities that has elements in common with mentoring, in which an older and more experienced adult in the community provides guidance and support to encourage positive development among youth. Recognizing this intersection between the roles of elder and mentor can inform how relationships of youth with mentors from outside the community, or with elders within the community acting in the role of mentors, can play an important role in engaging and empowering Native youth.

    The strengths perspective offers a framework for understanding the mechanisms underlying the role of mentoring in promoting positive development among Native youth,26, 27 and we offer a brief review of it here as it applies to this population. The strengths perspective is a holistic approach utilizing youths’ positive attributes including “promise, strengths, assets, hopes, and dreams, as well as our own strengths, talents, assets, potential, hopes, and dreams.”28 Further, the strengths perspective focuses on natural supports in the youth’s life, which often include support systems and resources within the surrounding environment. The strengths perspective recognizes that even distressed environments have strengths available. Focusing on strengths can also result in the seeking of additional support and services, thereby increasing supportive networks accessed by Native youth.25, 26 This perspective is reflected by other theoretical frameworks, including resilience, positive youth development, and critical mentoring.

    The strengths perspective can be directly applied to Native youth served by mentoring relationships. If a Native youth draws strength from spirituality, then spirituality can be a source of helping the youth develop a sense of purpose and serve as a means of connecting to others, including forming relationships.29 The cultural views of wellness and healing are also strengths-grounded.30 Culture values that include elders as mentors and view the culture and community as sources of support, resources, and skills are strengths-based.8

    The focus of this review is to examine research on natural and program-based mentoring for promoting strengths in Native youth, and to explore what we know about the promise and potential concerns associated with mentoring Native youth 5 to 18 years of age. In particular, this review addresses the following questions:

    1. What is the effectiveness of mentoring for American Indian and Alaska Native youth?

    2. What factors influence the effectiveness of mentoring for American Indian and Alaska Native youth?

    3. What pathways are most important in linking mentoring to outcomes for American Indian and Alaska Native youth?

    4. To what extent have mentoring initiatives for American Indian and Alaska Native youth reached and engaged the youth, been implemented with high quality, and been adopted and sustained?

    This review examines studies of mentoring from a variety of delivery methods, ranging from natural supports to formal mentorship programs, among American Indian and Alaska Native youth. While focusing primarily on implications for mentoring these two groups, this review will consider research on the broader population of Hawaiian Native and Native American youth generally in the United States. Because U.S.-based mentoring research on the populations is relatively nascent, the review will also draw on the richer body of research on indigenous populations in Canada, New Zealand, and other contexts where similar work is conducted. Such work conducted outside the United States can have direct or indirect implications for U.S. Native populations with regard to cultural comparability, as well as in considering barriers, challenges, and opportunities faced by youth in the United States and other contexts. While U.S. Native youth are not the same as other groups in every respect, the experiences of Native groups in other regions have important commonalities and merit considering research on mentoring in other contexts, including how such research can inform understanding of and practice in Native communities.

    When referring to “indigenous” populations, this review will draw on the World Health Organization (WHO) definition: “Indigenous populations are communities that live within, or are attached to, geographically distinct traditional habitats or ancestral territories, and who identify themselves as being part of a distinct cultural group, descended from groups present in the area before modern states were created and current borders defined.”31

    A literature search was conducted to identify journal articles, book chapters, and other types of reports relevant to one or more of the principal questions for this review, including searches of PubMed, ProQuest Dissertations & Theses, PsycINFO, and Google Scholar, using an established set of keywords. When searching for Native American youth, keywords, such as Native American, American Indian, Alaska Native, First Nation, and Indigenous People, were used. Sources were gathered from the United States, Canada, and New Zealand. This search identified a total of nineteen articles/reports/dissertations that met criteria for inclusion in this review.

  • What Is the Effectiveness of Mentoring for American Indian or Alaska Native Youth?


      It is important to critically consider the available evidence, and to assess the mentoring role broadly enough to incorporate community-based resources, such as extended family and elders, who commonly serve in this role.32 It is also critical to weigh important cautions in implementing mentorship interventions with this population. While mentoring maintains appeal in intervening and serving youth from this population, the hard work of committed mentors and programs cannot entirely address the systemic challenges faced by this group.


      This section reviews evidence for effectiveness of mentoring for Native youth. Overall, the research reviewed is limited to formal mentoring programs, such as those provided by organizations like Big Brothers Big Sisters. Other direct mentoring programs, such as Upward Bound, GEAR-UP and 4-H initiatives, as well as indirect mentoring programs, such as tribal schools and Out-of-School Time (after-school) programs, can be examples of important mentoring services focused on Native youth. However, research evidence of program effectiveness was not available for such programs. Due to the lack of research on Native youth, this review includes research from other populations (e.g., Hawaiian Native, indigenous groups in Canada and New Zealand). Further, the Native youth population is relatively small in number and because a requirement of this program review is that research be gathered from sources predominantly of Native youth, research is limited. This requirement eliminated some research conducted in larger cities or urban areas on programs that included a small minority of Native youth that amount to too small of the percentage of the sample to allow conclusions relevant to this review. As a result, although many Native youth in the United States live in urban areas, most of the research that qualified for this review comes predominantly from rural areas.

      Community-based one-on-one mentoring

      Research conducted within a traditional, Big Brothers Big Sisters community-based mentoring program in Canada suggests that mentoring as an intervention may have unique and important positive effects on the mental health of aboriginal youth living in large Canadian urban centers.33 Mentoring experiences and mental health and behavioral outcomes were compared between Aboriginal (AB) youth and non-AB youth (i.e., those for whom services were sought but a mentoring relationship not established during the course of the study) who were ages 6 to 17. The authors did not report the racial/ethnic composition of matches but noted that less than 1 percent of the mentors in the study were themselves Aboriginal. “Mentored youth within the AB group demonstrated significant reductions in parent-rated emotional problems and symptoms of social anxiety not found among mentored non-AB youth” (p. 61).32 The findings suggest that AB youth were much more likely than non-AB youth to indicate a high-quality and dependable mentoring relationship, and yet were less likely to be in a long-standing, on-going relationship with a mentor 18 months into the study.32

      Group-based mentoring

      A strengths-based, peer mentoring intervention in Canada involved youth participants in a research project to examine the impact of the program in supporting student transition from elementary to high school.34, 35, 36 The primary component is a weekly, group mentoring program for grade seven and eight students facilitated by two First Nations young adult mentors.34 In addition, eighth grade students are invited to participate in two full-day conferences prior to transition to high school, and a three-day, intensive outdoor experiential program focused on leadership and healthy relationship skills is offered.34 Finally, a peer mentoring program focuses on the development of healthy and positive relationships between junior (grade 9) mentees and senior (grades 10 through 12) mentors.34 Across three studies, data suggest that student participants and educators perceived that the programming contributed to improved academic performance and school attendance, strengthened relationships, an increased sense of belonging, and increased confidence and leadership skills34 following participation in the program. In a quantitative analysis over two years examining differences between mentored Native youth and youth not mentored,35 mentored youth who participated for two years demonstrated stronger benefits in terms of cultural identity and mental health relative to those who received mentoring for one year or who did not receive mentoring. At baseline, these groups were similar in terms of gender, family economic status, and First Nations ancestry, but the mentored group was younger and more likely to come from a two-parent family.

      Additionally, in a more in-depth, mixed-method case study of programs in 15 schools, elementary and secondary students, as well as educators and administrators, reported (via post-intervention surveys) an increase in student success in association with their program involvement. Furthermore, a survey of secondary students revealed an increased sense of belonging, peer connections, and skills in association with (but not necessarily due to) program involvement. Survey findings showed that 81 percent of participating youth felt more connected to their culture while at school.34 An important role of culture was suggested by interview data as well. Process evaluation data reinforce the importance of culturally relevant experiences and role models as key to the success of the program.

      Another group mentoring program used mixed methods to study Native youth within a rural tribal school. The data showed strong pre- to post-test increases in youths’ preference for education and career advancement.36 Significant change on various outcomes between pre- and post-data, as well as differences relative to the comparison group at post-test, provided evidence that the group mentorship partnership between two universities and one tribal school increased Native youth’s academic skills to be more successful in school. Outcomes include increased notation of educational factors (stay in school, ask lots of questions, take the right classes, get involved in school activities), and such results were similar to or higher than in the comparison group. Further, post-survey results showed increased mean scores on educational and career perspectives as well as cultural identity from pre- to post-test, and means scores that were higher than those of the comparison group at post-test. In a comparison of this group relative to a non-Native, non-tribal, rural school youth group, findings further supported the positive influence of the intervention on seventh and eighth graders’ educational perspectives.37

      An evaluation of another group mentoring program used qualitative data from individual and small group interviews, as well as drawing activities, to construct a case study examining the influence of mentoring on Native youth attendance, engagement at school, and positive lifestyle choices in Australia.38 Of the 126 participants, 55 were mentees and the remaining were mentors, parents/caregivers, elders, and school staff. Results provided support for the value of mentors as role models for education and career development.37

      Mentoring focused on domain-specific outcomes

      Mentoring programs focused on more specific content (e.g., STEM-related skills, business and entrepreneurship, health) also offer preliminary findings suggestive of effectiveness among Native youth. In a study of a STEM (i.e., Science, Technology, Engineering, Mathematics) intervention geared toward Native third to eighth graders, mentoring was provided in the school during lunchtime every other week, and through six field trips focused on science education.39 Of the 60 mentors in the program, half were Native, 15 were non-Native STEM professionals, and 15 were non-Native university students. Student mentees reported high levels of satisfaction, rating their mentors on items such as feeling excited, having fun, and enjoying activities, as well as trusting mentors’ advice. In pre- and post-test surveys, students reported increased interest in having a career in science, increased enjoyment of science activities at school, and an appreciation of the importance of math and science for their future. Youth also self-reported an increasing sense of “doing well in math and science,” increased encouragement by family to study science, and more frequent conversations with friends about science. Rates of school absenteeism among mentees decreased during the period of program involvement relative to the year prior to enrollment.

      A national study of Indigenous high school students within Canada40 examined virtual mentoring (i.e., interaction with mentor and content of program primarily delivered virtually) provided to high schoolers (grades 10 to 12) through community-university partnerships across Canada. In addition to two in-person conferences, content was primarily focused on online group-based challenges used to help students explore aspects of business with support of their mentors. Data were gathered across program implementation and various outcomes, including self-determination and pursuit of personal career goals. A subgroup of parents and caregivers of youth were also interviewed. Although a comparison group was not used, youth and parent/caregiver participants’ interviews reported improvements in learning and social and emotional development as a result of the program. Findings also suggest increased student engagement in school, the creation of strong mentorship bonds, and an increase in post-secondary enrollment for participants in twelfth grade.39

      Weight gain can be a concern among Native youth populations. A quasi-experimental trial was conducted with 151 children living in a rural First Nation reservation over two academic school years focused on health outcomes among Native Youth (ages 9 to 10). The evaluation documented improvements in healthy dietary choices knowledge, and attributed these changes to improved self-efficacy.41 Program participants also demonstrated significant decrease in BMI and waist circumference relative to those in the comparison group. Findings also indicate that youth participants reported improved body satisfaction relative to those in the comparison condition. In sum, data suggest promise of a peer-led approach to provide ongoing support around healthy lifestyle choices.

      The Talent Identification and Development in Education (TIDE) program was implemented in Canada through a focus on art, academics, vocational-technical, and interpersonal-social skills, in Canada.42 The program used assigned workplace representatives as mentors who focused on serving high school youth who had dropped out of school. More than half of the 88 youth enrolled in the program returned to high school, entered postsecondary programs at a college, or gained employment.41 As a result of this success, supplemental programs with similar goals have been developed to include more at-risk youth.


      1. Evidence of program effectiveness is derived from studies conducted with Native Youth in the United States and internationally (particularly Canada and New Zealand); findings provide preliminary indication that mentoring programs for Native youth can promote gains in a range of outcomes, including educational achievement and engagement, health (i.e., mental health, physical health), and social relationships.

      2. Programs do not show evidence of impact on addressing deeper systemic challenges, such as youth engagement in community or political activism, or indicators of community well-being (e.g., poverty rates) for participating youth; however, such outcomes to date have not been throughly examined.

  • What Factors Influence the Effectiveness of Mentoring for American Indian or Alaska Native Youth?


      Native American cultural values and practices are critical to the successful engagement of youth in relationships and with the broader society. Aboriginal views on youth success, including educational achievement, are holistic, rather than individualistic, by nature.43 The inclusion of family and community in youth development programs, such as mentoring, is thought to facilitate the empowering of traditional values within the program itself. Learning through mentoring, therefore, should focus on perspectives and values in accordance with this worldview, such as respect for all living things and personal and communal responsibility.44 This framing underlies considerations of what factors influence effectiveness of mentoring for this population.

      Conceptual work with this population reinforces the value of cultural norms and practices as important to resilience and coping with adversity. Qualitative research suggests that a key role in high engagement in school and educational success is the quality of mentoring relationships.45 Based on focus groups conducted with 20 Native adolescents, aimed at exploring their mentoring needs in achieving academic success in an American school system, insights emphasized the potential function that family, teachers, and peers can serve as mentors for this group. In conjunction with mentoring through a teacher-student relationship, participants highlighted the school environment, including presence of cultural values and connections to family, as critical to Native youth academic success. These findings suggest that it is important to prioritize engaging family members and elders in providing positive examples for their young people.46

      The celebration of cultural values and tradition has historically been a powerful component of Native cultures, and has served as a form of resistance to historical trauma. Indeed, a majority of the studies emphasize that mentoring programs would do well to incorporate family, community, and cultural events and activities into programming with Native peoples.

      Specific strategies for doing so include the following:

      • Focus on the development of quality and close mentor-mentee relationships.44,47

      • Incorporate traditional values to help foster a sense of giving back to the community, cultural identity, and pride in the culture.44, 47

      • Participation of mentors and staff in purposeful training on the culture, such as the role of family and community involvement, language, traditions, and history as well as potential ongoing impact of historical trauma on participants.38

      • Focus on the personal and cultural identity of youth, and include cultural components (via games and storytelling), as this has the potential to positively benefit other areas of their life, such as academics.56, 48

      In sum, research emphasizes the importance of applying an approach to mentoring with Native youth that leverages cultural tradition as well as individual and community strengths.56 For mentors, a focus on building a trusting, safe, and consist relationship with their mentees46 can be more important than sharing the mentee’s race or ethnicity.37 Through building positive relationships, Native youth can receive support in their personal, educational, and social development, which in turn can buttress healthy families and communities within this population.46

      In this section, we review the research on factors that moderate the effectiveness of mentoring for Native youth. Although a few quantitative studies indicate this process, lessons are more commonly derived from qualitative work illustrating the importance of educational conditions, culturally attuned program characteristics, and attention to resources within the community in supporting youth outcomes.


      Systematic review of mentoring

      A systematic review of mentoring focused on programs in New Zealand highlighted the importance of cultural appropriateness. Of the 26 studies that met the authors’ criteria (i.e., demonstrated program effects of formal mentoring, and identified an explicit research design), there was a positive association between involvement of family/extended family and program effectiveness. Surprisingly, findings for Maori youth suggest a negative association between cultural sensitivity and program helpfulness, and findings for Pacifica youth, suggest no association.49

      In addition to emphasizing the importance of incorporating family/extended family members into mentoring practices, this systematic review also emphasized three other considerations for addressing the complex task of incorporating cultural considerations into programming with Native youth. First, as mentioned earlier, Native youth come from many tribes, communities that each have unique cultural patterns. Thus, although some of the broader values of the Native culture are largely applicable across tribes, incorporating those aspects of culture most meaningful to those served is critical to understand and address. Second, cultural appropriateness is best captured by research that examines questions in a culturally sensitive way. Thus, if the research is drawn from values not shared by the Native culture, it is less likely to examine and perceive meaningful effectiveness for this group. Finally, mentoring programs that embed cultural values also must reflect culturally appropriate goals. According to the systematic review, the goals that are often most connected to effectiveness (e.g., interpersonal, psychological) may not be the focus of mentoring programs of Native youth.48

      Community-based one-on-one mentoring

      A study of Big Brothers Big Sisters community-based mentoring relationships suggests a variable effect of mentoring on Native, relative to non-Native, youth.33 In a comparison of non-mentored and mentored Aboriginal (AB) youth participants, overall “findings suggest that one-on-one mentoring may have positive effects on the mental health of AB youth, and may in fact be more beneficial to AB than non-AB, particularly in regards to decreasing emotional and anxiety difficulties” (p. 63).33

      Educational and learning conditions

      Although the Native population has historically experienced abuse from the education system via boarding schools (as noted in the introduction with reference to historical trauma), contemporary elders see the value of the parental and adult role in supporting children through a holistic model of education that incorporates mentoring as critical to success.50 Further, applied and engaged learning techniques are important in engaging Native youth through mentoring programs that involve an educational component. McCarthy and Benally51 found such hands-on, interactive methods to learning were particularly helpful academically for Navajo youth.

      Health and mental health benefits

      Research on the Cherokee Choices/REACH 2010 program explores the role of mentoring on wellness and health among this population,52 particularly as it relates to level of exposure to mentors, or dosage, in moderating program outcomes. Elementary school youth were divided into two conditions: those who participated with mentors within school (i.e., through lesson plans in the classroom to enhance cultural pride, self-esteem, emotional wellness, conflict resolution, and health comprehension) and those received school-based mentoring and also participated in after-school programming (which included weekly meetings focused on stress-management techniques and coping skills, teamwork, and cultural awareness). One year after the intervention concluded, both groups of children expressed gains in self-reported interest in school, academic grades, learning, and interaction with friends. However, those who also participated in after-school programming reported greater improvement than those restricted to the school day. These findings suggest that more prolonged mentoring (i.e., during the day versus after school) may be a factor that conditions the effectiveness of mentoring.

      Community involvement attentive to the needs and strengths of rural settings

      Because some Native youth reside in rural and, at times, isolated areas, mentoring programs have to consider specific operational practices.46 Due to the lack of resources, programs, and/or volunteers available in rural settings, forming partnerships with established community programs can be particularly helpful53 in facilitating effectiveness of mentoring for Native youth. Using elders or community role models to help plan and develop such partnerships can also serve as resources.46 Research in such isolated areas has demonstrated that through such partnerships, programs can successfully integrate volunteers and role models from surrounding communities. Recruitment practices can also be influenced by the rural environments. A more limited number of adult role models may require accommodations in the application process. Group mentoring may also facilitate effectiveness when volunteers or resources are limited.52, 43

      Mentoring by families and community elders

      The importance of relationship with family members and community elders for programs serving Native youth is further emphasized through qualitative work focused on Native youth development. In a systematic review focused on mental health and risk of substance use among youth, 8 of the 474 articles identified focused on interventions for Native youth in the United States, Canada, New Zealand, and Australia. Results suggested lack of research on mental health and substance use interventions for Native youth and emphasized the role of elders as role models31 for this group. Interviews with twenty Alaska Native youth, ages 11 to 18, exploring the role of relationships in resilience among this group, also highlights how engagement in kinship and friendship relationships can foster a greater sense of responsibility, competence, and self-worth.54 Participants reported greater success in core outcomes (i.e., grades, abstinence from alcohol and drugs, skill building) as a result of engagement in these relationships. Through a set of interviews with 23 elders, adults, and youth in Alaska,52 these themes were reinforced through the sharing of cultural stories by elders and adults with youth. Through this storytelling, youth reflections were captured in digital stories. Youth emphasized the power of their culture in overcoming problems, and a greater appreciation of their culture and cultural identity as a result of the sharing of stories.40


      1. Although limited primarily to qualitative studies, research suggests that effectiveness of mentoring for Native youth may be enhanced when programs actively engage family and community members. Such programs may provide greater opportunities for culturally attuned support and role modeling, as well as increased program engagement that, in turn, may be critical to youth outcomes of interest.

  • What Pathways Are Most Important in Linking Mentoring to Outcomes for American Indian or Alaska Native Youth?


      There are several concepts specific to mentoring with Native youth that are important to consider in evaluating pathways that link mentoring to outcomes for Native youth. Many qualitative studies have explored these processes and have highlighted culturally attuned components that are important for three (primary) reasons. First, culturally attuned components can benefit youth by strengthening their cultural identity and pride. Second, they utilize strengths available in rural environments, which may lack the resources necessary to host traditional mentorship programs. Third, they help strengthen a culture that is striving to overcome historical trauma and intersecting risks and challenges.


      Research that explicitly measures the pathways that link outcomes to Native youth is only beginning to emerge. Much of the work has been qualitative or focused on program evaluation, approaches which in many ways are more consistent with the larger philosophy underlying Native culture.

      The most common pathway to link mentoring to outcome for Native youth is to holistically incorporate the culture into the program. For example, programs that include elders31, 49 and the culture32 result in youth gaining a greater sense of connection to their culture, family, and community, greater sense of pride in being Native,36 and these changes in turn contribute to better academic achievement and attitudes toward school.

      As examples of this process, two studies are featured. First, Wexler and colleagues53 focused on ways by which elders and kinship members teaching and sharing their culture with youth can foster resilience. The life history interviews served as a form of mentorship as the elder taught and processed resilience and relatedness discussions.53 Similarly, Stevens and colleagues38 embedded Native culture into their STEM intervention by inviting mentees’ families to join them on the STEM-related field trips. The program is geared toward third to eighth graders, yet included parents and siblings to join in activities, which then had a ripple effect in facilitating sibling interest. Further, program facilitation during the school day provided an indirect yet important impact, as school absenteeism significantly decreased.38


      1. The primary pathway that has been studied linking mentoring to outcomes for Native youth is through intentional and community-based incorporation of Native culture into programming. This includes culture, traditions, practices, and community resources as tools that can facilitate youth and community engagement and resilience. At present, however, the incorporation of Native cultural elements has not been equally evaluated as a predictor of mentoring effectiveness.

  • To What Extent Have Mentoring Initiatives for American Indian or Alaska Native Youth Reached and Engaged the Youth, Been Implemented with High Quality, and Been Adopted and Sustained?


      The definition and philosophy that underlie mentoring provide a strong, and in some ways natural, fit with the perspectives and values that characterize Native American cultures. Yet, there is little published and empirically supported research of such work. This section of the program review will briefly discuss some of the challenges to conducting research in this area.

      There are significant factors influencing the lack of research, including:

      • The relatively small, yet geographically dispersed, population of Native youth.2, 3

      • The intensity of risks and historical trauma experienced by this group, and the related high risk of harm to Native youth, families, and communities that can be exacerbated by research or interventions that are not culturally attuned and strength-based.31, 55

      • Challenges to trust and relationships development due to historical trauma.33, 56

      • Often isolated geographic locations of many Native youth communities,38 which can translate to lack of instrumental resources (i.e., transportation, funding, volunteers) as well as community programs and social services.33, 50

      • The value that the Native culture places on connections to kin. Yet family members may benefit from additional supports in order to be able to serve as positive role models given the sustained cultural and historical trauma that has affected many Native communities.28, 45

      Drawing upon the strengths-based approach and the assets inherent to the Native culture can together make it possible to address the factors listed above. Several programs have successfully addressed these factors and achieved positive outcomes, as well as successfully shared their work through publication.


      Reached and engaged Native youth

      The material used for this review provides evidence of successfully reaching and engaging Native youth. In an effort to do so, even mainstream programs, such as Boys & Girls Clubs of America and Big Brothers Big Sisters26 have incorporated cultural values into elements of their programming, including the development of the program, selection of mentors, and program implementation. This process involved drawing on the values, traditions, language, and strengths of the local area. As tribes vary, so does their culture; thus, programs may need to holistically include the culture of the particular area to best serve their Native youth.

      Implemented with high quality

      By using qualitative research, the programs were able to use one-on-one interviews, focus groups, and other artifacts, such as youths’ artwork, to assess the quality of the mentoring programs. It was clear that the studies were able to gain youth-centered results, which appeared to be associated with advancing their individual program and sharing their results with others. Often the research appeared to be a method of further advancing the goals of the program, fostering relationship development, and advancing the culture (i.e., action research). This method of collecting data seemed to work well for the studies in this review as they appeared to be youth-centered, culturally sensitive, strengths-based, and future-driven.

      Adopted and sustained

      Even though studies highlighted through this review were typically based on one particular program or case study, themes can be generalized from these findings to considerations for adopting or sustaining programs with Native youth. For example, there was a common message of partnership and collaboration between the program developers and the tribal community. Also, successful programs demonstrate a strong understanding of the barriers challenging the youth served. For example, due to their presence in remote and isolated areas, programs have to consider transportation barriers, which may result in locating the program at the youth’s school.33, 35 As another example, cultural attunement, as demonstrated by incorporating the family unit into the youth program, resulted in indirect support to the family, which translated to interest by younger siblings in the program going forward.35 Existing programs, such as Boys & Girls Clubs, Upward Bound, 4-H, GEAR UP, and tribal schools, are other possible examples of broad-reaching programs that have incorporated culture into their existing programming and have been adopted and sustained over time.

      In addition to building partnerships between Native communities and external agencies that offer mentoring programs, another approach to promoting adoption and sustainability is to build mentoring initiatives from within Native communities. For example, an experimental study of entrepreneurship education, as supported by adult mentors from within a Native community, has also shown potential for youth development across psychosocial, behavioral health, educational, and economic outcomes.57 “Arrowhead Business Group” was cocreated through a tribal-university partnership targeting known protective factors for substance use and suicide at individual, peer, and community levels.56 This initiative provides a 16-lesson curriculum made up of discussions, games, hands-on learning, and multimedia, which targeted Native adolescents ages 13 to 16 years who were currently enrolled in high school. Over an eight-month period, the program focused on entrepreneurship and business development, life skills, self-efficacy, and finance. Using a repeated measures design with 400 adolescents in the intervention arm and 200 youth in the control arm, participants were administered a survey at baseline, immediately post-intervention, and 6, 12, and 24 months following the program. Although outcome data are not yet available, process evaluation data, including attendance records, suggest that locating such an initiative in a tribal-owned site supports engagement and participation and may facilitate desired impact.


      1. Although there are many reasons why mentoring may be an excellent resource for Native youth, research is lacking on how to reach and engage youth and on how to sustain programs over time; however, programs that incorporate the Native culture appear to be more accepted and have more perceived value by the youth and their communities.

      2. Consideration of context (i.e., rural and remote settings) in which Native youth often reside may require creative program practices to provide mentoring benefits.

  • Implications for Practice

    Mike Garringer (MENTOR) and Christian Weaver (formerly of Boys & Girls Clubs of America)

    As noted in the main evidence review on mentoring for American Indian and Alaska Native youth, tribal communities, both on and off of tribal lands, have a long and successful history of weaving natural mentoring relationships into the kinship structure of their culture and in the elevated role that elders have as teachers, problem solvers, and sources of wisdom and guidance for their communities, especially for young people. Thus, any discussion about providing additional, programmatic mentoring for tribal youth must begin with a recognition of these natural mentoring traditions and serious thought about how to honor these existing ties, tribal traditions, and indigenous ways of knowing more broadly in the delivery of services.

    This review also notes that, beyond the positive presence of elders and natural mentors, tribal youth may also benefit in myriad ways from additional mentoring provided through a program. Programmatic mentors, whether from within the Native community or from outside it, can play a vital role in helping connect these youth to educational opportunities, support the transition to the world of work, allow youth to work through traumas and personal setbacks, and provide additional or targeted support in overcoming serious challenges, such as reducing juvenile justice involvement or bolstering mental health services. While the traditional networks of supportive adults can never be replaced or duplicated in the lives of Native youth, the evidence reviewed in the previous pages makes a compelling case for also offering programmatic mentors to their lives as needed, regardless of whether those mentors are from within the tribal community.

    But the review also notes that effectively integrating mentoring services into Native communities can present challenges, particularly when those developing and implementing the services come from outside the community. In these instances, program staff and mentors must work diligently to meet indigenous communities where they are, and ensure that the programming offered is aligned with community values, culture, goals, and institutions. Strategies for doing this are provided later in this section.

    The recommendations offered to practitioners here draw on the research review itself, and practitioner experience, most notably from the work of Boys & Girls Clubs of America Native Services, which offers mentoring and other integrated services at nearly 200 Clubs in Native communities, and recently celebrated 25 years of supporting Native youth. We hope these recommendations will help staff and mentors more effectively meet the needs of Native youth, both in and out of indigenous communities, although our emphasis here is on services offered within communities on reservations or in schools or other settings with large numbers of Native youth and adults.


      Regardless of whether staff and mentors come from within the tribal community, there are several considerations to keep in mind that will ensure that mentoring relationships are meaningful, safe, and culturally relevant and responsive for Native American youth:

      • Create a safe space within the program and the relationship. Because Native youth have so commonly experienced trauma and challenging circumstances, or are being cared for by adults who have, it is important to ensure a safe and nurturing mentoring relationship. This history of trauma is thoroughly discussed in the main sections of this review and can include both historical traumas and events from the mentees’ own lives. All mentors for AI/AN youth should think carefully about how the activities they engage in might affect or be perceived by youth or their families and ensure that they don’t deepen or add to trauma already experienced. Mentors must also be willing to engage with the circumstances of the young person’s life without judgment or without inadvertently disrespecting the community’s values or sovereignty (this is most critical to mentors coming from outside the youth’s specific tribal community). Youth must feel like the program and the relationships it provides are safe places where they can share their feelings and emotions, process prior traumas, and feel unconditional love and support. One practice that might help both staff and mentors who are external to the community with creating this environment is to partake in cultural competence training that uncovers their biases. Such training can raise awareness of how personal values might come across negatively or be perceived as judgmental, condescending, or antithetical to the youth’s culture, ways of knowing, or family structures. Such training can also teach mentors, especially non-Native mentors, about subtleties of nonverbal communication, such as eye contact, touching, and pauses or silence in conversations. These things may be perceived very differently in tribal cultures. Perhaps the best things a mentor can do to support a young person is to approach the relationship with humility and self-awareness and be open to the idea that they have a responsibility to be vigilant about not sending the wrong message or adding to traumas already experienced.

        One resource that might help mentors (especially those from outside the community) with their cultural competency in serving indigenous youth is A Critical Orientation for Supporting and Inspiring Native Youth, an online learning module also developed by the National Mentoring Resource Center. This self-paced learning opportunity includes critical information about native culture and traditions, as well as a section that helps mentors explore their own cultural identity and key topics such as structural racism, unconscious bias, and cultural appropriation. This resource can be found on the NMRC website here. It is worth noting that even mentors from within Native American communities can have viewpoints or differences in experiences (especially across different tribal histories) that can leave them vulnerable to creating “unsafe” or culturally insensitive moments in these relationships. Thus, this awareness of trauma and commitment to safety within the relationship is something all mentors for these youth should strive for.

      • Get to know the specific challenges that each mentee is facing and the strengths they bring to the relationship. One of the common missteps of mentoring programs generally is to treat youth from similar demographic backgrounds as being fairly homogenous in terms of their circumstances and needs. The reality is that all mentees bring their own mix of personality and history to the relationship. Mentors and staff members need to take the time to get to know each individual young person—their story, their strengths, challenges, and hopes for the future. Mentors should pay attention to the intersectionality that resides within Native youth, where their status as an indigenous person is also wrapped up with other identities and personal circumstances that define who they are and where they want to go in life. It can be particularly helpful for mentors, but especially those from outside the tribal community, to work with the young person to map out their web of supportive adult relationships. This can help mentors identify elders and other adult community members who can identify the youth’s strengths, assets, and challenges and who can work collaboratively with the mentor and program to support the young person’s growth.

      • Use a variety of strategies for integrating indigenous culture into mentoring activities. While ideally the program as a whole will offer activities that build on or integrate Native culture, mentors should also remember that they have a personal obligation to integrate Native traditions, language, skills, ways of knowing, and arts into mentoring activities and reflections each time they meet with their mentee. While this may seem like advice mostly for mentors from outside the tribal community, the reality is that even indigenous mentors can be so caught up in helping tribal youth succeed in the broader world that they de-emphasize the value that can come from reinforcing Native identity and honoring historical practices and ways of knowing. Building indigenous culture into the match can be done in overt ways, such as working on traditional arts or crafts together or by participating together in cultural events, such as a powwow. But it can often be done in subtle ways, such as integrating indigenous knowledge into work on school assignments or thinking about traditional tribal roles and responsibilities when engaging in career exploration. As noted throughout the evidence review, this integration of culture can build positive ethnic identity and racial self-regard for the young person, which serve as powerful resilience factors that can boost confidence and esteem and negate the harm of stereotype threat. Non-Native mentors can also share their culture and traditions with mentees so that young people are exposed to a variety of values and ways of thinking—provided that it is done in a mutually respectful and positive way. Mentors may want to let the youth take the lead in sharing their culture—especially if the mentor is from outside the indigenous community—but programs are wise to be mindful of not placing the youth in a position of being an “ambassador” of their entire culture to outside adults, as that may make them uncomfortable or seem intrusive or disrespectful.

      • Provide exposure to new experiences and opportunities. Even when integrating indigenous culture in mentoring activities, there can be tremendous value, especially in programs in geographically isolated tribal communities, in exposing Native youth to the broader world and offering new experiences. All rural youth, but especially Native youth, struggle to engage in new activities outside of their community. Such experiences might include job shadows or other opportunities to see careers that are rare locally in action, tours of college campuses, or exposure to arts or culture that is uncommon to their local communities (e.g., trying Thai food, spending the day at a science museum, or attending a symphony performance). Mentors should keep in mind, however, that there might be a fine line between efforts to expand youths’ horizons and a perception that mentoring activities work to negate the youth’s culture and community. As noted above, even Native American mentors can fall into a trap of sending the inadvertent message that tribal traditions and values are antiquated. Mentors must ensure that these types of activities are, once again, safe, supportive, and reaffirming of indigenous culture.

        As noted in the review, the concept of mentoring is closely aligned with Native cultural conceptions of the role of elders and the broader adult community in explicitly teaching and supporting youth through collective relationship structures. Native cultures have been mentoring youth intentionally far longer than modern non-Native society seems to recognize. Western notions of one mentor guiding one youth must be tempered and adapted to communities where the “we” is far more important than the “me” and where young people are cared for in different ways. But as the review authors note, mentoring programs have tremendous potential to not only address community needs but to also elevate Native strengths, traditions, and culture in ways that empower Native youth and strengthen the web of support surrounding each young person.


      Keeping Native strengths and traditions at the center of the mentoring program can be a struggle for program developers and those who run programs on the ground; this is especially critical to do when those developers come from outside the tribal community. Even when programs are developed by those within tribal communities, there can be challenges integrating what can be a Westernized notion of mentoring if not appropriately implemented as noted above given a Native cultural context that is almost the polar opposite of that approach.

      We begin here with some steps that program developers from outside tribal communities can take to ensure that the implementation of their services will be well-received:

      • Recognize that each Native community is unique and develop strategies that honor and integrate this unique cultural context. According to the Bureau of Indian Affairs, there are 573 federally recognized Indian tribes, bands, nations, pueblos, rancherias, communities, and Native villages in the United States (229 are located in Alaska with the remaining located in 33 other states). There is great diversity among AI/AN communities and tribes, with each having very different histories, values, strengths, and ways of supporting members of the community, including youth. Thus, the first step to designing a mentoring program that will resonate with the community is to set aside preconceived notions and learn about the nuances and traditions that make the community in question unique and special.

      • Community leadership, understanding, and buy-in is critical to success. As mentioned in the introduction, Native communities wrestle with the historical legacy of trauma and victimization at the hands of the U.S. government and other state and regional institutions—a trauma that unfortunately continues to this day. As such, tribal communities may be especially wary of services brought to the community from the outside and are often reluctant to support new programming without a clear understanding of the motives and goals of the effort. External developers can address this by approaching tribal government, elders, and other authoritative stakeholders in the community from a position of true collaboration and by listening intently to what these leaders have to say about the history, strengths, and values of their community, as well as the needs.

        Respecting tribal sovereignty and autonomy is essential to building these relationships and getting this dialogue started on the right foot. A mentoring program, in general, should always be presented in any community as an effort to build on community strengths and integrate with existing youth–adult relationships and structures, not as something only needed to “fix” deficits. In the early stages of program development, practitioners should ask the community what they would like to see in terms of outcomes or experiences for the youth served, codeveloping program goals and recognizing when certain aspects of the program are simply not aligned with local values or traditions. In a world that increasingly emphasizes rigid fidelity to “evidence-based” practices, it can be hard for practitioners to deviate from a set program model or intervention strategy. But the reality is that when programs do not find that common ground with Native community leaders and honor the history of negative experiences, they are destined to fail. Practitioners should also note that their relationship with tribal leadership needs to be maintained over time, not just at the initial stages of program development. Frequent communication with the community’s leadership can help avoid misunderstandings and correct future missteps.

      There are also a few things for program developers of all types, including those from within tribal communities, to keep in mind as they develop services:

      • Recognize the challenges that rural isolation can also bring and choose partnerships accordingly.As noted in the review, many Native communities are located in geographically isolated communities on reservations or other land far from large population centers. In these contexts the resources available to start and maintain a program, as well as the types of activities that mentors and youth can engage in may be constrained. While this isolation will not be news to individuals who live in those communities, it does mean that the program they ultimately develop will, by necessity, be less of a “solo” effort than it might be otherwise. The evidence review here suggests partnering with existing services and institutions in the community in an effort to maximize resources and integrate the mentoring into other services and supports that are working in the community. Practitioners may have to get even more creative in how they structure mentor-mentee interactions and in how they schedule program activities in order to overcome some of the challenges of rural isolation. This includes carefully thinking through transportation and other logistical considerations. Programs will want to ensure that mentors can meet frequently and consistently so that youth get the mentoring experience they deserve and don’t experience feelings of abandonment or disrespect from mentors skipping meeting times.

        At the other end of the geographic isolation spectrum are mentoring programs that are developed from from tribal communities or reservations, where Native American youth are integrated into mentoring programs serving a general student body or are offered a discrete mentoring program within a larger non-Native school or community context. In these instances, youth are still geographically isolated, but in this case isolated from a larger community of supportive adults and services who already understand who they are and how to serve them in culturally relevant ways. In such cases, program developers may have to partner with external Native American organizations to create that tribal community and give youth access to more Native mentors and easier ways to integrate cultural traditions. Almost every urban area in the United States will have Native American community groups who bring indigenous members of the community together for cultural events, ceremonies, arts exploration, and other activities grounded in tribal traditions. These organizations make excellent partners for programs serving tribal youth outside of reservation communities and can be an excellent source of both Native mentors and elders who can support the work of non-Native mentors. So whether it’s a program working in an isolated tribal community or one serving tribal youth isolated within an urban environment, program developers will need to make specific partnerships and connections to other services so that youth and mentors can meet as intended and effectively integrate tribal culture when they do.

      • Structure match activities so that they can build on traditions and existing relationships within the tribal community. As noted above, each Native community will have its own knowledge, skills, traditions, and ways of finding meaning that can be incorporated into a mentoring program. Integrating cultural nuances into program activities, even in models where mentors and mentees largely meet out on their own, can increase the relevance of the program to youth (and demonstrate its value to perhaps skeptical elders and tribal adults). Each Native community will also have unique ways of thinking about, and definitions around, relationships, something that external service providers may not know and that internal providers may fail to fully maximize. This tapestry of relationships includes the relationships of youth with elders and other community members (note that in some indigenous communities, everyone who is not an elder may be considered a “youth,” which can sometimes complicate ideas about program delivery). Also included are the relationships between men and women, generally, and among parents, their children, and the rest of the community. Learning about the ways the community defines and, in turn, facilitates nurturing relationships can help a program figure out how to best position mentors to have a positive impact and work effectively with these other relationships and their dynamics. For example, these programs might find it much more common to have elders or other adults accompany the mentor and youth on outings or to integrate the perspective and roles of other adults into how to best support a youth in wrestling with a specific problem. These programs might also engage in more group activities, or offer secondary services to parents or other adults in the youth’s life. Regardless of the tribal community being served, and whether the developers came from within or from outside that community, chances are these programs might benefit from more group activities, a greater involvement of other adults in the match, and more emphasis on teaching or relearning tribal traditions or skills than one might expect from a more traditional “Western” mentoring program where the mentor alone may be viewed more as the agent of positive change. Practitioners serving tribal youth should consider a collectivist approach that fits the feeling of community already inherent in the lives of these youth.
  • Resources and Additional Reading

    For practitioners who are looking for even more nuanced guidance in this area, we suggest these two resources, generously provided by Boys & Girls Clubs of America Native Services, available for download on the National Mentoring Resource Center (NRMC) website:

    And, as noted above, the following e-learning module developed by the NRMC will also help both staff and mentors deliver mentoring relationships that are safe, purposeful, and culturally relevant. Practitioners might also find value in these online resources:
  • References

    1. U. S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration [SAMHSA] (2009, January). CultureCard: A Guide to Build Cultural Awareness. American Indian and Alaska Native. Retrieved from

    2. United Nations (2017, August 8). International day of the world’s indigenous peoples [press release]. Retrieved from

    3. United States Census Bureau (2015, November 2). Facts for Features: American Indian and Alaska Native Heritage Month: November 2015. Retrieved from

    4. National Congress of American Indians (n. d.). Tribal Nations and the United States: An Introduction. Retrieved from

    5. Weinberger, S. G. (2017, December). The impact of mentors on the well-being of American Indian children. Retrieved from

    6. Fleming, J., & Ledogar, R. J. (2008). Resilience and Indigenous spirituality: A literature review. Pimatisiwin: A Journal of Aboriginal and Indigenous Community Health, 6 (2), 47–64.

    7. Koithan, M., & Farrell, C. (2010). Indigenous Native American Healing Traditions. The Journal for Nurse Practitioners: JNP, 6, 477–478.

    8. Morrison, C., Fox, K., Cross, T., & Paul, R. (2010). Permanency through Wabanaki eyes: A narrative perspective from “the people who live where the sun rises.” Child Welfare, 89(1), 103–123.

    9. U.S. Department of Health and Human Services, SAMHSA’s Center for the Application of Prevention Technologies. Using Strengths to Address Alcohol Abuse and Suicide Among American Indian and Alaska Native Youth. Retrieved from

    10. Shegog, R., Rushing, S. C., Jessen, C., Lane, T. L., & Gorman, G. (2017). Native IYG: Improving psychosocial protective factors for HIV/STI and teen pregnancy prevention among youth in American Indian/Alaska native communities. Journal of Applied Research on Children, 8(1), 1–29.

    11. Camera, L. (2015, November 6). Native American Students Left Behind. Native youth post the worst achievement scores and lowest graduation rates of any student subgroup. U.S. News & World Report. Retrieved from

    12. Allen, J., Mohatt, G. V., Markstrom, C. A., Byers, L., & Novins, D. K. (2012). “Oh no, we are just getting to know you”: The relationship in research with children and youth in indigenous communities. Child Development Perspectives, 6, 55–60. doi: 10.1111/j.1750-8606.2011.00199.x

    13. Lechner, A., Cavanaugh, M., & Blyler, C. (2016, August 24). Addressing Trauma in American Indian and Alaska Native Youth. Mathematica Policy Research Reports. doi: 10.13140/RG.2.2.28558.36162. Retrieved from

    14. Ehlers, C. L., Gizer, I. R., Gilder, D. A., Ellingson, J. M., & Yehuda, R. (2013). Measuring historical trauma in an American Indian community sample: contributions of substance dependence, affective disorder, conduct disorder and PTSD. Drug and alcohol dependence, 133(1), 180–187. doi: 10.1016/j.drugalcdep.2013.05.011

    15. Goldsmith, D. J. (2002). In the best interests of an Indian child: The Indian Child Welfare Act. Juvenile and Family Court Journal, 9–17.

    16. Earle, K. A. & Cross, A. (2001). Child Abuse and Neglect among Indian/Alaska Native Children: An Analysis of Existing Data. Portland, OR: National Indian Child Welfare Association.

    17. Engel, M. H., Phillips, N. K., & DellaCava, F. A. (2010). Cultural difference and adoption policy in the United States: The quest for social justice for children. The International Journal of Children’s Rights, 18, 291–308. doi:10.1163/092755609X12488514988991

    18. Mooradian, J. K., Cross, S. L., & Stutzky, G. R. (2007). Across generations: Culture, history, and policy in the social ecology of American Indian grandparents parenting their grandchildren. Journal of Family Social Work, 10, 81–101. doi: 10.1300/J039v10n04_04

    19. O’Brien, K., Pecora, P., Echohawk, L., Evans-Campbell, T., Palmanteer-Holder, N., & White, C. R. (2010). Educational and employment achievements of American Indian/Alaska Native alumni of foster care. Families in Society: The Journal of Contemporary Social Services, 91, 149–157.

    20. The DASIS Report (2003). American Indian/Alaska Native treatment admissions in rural and urban areas: 2000. Office of Applied Studies, Substance Abuse and Mental Health Services Administration (SAMHSA), 1–4. Retrieved from

    21. Leavitt, R. A., Ertl, A., Sheats, K., Petrosky, E., Ivey-Stephenson, A., & Fowler, K. A. (2018). Suicides among American Indian/Alaska Natives - National Violent Death Reporting System, 18 States, 2003–2014. MMWR: Morbidity and Mortality Weekly Report, 67, 237–242. doi: 10.15585/mmwr.mm6708a1

    22. Tyler, J. H., & Lofstrom, M. (2009). Finishing high school: Alternative pathways and dropout recovery. The Future of Children, 19, 77–103. doi: 10.1353/foc.0.0019

    23. Tierney, W. G., Sallee, M. W., & Venegas, K. M. (2007). Access and Financial Aid: How American-Indian Students Pay for College. Journal of College Admission, 197, 14–23.

    24. Sachs, S. M. (2017, April). Returning the people to the circle: An overview on overcoming the fracturing of American Indian communities. Presented at the Western Social Science Association Meeting, San Francisco: CA.

    25. >National Mentoring Resource Center. (n. d.) What is mentoring? Retrieved from

    26. Early, T. J., & GlenMaye, L. F. (2000). Valuing families: Social work practice with families from a strengths perspective. Social Work, 45, 118–130. doi: 10.1093/sw/45.2.118

    27. Saleebey, D. (ed.) (1992) The strengths perspective in social work practice. New York: Longman.

    28. University of Kansas (2013). What is the strengths perspective? Retrieved from

    29. Cheon, J. W., & Canda, E. R. (2010). The meaning and engagement of spirituality for positive youth development in social work. Families in Society: The Journal of Contemporary Social Services, 91, 121–126. doi:10.1606/1044-3894.3981

    30. Voss, R., Douville, V., Soldier, A., & Twiss, G. (1999). Tribal and Shamanic-Based Social Work Practice: A Lakota Perspective. Social Work, 44(3), 228–241.

    31. World Health Organization. (n. d.) Indigenous populations. Retrieved from

    32. Antonio, M. C. K., & Chung-Do, J. J. (2015). Systematic review of interventions focusing on indigenous adolescent mental health and substance abuse. American Indian and Alaska Native Mental Health Research, 22(3), 36–56. doi:10.1007/s10935-016-0441-8

    33. DeWit, D. J., Wells, S., Elton Marshall. T., & George, J. (2017). Mentoring relationsihps and the mental health of Aboriginal youth in Canada. The Journal of Primary Prevention, 38, 49–66. doi: 10.1007/s10935-016-0441-8

    34. Crooks, C. V., Chiodo, D., & Thomas, D. (2010). Strengths-based programing for first nations youth in schools: Building engagement through healthy relationships and leadership skills. International Journal of Mental Health and Addiction, 8(160), 160–173. doi: 10.1007/s11469-009-9242-0

    35. Crooks, C. V., Burleigh, D., Showshoe, A., Lapp, A., Hughes, R., et al. (2015). A case study of culturally relevant schoolbased programming for First Nations youth: Improved relationships, confidence and leadership, and school success. Advances in School Mental Health Promotion, 8, 216–230.

    36. Crooks, C. V., Exner-Cortens, D., Burm, S., Lapointe, A., & Chiodo, D. (2017). Two years of relationship-focused mentoring for First Nations, Métis, and Inuit adolescents: Promoting positive mental health. Journal of Primary Prevention, 38, 87–104. doi: 10.1007/s10935-016-0457-0. doi:10.1007/s10935

    37. Aschenbrener, C., & Johnson, S. (2017). Educationally based, culturally sensitive, theory-driven mentorship intervention with at-risk Native American youth in South Dakota: A narrative review. Journal of Child and Family Studies, 26, 14–27.

    38. Peralta, L., Cinelli, R., & Bennie, A. (2018). Mentoring as a tool to engage aboriginal youth in remote Australian communities: A qualitative investigation of community members, mentees, teacher, and mentors’ perspectives. Mentoring & Tutoring: Partnership in Learning. Advance online publication. doi: 10.1080/13611267.2018.1445436

    39. Stevens, S., Andrade, R., & Page, M. (2016). Motivating young Native American students to pursue STEM learning through a culturally relevant science program. Journal of Science Education and Technology, 25, 947–960. doi:10.1007/s10956-016-9629-1

    40. Bourassa, C. (2017). In.Business: A national mentorship program for indigenous youth. Retrieved from

    41. Eskicioglu, P., Halas, J., Sénéchal, M., Wood, L., McKay, E., Villeneuve, S., Shen, G. X., Dean, H., & McGavock, J. M. (2014). Peer mentoring for Type 2 Diabetes prevention in First Nations children. Pediatrics, 133, e1624-e1631. doi: 10.1542/peds.2013-2621

    42. McCluskey, K. W., Baker, P. A., & McCluskey, A. L. A. (2005). Creative problem solving with marginalized populations: Reclaiming Lost Prizes through in-the-trenches interventions. Gifted Child Quarterly, 49, 330–341. doi: 10.1177/001698620504900406

    43. Henderson, J. S. Y. (Sákéj). (2000). Ayukpuchi: Empowering Aboriginal thought. In M. Battiste (ed.), Reclaiming Indigenous voice and vision (pp. 248–278). Vancouver: University of British Columbia Press.

    44. Klinck, J., Cardinal, C., Edwards, K., Gibson, N., Bisanz, J., & da Costa, J. (2005). Mentoring programs for aboriginal youth. Pimatisiwin: A Journal of Aboriginal and Indigenous Community Health, 3, 109–130.

    45. Strong, Z. H. (2013). Native American Youth Voices on Success, Identity, and Cultural Values: Educational Success and Positive Identity Development through Culturally Responsive Mentoring (unpublished master’s thesis). Seattle, WA: University of Washington.

    46. Sinclair, R. & Pooyak, S. (2007). Aboriginal Mentoring in Saskatoon: A Cultural Perspective. Indigenous Peoples’ Health Research Centre and Big Brothers Big Sisters of Saskatoon.

    47. Jones, D., & Skogrand, L. (2015). Informing 4-H youth development in southeast Alaska native villages. Journal of Community Engagement and Scholarship, 8(1), 36–44.

    48. Kulis, S. S., Robbins, D. E., Baker, T. M., Denetsosie, S., & Deschine Parkhurst, N. A. (2016). A latent class analysis of urban American Indian youth identities. Cultural Diversity and Ethnic Minority Psychology, 22(2), 215–228. doi: 10.1037/cdp0000024

    49. Farruggia, S. P., Bullen, P., Solomon, F., Collins, E., & Dunphy, A. (2011). Examining the cultural context of youth mentoring: A systematic review. Journal of Primary Prevention, 32, 237–251. doi: 10.1007/s10935-011-0258-4

    50. Kahn, C. B., Reinschmidt, K., Teufel-Shone, N., Oré, C. E., Henson, M., & Attakai, A. (2016). American Indian elders’ resilience: Sources of strength for building a healthy future for youth. American Indian and Alaska Native Mental Health Research, 23(3), 117–133.

    51. McCarthy, J., & Benally, J. (2003) Classroom management in a Navajo middle school. Theory Into Practice, 42, 296–304. doi: 10.1207/s15430421tip4204_6

    52. Bachar, J. J., Lefler, L. J., Reed, L., McCoy, T., Bailey, R., & Bel, R. (2006). Cherokee Choices: a diabetes prevention program for American Indians. Preventing Chronic Disease, 3(3), 1–9.

    53. Wexler, L. (2011). Intergenerational dialogue exchange and action: Introducing a community-based participatory approach to connect youth, adults and elders in an Alaskan Native community. International Journal of Qualitative Methods, 10, 248–264.

    54. Wexler, L., Moses, J., Hopper, K., Joule, L., Garoutte, J., & The LSC CIPA Team (2013). Central role of relatedness in Alaska Native youth resilience: Preliminary themes from one site of the Circumpolar Indigenous Pathways to Adulthood (CIPA) Study. American Journal of Community Psychology, 52, 393–405. doi:10.1007/s10464-013-9605-3

    55. Dorgan, B. L. (2010). The tragedy of Native American youth suicide. Psychological Services, 7(3), 213–218.

    56. Wallis, J. A. M., Riddell, J. K., Smith, C., Silvertown, J., & Pepler, D. J. (2015). Investigating patterns of participation and conversation content in an online mentoring program for northern Canadian youth. Mentoring & Tutoring: Partnership in Learning, 23, 228–247. doi: 10.1080/13611267.2015.1072395

    57. Tingey, L., Larzelere-Hinton, F., Goklish, N., Ingalls, A., Craft, T., Sprengeler, F., McGuire, C., & Barlow, A. (2016). Entrepreneurship education: A strengths-based approach to substance use and suicide prevention for American Indian adolescents. American Indian and Alaska Native Mental Health Research, 23(2), 248–270.

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.

Mentoring for Youth with Disabilities
PDF button Facebook button Twitter button

October 2018

This review examined research on mentoring for youth (ages 25 and younger) who have a disability, including physical, cognitive, learning, and developmental disabilities, and excluding psychiatric disabilities which have been discussed elsewhere.1 It addressed four questions:

  1. What is the documented effectiveness of mentoring for youth with disabilities?

  2. What factors condition or shape the effectiveness of mentoring for youth with disabilities?

  3. What are the intervening processes that are most important for linking mentoring to outcomes for youth with disabilities?

  4. To what extent have efforts that provide mentoring to youth with disabilities reached and engaged targeted youth, been implemented with high quality, and been adopted and sustained by host organizations and settings?

The review found a total of 40 studies addressing these questions. Benefits of mentoring program participation for youth with disabilities include improved employment and career-related decisions, transitions to adulthood (as well as college and work), postsecondary education goals, and independent living skills.

Although the research in this area is still relatively new, it suggests the following takeaways:

  • Potential benefits of mentoring programs for youth with disabilities include several in the areas of academic and career development, employment, psychosocial health and quality of life, transition, and life skills.
  • Although various types of mentoring models were used in these studies, it is unclear which formats work best for youth with disabilities.
  • Results suggest several potential processes occur between mentoring provision and ultimate outcomes (i.e., mediators), such as self-determination, and some factors could influence, or moderate, the effects of mentoring for youth with disabilities, including gender and ethnicity.

The review concludes with insights for practitioners that highlight a number of factors to consider when developing and implementing mentoring programs for youth with disabilities. This commentary suggests that programs looking to serve youth with disabilities consider accessibility factors that would better enable mentees to participate in activities offered, which may include not only physical access to facilities but also access to program materials in various formats. Furthermore, programs are advised to consider expanding the age ranges of youth they serve in order to meet the needs of youth with disabilities, who often need support during their transitions into adulthood (e.g., transition to independent living).

  • Introduction

    Worldwide there are an estimated 93 to 150 million children and youth with disabilities. This number is expected to rise given medical advancements that promote higher survival rates and life expectancy.2 Within the United States, there are approximately 6.7 million students aged 3 to 21 who receive special education services.3 Thirty-four percent of these students have a learning disability, 20 percent have a speech or language impairment, 9 percent have autism, 6 percent have a developmental delay, 14 percent have other health impairments, and the remainder face other types of physical disabilities.3, 4

    Young people with disabilities encounter many challenges and barriers to participating in society. For instance, they often experience social isolation and physical exclusion,5 are at risk of abuse and poor developmental outcomes, and are less equipped with the emotional, social, and cognitive resources to fully achieve positive life outcomes.2, 5 Many youth with disabilities also lack educational and employment opportunities.2, 6 For instance, youth with disabilities are underrepresented in higher education and have a lower probability of completing school than children without disabilities.7, 8 Furthermore, they are at risk of living below the poverty line9 and are more likely to encounter extreme social and economic disparities relative to youth without disabilities.4 Negative attitudes, discrimination, lack of resources and supports, and inaccessible environments contribute to these trends.10, 11

    The UN Convention on the Rights of Persons with Disabilities aims to enhance inclusion and participation of youth with disabilities toward realization of their human rights.12 Mentoring is one promising mechanism that could help achieve this goal by enhancing youth’s inclusion in society.13, 14, 15, 16, 17 Mentors can serve as role models and share experiences while helping to support youth in their academic, career, and psychosocial development5, 15, 18 and in their transition to adulthood. Mentors can help teach or advise youth, offer support and coping strategies, and help them to feel less alone.19

    Until recently, most mentoring programs did not include or specifically target youth with disabilities.14, 20 Therefore, the number of youth with disabilities in the United States who are engaged in mentoring is largely unknown. Studies focusing on mentoring for youth with disabilities show potential benefits on the transition to postsecondary education and employment,14, 21 self-esteem, social competence,22 and independent living skills.23 Having mentors for youth with disabilities also may be important for the development of social capital, self-determination, quality of life, and career and employment goals.14, 15, 21, 24, 25, 26, 27

    Scope of Review

    For this review, disability is defined as follows (using the World Health Organization’s definition)8: “Disability is an umbrella term covering impairments, activity limitations, and participation restrictions. An impairment is a problem in body function or structure; an activity limitation is a difficulty encountered by an individual in executing a task or action; while a participation restriction is a problem experienced by an individual in involvement in life situations. Disability is, therefore, not just a health problem. It is a complex phenomenon reflecting the interaction between features of a person’s body and features of the society in which they live.”

    Furthermore, the National Mentoring Resource Center defines mentoring as “relationships and activities that take place between youth (i.e., mentees) and older or more experienced persons (i.e., mentors) who are acting in a nonprofessional helping capacity, whether through a program or, more informally, to provide support that benefits one or more areas of the young person’s development” (for further details, see What is Mentoring?).

    Studies were included in which:

    1. Youth participants are age 25 and under, or the average age of the sample was 18 or under; or findings were delineated by age, with findings outlined for a subsample of youth 25 and under. This target age was expanded from that used in other NMRC reviews because youth with disabilities are often delayed in their transition to adulthood relative to youth without disabilities.28 We sought to include youth up to age 25 years to capture the “other side” of their transition to adulthood.

    2. At least 80 percent of study participants have a disability (using the World Health Organization’s8 definition of disability), or the authors conducted analyses that examine youth with disabilities as a distinct group.

    3. The study was a report of quantitative or qualitative empirical research with sufficient methodological detail included to be able to assess study rigor and findings.

    4. The study reported findings that bear on one or more of the four core questions for the review and examined either (a) an intentional, structured intervention or program involving mentoring, or (b) natural mentoring relationships occurring with youth with disabilities.

    Mental health conditions were excluded as a disability because an NMRC review has already been conducted on mentoring youth with mental health challenges.1 Temporary disabilities (e.g., cancer and the youth has fully recovered) and chronic illnesses/conditions that are not classified as a disability using the World Health Organization’s definition8 were also excluded. We also excluded chronic illness (e.g., chronic pain, diabetes) because a systematic review of peer support interventions for youth with chronic illness is already reported elsewhere.29 In addition, we excluded program descriptions and studies that did not have empirical findings.

    A literature search was conducted to identify potentially eligible journal articles, book chapters, and other types of reports, including searches of PubMed, ProQuest, PsycINFO, and Google Scholar, using an established set of keywords. Keywords used in the searches included “disability” (and a broad list of various types of physical, developmental, cognitive, and intellectual disabilities), “children and youth,” and “mentoring.” After two authors independently applied our inclusion criteria, we found a total of 40 studies addressing these questions with most of the studies focusing on youth aged 25 and under.

  • What are the Demonstrated Effects of Mentoring on Youth with Disabilities?


      There are several reasons to suggest that youth with disabilities could benefit from mentoring relationships. Youth with disabilities are a vulnerable population with unique social, developmental, educational, and vocational needs.14 Children with disabilities are more likely to report being victims of peer aggression and social exclusion because they often do not have the protective function of friendships.10, 30 They are bullied at disproportionally higher rates compared to youth without disabilities and are nearly three times as likely to experience social exclusion including limited social integration, fewer friends, and lower levels of friend support.10, 31 Given that they often encounter social isolation and exclusion, social support is an area of particular need for youth with disabilities. Therefore, mentoring may be a good resource for them to build friendships and other social networks that facilitate their development. Research also suggests that mentoring can improve academic and employment outcomes—areas of need for many youth with disabilities. This section presents findings on the potential benefits of mentoring for youth with disabilities.


      First, we discuss the following broad outcome areas that were explored in this review: academic and career development; employment; psychosocial health, quality of life, and protective factors; and transition and life skills. Next, we describe the types of mentoring models used within the studies found in our review.

      Academic and career development.

      Six studies in this review assessed the benefits of mentoring for academic outcomes and career development. For example, Kolakowsky-Hayner et al.32 used a pre-post survey to evaluate a community, group-based mentoring program (i.e., Back on Track to Success) to help 131 youth (aged 16 to 26) return to work and school after a brain or spinal cord injury. Participating youth reported that mentoring was beneficial for achieving postsecondary educational goals. In another study using a repeated measures design, Bell33 explored the effects of an online mentoring program for transition-age youth with blindness and found a significant increase in efficacy to make career-related decisions compared to their efficacy at the beginning of the program. Similarly, Kim-Rupnow and Burgstahler34 evaluated a community-based online mentoring program using a cross-sectional post-survey design and found a significant improvement in knowledge of career options. O’Mally and Antonelli18 used a longitudinal design to explore how a one-on-one career mentoring program benefited college students who were legally blind and found a pattern of improved career adaptability among participants over time. Burgstahler and Chang’s35 study assessed the impact of the Access STEM/DO-IT online program among youth with various types of disabilities. Using a case study design, they found that youth improved their career options (e.g., interest in STEM) over time. Finally, Powers et al.36 assessed the impact of an online group-based program on youth with various types of disabilities using a RCT (randomized controlled trial) design and reported significant improvements in educational planning among participants compared to controls.36


      Six studies in this review found that mentoring was associated with employment-related improvements, specifically improved knowledge of employment services and supports,37, 38 transition to employment,32, 37 knowledge of employment preparedness34 (e.g., the key skills needed to apply for a job), increased job-seeking self-efficacy and assertiveness in job hunting,18 as well as improvements in employability.39 Francis et al.37,38 used a mixed-method design (i.e., surveys and qualitative methods) to evaluate a group mentoring program serving youth with various types of disabilities. They found improved self-rated knowledge of employment services and supports and different types of competitive employment positions.37, 38 Kolakowsky-Hayner et al.32 evaluated a community-based group mentoring approach for youth with a brain or spinal cord injury. The researchers used a pre-post survey design and reported that promising numbers of program participants showed progress toward achieving their goals of postsecondary education, employment, and community independence. Another study focusing on youth with various types of disabilities34 involved an online group mentoring program and found significant improvements in perceived Internet and computer skills, career options, employment preparedness, perseverance, self-esteem, social skills, self-advocacy, and independence. O’Mally and Antonelli’s18 study focused on a one-to-one career mentoring program serving youth with vision impairments. Using a longitudinal design, they found that participants experienced increased job-seeking self-efficacy and career adaptability, and made significant gains in assertiveness in job hunting. Another study39 found that youth with spinal cord injury participating in an online, phone-based mentoring program experienced significant improvements in student education planning and transition awareness.

      Psychosocial health, quality of life, and protective factors.

      Fourteen studies included in this review found positive mentoring outcomes among youth with disabilities in areas which are important protective factors, including self-determination,25, 40 self-efficacy,18 social and emotional support,41, 42, 43 self-advocacy,25, 34, 42, 44 self-esteem,34, 45, 46 self-confidence,16, 47 and sense of community.39, 47 Studies showing benefits in these areas used a wide range of methods (e.g., quasi-experimental, pre-post, qualitative) and focused on youth with various disabilities including intellectual disability,40, 43, 45 vision impairments,18 pediatric liver transplant,41 deafness,42 autism,46 and various additional disabilities.34, 44, 48 Studies also included a wide range of mentoring models, such as one-to-one mentoring18, 41, 43 taking place at a college, out-patient or employment setting; group-based mentoring;34, 40, 44 e-mentoring;25, 34, 48, 49 naturally occurring mentoring;42, 45 and mixed models.34 We did not note any strong patterns in outcomes based on methodological design, type of disability, or type of mentoring.

      Eleven studies showed improvements in quality of life and social connections. Specifically, four studies41, 46, 50, 51 using various methodological designs (i.e., pre-post, case study, RCT) showed enhanced quality of life among youth with autism, brain injury, pediatric transplants, and other various types of disabilities (i.e., learning, emotional, behavioral). One study found an improved ability to manage their condition49 among those with juvenile arthritis. Studies focusing on youth with autism that used pre-post designs showed improvements in social anxiety46 and empathy.52 A qualitative study further found that role modeling42 helped youth with deafness. Improvements in social skills,34 social acceptance,47 social connectedness,46, 47, 53 ability to make new friends,22 and positive attitudes toward disability33 were seen for youth with autism, developmental disabilities, learning disabilities, physical disabilities, blindness, and other types of disabilities. These studies used survey, pre-post, and qualitative designs.

      Transition and life skills.

      Five studies using a range of designs (e.g., 3 RCTs, pre-post surveys) highlighted that mentoring yielded improvements in skills related to the transition to adulthood and life skills. Specifically, Powers et al.16, 36, 51 used an RCT design in three separate studies to assess the impact of mentoring programs for youth with various types of disabilities. They found significant differences between the treatment and comparison groups at post-intervention and/or follow-up in transition-related goals and planning, accessing transition services,51 engagement in independent living activities,51 and knowledge about strategies to promote independence,16 as well as significant improvements over time in program participants relative to the control group in transition awareness.36 Kolowsky-Hayner’s32 evaluation of a community, group-based mentoring program for youth with brain and spinal cord injury used a pre-post survey and found improvements in community independence32 over time.

      Studies within this review also reported on benefits of mentoring for the development of life skills. For example, Powers16 evaluated an online and in-person mentoring program for youth with physical disabilities, using an RCT design and found significant improvements in daily living skills (i.e., choice management, problem-solving)16, 54 compared to controls. Kramer et al.54 similarly found improvements in problem-solving over time for a one-to-one e-mentoring program for youth with intellectual and developmental disabilities. Another study evaluated an in-person, group-based mentoring program for youth with intellectual disabilities using a pre-post design and found significant improvements in self-regulation and assertiveness.40 Two studies focusing on youth with physical disabilities participating in online mentoring programs found significant improvements in typing skills55 as well as Internet and computer skills.34

      Types of mentoring models.

      Of the studies included within our review, five different types of mentoring models were studied. Two studies involved naturally occurring mentoring,42, 45 fifteen involved one-to-one mentoring, (see references 17, 18, 22, 24, 26, 41, 43, 46, 50, 53, 56, 57, 58, 59, 60) and eleven studied group-based mentoring programs(see 13, 16, 32, 34, 36, 37, 38, 40, 44, 51, 52). Online or e-mentoring was studied in 21 evaluations (see 13, 16, 25, 33, 34, 35, 36, 39, 47, 48, 49, 51, 54, 55, 58, 61, 62, 63, 64, 65, 66), revealing both the promise and relevance of this modality given that it helps to address many of the barriers that youth with disabilities often encounter in traveling to meet a mentor. Six studies had mixed models (i.e., combined approaches).13,16,34,36,51,58 The studies of naturally occurring mentoring focused on youth with deafness and youth with learning disabilities, whereas the one-to-one based mentoring interventions focused on youth with a wide variety of disabilities including intellectual, learning, and developmental disabilities; autism; blindness; pediatric transplants; acquired brain injury; and various other (i.e., mixed) types of disabilities. The group-based models focused on youth with physical and intellectual disabilities, acquired brain injury, spinal cord injury, autism, and various other types of disabilities. Finally, studies using an e-mentoring approach focused on youth with cerebral palsy; spina bifida; intellectual, learning, and developmental disabilities; blindness; spinal cord injury; juvenile arthritis; and various types of physical disabilities.


      1. Research on mentoring programs and interventions for youth with disabilities shows that there are potential benefits of mentoring on academics, employment, psychosocial health and quality of life, and transition-related and life skills.

      2. Given the various mentoring formats and disability types included in the studies, it is difficult to draw conclusions about what formats work best for which types of youth.

      3. The limited number of RCTs conducted and the various types of outcomes explored in studies to date only allows for tentative conclusions about the effectiveness of mentoring programs for youth with disabilities.

  • What Factors or Conditions Influence the Effectiveness of Mentoring for Youth with Disabilities?


      The impact of mentoring—both for youth with and without disabilities—can vary due to individual-level factors (e.g., gender, age, level of commitment), relationship-level factors (e.g., parent support), and program factors (e.g., duration).16, 67, 68 Also, the impact of mentoring has been thought to depend on program practices (e.g., training, supervision, characteristics of the mentor).69 For example, the impact of a mentoring program may be stronger if the program includes structured training with continued intermittent training and supervision67 or if it involves mentors who also live with disabilities.69 The impact may also depend on program location and organizational culture and climate.


      Research on children and youth with disabilities has not formally tested moderation (i.e., factors influencing the extent to which youth benefit), but qualitative and small-scale exploratory studies suggest some potential factors.

      Demographics and type of disability.

      For example, in one study of secondary and postsecondary students (n=189) with a university-defined disability (e.g., autism, learning disability), the effects of a virtual mentoring program to keep students involved in science, technology, engineering, and mathematics (STEM) were found to vary by type of disability and race/ethnicity.25 More specifically, minority students did not experience gains in self-determination that were apparent for nonminority students. Also, students with attention deficit hyperactivity disorder (ADHD) actually decreased in their math-related self-efficacy, whereas students without ADHD made gains in this area.25 Burgstahler and Chang61 also found gender differences in the perceived value of a mentoring program for students with disabilities. Male DO-IT program participants reported more interest, or saw more value, in STEM areas of career goals and financial security, while females reported more interest in program areas related to independent living.61 Thus, although no studies in our review formally examined moderators, these qualitative findings suggest that race/ethnicity, gender, and type of disability may influence program effectiveness and should be prioritized as potential moderators in future studies.


      A small group of studies suggest that communication between mentors and youth may influence the impact of mentoring: stronger communication may foster stronger program benefits. One study involving the DO-IT program found that youth particularly enjoyed “having conversations about their work plans for the future”.61 It was these types of conversations that fostered youth satisfaction and, thus, potentially influenced educational and employment outcomes.13 One e-mentoring study with a small group of mentoring dyads (n=9)53 reported that the type of communication style within the mentoring relationship affected its success.53 Another e-mentoring study found that unsuccessful mentoring pairs used a more formal and distant communication style, whereas successful dyads had mentors who used a more informal and supportive communication style.64

      Communication was also highlighted as important in an evaluation of a mentoring program for young adults with intellectual disabilities on a college campus (n=24 participants across three focus groups).57 One of the main themes in this study was that program effectiveness was perceived to be affected by the strength of communication and collaboration across stakeholders (e.g., with parents and professors).57 Combined with the findings highlighting the importance of strong mentor-mentee communication, these findings suggest that the effects of mentoring may be strengthened by strong communication on both the dyadic and programmatic levels.


      While the studies within our review did not formally assess factors influencing the effectiveness of mentoring, they suggest some potentially important factors and set the stage for the next phase of research. Next steps for the field should include formally examining the influence of some of these potential moderators.

      1. Gender and race/ethnicity of program participants may strengthen or weaken program effectiveness; however, research to understand whether mentoring programs should be designed to target youth with specific types of disabilities or whether a more generic approach could be similarly effective is lacking.

      2. Communication (e.g., strength, style) between mentor and mentee emerged as another potential factor that may affect outcomes of mentoring for youth with disabilities.

  • What Processes Are Most Important in Linking Mentoring to Outcomes for Youth with Disabilities?


      Numerous mentoring conceptual frameworks, models, or theories have proposed possible pathways through which mentoring can benefit youth (see Rhodes;70 Parra et al.71). In the field of mentoring for youth with disabilities, some researchers have begun to apply these theories and others (e.g., relational cultural theory) to their research but, to date, there is no single dominant theory for how mentoring impacts youth with disabilities. The studies reviewed in this section are suggestive of some of the important processes that may ultimately lead to positive youth outcomes, such as work, school, or relationship improvements. Where possible, we group studies together by mediator (i.e., the processes through which mentoring achieves its benefits) and/or type of disability.


      Among the studies within our review, there were no direct investigations of processes through which mentoring may influence outcomes for youth with disabilities. However, a group of empirical descriptive and qualitative studies provide some early evidence for important processes that youth experience in these relationships and that may be key in contributing to program effects. Qualitative data are often a first step in understanding how an intervention, such as mentoring, makes an impact. Future studies will need to test these potential pathways.

      Social processes.

      Mentoring relationships can have effects on social processes and relationships (e.g., improved social skills, improved relationships with parents or peers). In addition, the impact of mentoring on specific youth outcomes, such as employment and college success can be achieved through effects on other relationships, making these relationships an important process in linking mentoring with outcomes. For example, using data from the classic Big Brothers Big Sisters evaluation,68 Rhodes et al.70 found that the impact of mentoring on academic outcomes occurred, in part, through improved relationships with parents.

      One small qualitative study of 22 adolescents with physical disabilities, such as spina bifida or cerebral palsy, were connected with 5 mentors online for 25 sessions over 6 months.55 Youth’s responses to semistructured interviews postintervention suggest that they felt the program increased their social connections with other teens, reduced their feelings of loneliness, and increased their feelings of social acceptance.47 A second qualitative study with adolescents with autism spectrum disorders reported that both youth and other stakeholders (e.g., parents, mentors, staff) had improved social connectedness and willingness to take social risks.46 These social outcomes could certainly foster distal improvements in a wide range of areas.

      Learning processes.

      Another important process that research suggests could mediate the ultimate outcomes of mentoring is knowledge or learning. Two studies suggest that youth learn important information though mentoring. One study examined the use of peer mentors for youth in a cosmetology program and found improvements in work-related performance; one youth reported that a peer mentor provided them an opportunity to learn and “ask questions that14 they may have been hesitant to ask before working with (their mentors)”.58 There was learning taking place through the provision of the peer mentorship program, which focused on praise, corrective feedback, and demonstrations.58 These types of processes could then support further positive outcomes.

      Another mentoring program, the Family Employment Awareness Training (FEAT), focused on improving competitive employment for youth with disabilities and reported that expectations and knowledge improved through the program.37, 38 Relatedly, Barnard-Brak and colleagues24 conducted a study involving 43 high school students attending a one-to-one mentoring program aimed at improving academic outcomes for students with a variety of disabilities. Participation improved youth’s attitudes toward help-seeking (e.g., requesting accommodations).24 As stated earlier, theories have suggested cognitive, emotional, and modeling pathways to ultimate outcomes in mentoring, and some of these preliminary qualitative studies suggest that examining attitude change and enhanced knowledge may similarly inform our understanding of how mentoring influences outcomes for youth with disabilities.


      Some research suggests that self-determination is an outcome of mentoring participation. One study further suggests that it also may help to explain how mentoring achieves impacts on quality of life.51 “Take Charge” is a mentoring program for youth enrolled in special education and involved in the foster care system. An evaluation of this program reported that self-determination partially mediated, or explained, effects on enhanced quality of life for participants.51 Another study found that e-mentoring was empowering for youth with special needs.65 While the study did not test empowerment, or self-determination, as a potential mediator, it could be important in fostering other positive outcomes. Although the field of mentoring for youth with disabilities is in its infancy, these two studies together suggest that fostering the process of enhancing self-determination and empowerment may be important; however, research in this area is notably limited.

      Emotional support.

      Finally, emotional support is a common process that is addressed by mentoring programs for many different groups of youth, as it is a key process in youth development. One qualitative study on youth with hearing impairments reported that emotional support, in addition to advice-giving and role modeling, were important for mentees in achieving career success.42 Informal mentors provided a foundation for the mentees to break through common barriers to career success (e.g., lack of self-belief) for deaf youth. The relationships assisted these youth by advocating for additional needed support services and having faith and belief in them as they struggled to move forward in their lives.


      1. Potential areas for formal tests of mediation roughly map onto previous conceptual models of youth mentoring, namely a socioemotional mediating pathway, a cognitive pathway, and a modeling pathway.

      2. Overall, the field of mentoring interventions for children, youth, and young adults with disabilities needs to move beyond qualitative research to rigorously test potential mediators that have emerged as important in qualitative studies.

  • Have Mentoring Programs and Supports for Youth with Disabilities Reached Intended Youth, Been Implemented with High Quality, and Been Adopted and Sustained?


      Studies included in this review focused on mentoring programs designed specifically for youth with disabilities. These programs have shown some evidence of reach through their ability to enroll a targeted number of participants. However, with the exception of the DO-IT program, most of the mentoring programs discussed in this review have not been adopted on a wider scale. Our review indicates that relatively little is known about best practices for setting up a sustainable and effective mentoring program for youth with disabilities.


      Challenges in mentoring youth with disabilities.

      Several studies in this review highlighted challenges encountered by programs and the mentors they support when serving this population. Some mentors found it difficult to engage youth and to develop a rapport with them,72 particularly engaging younger mentees in career development conversations. This may have been a result of youth with disabilities often starting to think about employment and careers at a later stage compared to youth without disabilities.28 Communication style was highlighted in Shpigelman and Gill’s64 study where they noted that unsuccessful mentoring was associated with a more formal style and distant tone. Others similarly reported challenges common to mentoring other populations of youth—for example, that having a mismatch in the values, work styles, or personalities of the youth and mentor, combined with distancing behavior64 hindered communication. Pham26 found that building positive mentoring relationships requires sustained rather than time-limited or random efforts. Mentors need to think about communicating in a way that enhances trust and reduces feelings of alienation.26 This work suggests that additional and/or tailored training and support beyond that which is provided in more typical mentoring programs is needed when working with this population to ensure that mentors are adequately prepared for the unique challenges that these youth are experiencing.

      Other studies highlighted difficulties with the accessibility of the program setting,73 challenges in arranging transportation,17, 73 and/or overprotective parents.73 Other hurdles specific to mentoring youth with disabilities include that there is not a “one-size-fits-all” approach; mentoring programs should be designed specifically for the youth and type of disability that they are targeting.3 It is also difficult to assimilate persons from diverse backgrounds, needs, and abilities into one mentoring program.42 Barnfather et al.55 also noted that the age and ability level of the participants need to be considered when matching them. Others (e.g., Pham26) discovered that a program’s own evaluation efforts may be challenged given that youth with learning disabilities, autism, and intellectual disabilities had difficulties completing some of the self-assessment outcome measures. Bedell et al.74 found that youth with acquired brain injury had social participation barriers and fewer strategies for overcoming them compared to youth without disabilities. Because mentoring is an inherently social activity, some youth with disabilities may need support and resources beyond that which is provided by mentoring to help them achieve their goals. Other challenges noted by programs mentoring youth with disabilities include lack of mentor training, length of time for mentoring,17, 49, 64, 75 difficulty locating mentors,76 and challenges specific to e-mentoring (e.g., connectivity, security, privacy).25

      Reach and engagement.

      Very few studies in this review reported on reach and engagement (i.e., participation in the program). Of those that did, one study reported a mean engagement of 8.53/10 (i.e., participants self-reported on their engagement level with the program).49 The Kramer et al.54 study reported that mentees had high rates of attendance (87 percent) in peer mentoring calls and high rates of engagement within these calls, suggesting that mentoring is a promising approach for engaging this population. They also noted that some mentors with disabilities relied on a script to maintain engagement with participants.54 The highest rates of fidelity were achieved when addressing objectives related to participants’ unique interests and strengths or goals.54

      Powers et al. found that having experiential and hands-on activities (e.g., visiting colleges, shadowing professionals on job sites, touring STEM clubs and organizations, volunteering) helped to increase engagement in the mentoring relationship.72 Francis et al.37 further noted that having small group activities helped with engagement of participants. Others57 reported that clearly articulating the expectations of mentors at the outset helped with participant engagement. Requiring a social component (i.e., meeting face-to-face) is also a catalyst for more frequent, spontaneous, natural interactions, helping to foster a meaningful bond.

      Three studies highlighted that e-mentoring can help to reach and engage youth because it uses a convenient format, can reach youth in remote locations, and is anonymous.15, 25, 77 Gregg et al.,25, 48 for example, found that a collaborative use of online learning modules, in which mentors and participants met to complete these modules, was essential to participant engagement (i.e., participation in the program). Another aspect of engagement, noted in two studies,16, 78 was the importance of having family supports and engaging family members in the intervention.

      Quality of implementation.

      Six studies within this review reported that their mentoring program or intervention was feasible and acceptable to the participants.17, 22, 46, 54, 66, 74

      Adoption and sustainability.

      Some studies within this review reported on the adoption and sustainability of mentoring programs for youth with disabilities. For example, Stumbo et al.15 found that e-mentoring can be used to create and sustain a community that benefits both peers and mentors. These authors argue that engaging youth early on (i.e., beginning of college) can help to sustain their participation in mentoring over the longer term.15 Francis et al.37, 38 suggested that their program (FEAT) could be formatted as a professional development program for employment agencies to reach vocational rehabilitation counselors, job coaches, and other employment-related professionals. The authors also highlighted the potential for this program to expand to other states. They argue that expanding into schools would provide a sustainable foundation for teachers to empower their students.37, 38 Burghstahler and Crawford77 noted that the steps involved in sustaining an e-mentor community include: establishing goals for the program; selecting appropriate technology for the communication; developing the community structure; developing guidelines for protégés, mentors and parents; standardized procedures for recruiting, screening, and orienting participants; providing supervision and ongoing support of mentors; managing the mentor-mentee discussions; and evaluating the program.

      Furthermore, Kramer et al.54 emphasize that community-based organizations adopting e-mentoring should consider partnering with local colleges or vocational training institutions, which could help provide students with valuable hands-on experience and ensure that they also have access to qualified personnel.54 E-mentoring could help to sustain the program because of its ease of access.


      1. Several mentoring programs that are designed specifically for youth with disabilities appear to have successfully engaged substantial numbers of youth on a local level; however, most of these programs have not been adopted on a larger scale.

      2. Research on the factors influencing the adoption and longer-term sustainability of the programs is lacking.

      3. Challenges in mentoring youth with disabilities are similar to those found in mentoring programs for youth without disabilities, with the exception of the accessibility of the program.

  • Implications for Practice

    (Mike Garringer, MENTOR: The National Mentoring Partnership and Genelle Thomas, Partners for Youth with Disabilities)

    As noted in the review of the research presented on the preceding pages, there is considerable evidence that mentoring relationships can be beneficial to youth with disabilities in a wide range of aspects of their lives, including their education, careers, engagement with the community at large, and their own sense of identity, direction, and purpose. This evidence is especially important given the high prevalence of disability within the U.S. population. With one in five people experiencing some type of disability, it is a virtual guarantee that all mentoring programs are serving youth with disabilities (even when the program does not proactively recruit youth with disabilities). Therefore, whether a mentoring program explicitly engages youth with disabilities or whether the inclusion of youth with disabilities has occurred in a less intentional way, all mentoring programs should be prepared to adopt an inclusive approach to ensure that youth with disabilities are being served in a meaningful, equitable way.

    Despite the evidence of the value of mentoring for youth with disabilities, the review also notes challenges that practitioners can face in providing meaningful mentoring to these youth. Here we attempt to review some of the programmatic and relationship factors that can maximize the benefit of mentoring for youth with disabilities, building on the content of the review to support practitioners in developing inclusive, responsible, and meaningful mentoring services.


      A number of included studies discussed the challenges mentoring programs encountered with accessibility-related issues. The most foundational step a program can make around disability inclusion is to ensure a physically accessible environment. Program meetings and events should only be held in accessible locations and if transportation is provided, that transportation should be accessible to everyone. Examples of very basic accessibility include accessible bathroom facilities, clear signage, a level entrance to a building, accessible parking, meeting rooms with enough space for wheelchair access, and an elevator if a meeting is held above the first floor.

      Accessibility can also include less obvious (but still important) steps, such as having materials available in alternate formats, asking about and providing reasonable accommodations, and providing a low-stimulus area. To download a no-cost tip sheet and checklist, “Disability Inclusion Tips for Youth Sports and Recreation Programs,” go to:

      Online or electronically delivered mentoring models also need to pay attention to accessible design. Although they may not have physical spaces that youth and mentors visit in person, they certainly offer virtual spaces that need to be just as accessible. Please see section 5 on e-mentoring platforms below for further discussion about how virtual mentoring programs can ensure accessibility for all.

      In addition to physical accessibility, programs should ensure programmatic accessibility by understanding and using Universal Design for Learning principles. Universal Design for Learning (UDL) is a framework to improve and optimize teaching and learning for all people based on scientific insights into how humans learn. When programs take a proactive approach to developing trainings and activities, all youth participants are more engaged and retain more information. Programs can gain knowledge around UDL through CAST ( Programs and professionals can evaluate their own accessibility by accessing no-cost online inclusion self-assessments on the Partner’s for Youth with Disabilities (PYD) Pathways to Inclusion eLearning network. By registering on, individuals can measure their individual and organizational strengths and areas for growth in key areas, allowing them to focus on their greatest area of need for future training. Additional training materials can be found at no cost on the PYD website ( or through online courses on the Pathways to Inclusion network for a nominal fee. Readers should also note that there are UDL concepts for physical spaces as well, making this a principle that can be woven into all efforts to make a program’s physical, virtual, and educational spaces accessible and valuable to all.


      Included in this review were several examples of mentoring programs that intentionally served youth with disabilities at key transition points. This included examples related to educational and career transitions, as well as transitions out of services, such as leaving the child welfare system. Mentors can be tremendous assets in supporting transition planning and in helping mentees navigate other services and acclimate to new environments and routines. This can include transitions that are both sudden (e.g., the Back on Track to Success program32 that worked with youth who had experienced a spinal cord injury and needed help returning to familiar activities with new limitations), as well as those that are known well in advance (e.g., the work of Powers and colleagues focused on transitions to independent living as youth aged out of juvenile services).16, 51

      One key point related to transitions that practitioners and program developers should keep in mind is that these transition points often happen at later ages for youth with disabilities than they do for their peers. Due to the Individuals with Disabilities Education Act, some youth with disabilities choose to remain in high school until age 22, which extends their transition period compared to their peers without disabilities. Similarly, there may be delays in entering the workforce, living independently, or in other major life milestones. Programs may want to expand the age ranges they serve so that they can meet youth with disabilities where they are on their progression toward key milestones. Many service providers increase the upper limit of the age range they serve, with some including youth up to age 26 to reflect the reality of the timelines that youth with disabilities may experience. Unfortunately, many youth age out of “juvenile” services long before they are practically ready, so keep in mind that mentoring that supports these stressful transition points may be very beneficial to these youth.


      Many youth with disabilities face barriers to making independent decisions. Even well-intentioned supports in healthcare, education, and family life can leave youth feeling like they have little say in how they participate in the world and the paths that are open to them. Mentoring programs, and mentors, may be uniquely positioned to help young people with disabilities think about and strategize about life plans that are important to them. The research review offers several excellent examples of this type of programming in action, especially in the Take Charge36 and My Life51 programs, which support the transition out of the foster care system. These programs offer intensive transition planning and dogged pursuit of specific goals set by the youth (along with very little coaching from the mentors or staff about what is a “worthy” goal to pursue). Many of the features of these programs are designed to give the mentee the authority to guide activities, seek additional resources, and set timelines and milestones. This type of approach is often a breath of fresh air to youth who are used to authority figures telling them what their experiences will be based on their disability. Unsurprisingly, reports of self-determination predicted, in part, the other outcomes of the Take Charge program,36 particularly perceptions of overall quality of life.

      One simple way that programs can start a self-determination approach is to give youth with disabilities extensive say in who they are matched with. For example, some may want a mentor with a similar disability who can teach them how to overcome the specific barriers that their disability provides. Other youth with disabilities may want a mentor with a specific skill or who can connect them to career opportunities. What is important is that the program is not prescriptive in the type of mentor they offer the young person. By allowing these mentees to say, “This is what I want to achieve and here is who I want to help me get there,” programs are giving a gift that goes well beyond the support the mentor actually provides.

      Programs can also ask mentors to explicitly engage mentees in activities that help them envision potential directions for their lives and allow for focused goal setting and asset mapping. This type of activity can greatly improve mentee feelings of self-competence, agency, pride, and life satisfaction, even if they don’t reach their ultimate goals. A self-determination approach emphasizes the quality of the journey, not just the destination.


      Another skill mentors can teach that pairs well with a self-determination approach is the concept of self-advocacy. This empowers youth to stand up for themselves more effectively when interacting with institutions, such as schools or workplaces, and in their personal relationships, including with parents and other adults. Teaching youth to identify and respond to situations where their rights are discounted or when their decisions are negated will help them not only fight back against discrimination but also access the proper supports and resources to pursue their goals.

      This self-advocacy may be especially important for older youth who are on the cusp of aging out of services or special protections under the law. Youth with disabilities under the age of 18 are entitled to certain services, accommodations, and protections under the law, but upon entering adulthood are only eligible for services and protections and often have to request or even fight for them in the systems and institutions they will interact with in their young adulthood. Practicing these skills with a mentor can help prepare youth for situations where they need to push back against authority or protect their rights. It can also build self-confidence and feelings of self-worth.

      While a self-advocacy approach can be instrumental in building a sense of self and in achieving personal goals, mentoring programs may also want to encourage youth with disabilities to go beyond their own journey and engage in civic activities, advocacy, and activism that improves the well-being of all people with disabilities or other underserved groups. One of the most popular ideas in recent years in the youth mentoring field is that of “critical” mentoring, which builds on the concepts of critical race theory, pedagogy of place, and other critical perspectives to focus the impact of mentoring beyond the individual to larger communities and groups of people. Most notably, this work has been championed by academics like Torie Weiston-Serdan, whose seminal work Critical Mentoring: A Practical Guide argues that mentoring programs have an obligation to not only help youth cope with the negative impact of living in “toxic” environments (both literally and metaphorically), but to also help youth do transformative work at the community level in an effort to, as she phrases it, “clean the air and purify the water.”

      This approach may have particular appeal to youth with disabilities who, as noted above, have often experienced frustrations with institutions, agencies, service providers, and a society that generally is not inclusive and can neglect, if not outright ignore, their needs. They may find tremendous purpose and passion in advocacy or activism that helps address causes of systemic discrimination or disenfranchisement. They may also have passions that are totally unrelated to disability, but from which they have been excluded from having and acting on their voice. Mentors can be especially supportive in helping young people understand the root causes of systemic discrimination and underrepresentation and, in turn, develop strategies to combat these things in the real world. Youth may also find supportive peer relationships and a broader community by engaging in activism and other forms of civic engagement. So, while much of the mentoring journey should be focused on personal development and growth, programs are encouraged to remember that many youth often relish the opportunity to channel their passions to change the often dismissive world they were born into.


      Of note in this review are the numerous examples of online mentoring programs and the use of technology to supplement and support mentoring relationships that also meet face-to-face. Online communication platforms can help youth with disabilities overcome many barriers to accessing the help of a mentor, particularly those that involve limited physical mobility. The ability to communicate with a mentor without leaving home can give these youth another pathway to getting the support they need, particularly in instances where there are transportation barriers, such as a lack of accessible public transit, instances of inclement weather that disproportionately impact those with disabilities, or for isolated rural youth who can find it especially challenging to meet face-to-face.

      Increasingly, all youth, but especially youth with disabilities, are comfortable using digital platforms as the primary way of communicating. While this can be unfamiliar territory for older mentors, many youth today may prefer text-based communication to in-person meetings or even talking on the phone. Text-based communication can be helpful for youth who have trouble communicating orally because of a disability or who face anxiety bringing up certain subjects in person. In fact, many youth may prefer to discuss difficult or painful experiences and fears within the relatively safe space of a “chat,” where the distance between the participants can somewhat mask feelings of pain and frustration and where they have more control over the flow and depth of the conversation. Mentors working with youth with disabilities are encouraged to accept these communication alternatives and recognize that online platforms can actually enhance the relationship and the mentor-mentee bond rather than

      Online platforms also offer another advantage for youth with disabilities: access to a wider pool of mentors. Online chat groups and message boards can expose youth with disabilities to a chorus of supportive voices and other perspectives, which can be especially helpful in career exploration or transition-focused programs. This wider pool of mentors can help offset the impact of a mentor-mentee pair that is not “meshing” as intended, while also providing access to more social capital and networking opportunities that can help with career transitions or academic pursuits.

      Of course, one key to providing meaningful online mentoring opportunities to youth with disabilities is doing so on platforms that are designed with their needs and limitations in mind. Programs offering some form of e-mentoring should work with a competent designer who understands online disability issues and can ensure that the platform or technology will be accessible and easy to use for a variety of potential disabilities. This is yet another area where the principles of UDL can play a role in ensuring that technology platforms work not only for youth with disabilities but for all users. Common elements of accessible design include making text high-contrast with the background for low-vision or colorblind users and providing alternative text and transcripts for page elements such as images and any audio or video files. Once again, qualified designers can help ensure that all elements of online platforms will work with screen readers and other assistive technology, meeting the needs of all users.

      One good example of a well-designed platform that is not only functional for youth with disabilities, but also hits on some of the additional benefits of e-mentoring noted here, is the Campus Career Connect platform ( developed by Partners for Youth with Disabilities. Campus Career Connect (C3) was created to aid transitioning young adults with disabilities from school to work and connect them to mentors within their desired career field. By promoting job readiness, inclusion, and advocacy training and advice, C3 mentors help make the transition from school to employment positive and socially impactful. Mentoring on C3 can be found through the platform’s use of online events, local job listings, networking, résumé building, soft- and hard-skill coaching, and an interactive forum space for questions and advice. C3 was designed by Thunder Media and was created to be fully accessible and meet the standards set by the Web Content Accessibility Guidelines (WCAG).


      It goes without saying that all young people in mentoring programs deserve a relationship that is safe, supportive, and aligned with their needs and dreams. But youth with disabilities may be especially sensitive to experiences that fall short of this type of responsiveness. They may have a long history with “helping” services that are anything but helpful, and may be especially sensitive to feelings of rejection, bullying, and isolation from their peers. This means that mentors who are paired with youth with disabilities must possess some special characteristics that reduce potential harm and allow them to effectively serve youth with disabilities:

      • They must commit to sticking with the match through all the ups and downs they may experience. All mentors commit to this at some level, but youth with disabilities might especially need someone to be a stable, constant, unwavering presence in their lives. This makes mentors who are likely to be mobile in their lives, or who seem unsure about taking on the challenge, a poor fit for mentoring a young person with a disability. These mentors also must be patient, especially with the possible extended transition to adulthood noted earlier. Programs must emphasize the nonnegotiable nature of meeting frequency and longevity of these relationships to prospective mentors.

      • They must commit to learning about their mentee’s disability and the impact that disability has on their life and their pursuit of goals. This includes becoming aware of how to talk about the disability, either with the youth directly or with others, in ways that do not add to the stigma that these youth may already be experiencing. Mentors can follow the youth’s lead in how they talk about and respond to disability within the relationship—an approach that can empower youth and allow them to determine what the relationship looks like. Programs can support mentors in this endeavor by providing trainings and learning materials related to disability etiquette and inclusive communication.

      • They must have the capacity to express empathy, understanding, and compassion, while also challenging their mentee to grow and expand their horizons. This is a delicate balancing act, but mentors can be instrumental in encouraging mentees with disabilities to try something new or to take a risk—something other adults in their life may have discouraged. Mentors also have to be able to express empathy and understanding for those times when the youth may face barriers related to their disability. And most critically for mentors who themselves have a disability: they must be willing to talk about their own journey. While the research noted in this review is unclear on whether youth benefit more from having a mentor with a shared disability, there were qualitative examples in the literature (most notably Powers and colleagues36, 51) where having a mentor who was willing to share their personal journey of overcoming adversity related to their disability was absolutely critical in helping the young person feel hopeful and inspired for the hard work that may lay ahead for them.

    • Resources

    • References

      1. Munson, R., & Railey, J. (2016). Mentoring for youth with mental health challenges. Boston, MA: National Mentoring Resource Center.

      2. UNICEF (2013). State of the world's children: Children with disabilities. Retrieved from

      3. National Centre for Education Statistics. (2018). Children and youth with disabilities. Retrieved from

      4. United Nations. (2015). Youth with disabilities. Retrieved from

      5. McDonald, K., Balcazar, F., & Keys, C. (2005). Youth with disabilities. In D. L. DuBois & M. J. Karcher (Eds.) Handbook of Youth Mentoring (pp. 493–507). Thousand Oaks, CA: SAGE.

      6. Lindsay, S., McPherson, A., & Maxwell, J. (2017). Perspectives of school-work transitions among youth with spina bifida, their parents and health care providers. Disability & Rehabilitation, 39, 641–652. doi:10.3109/09638288.2016.1153161

      7. Organisation for Economic Co-operation and Development (OECD). (2015). Education at a glance. Retreived from

      8. World Health (2015). Disabilities. Retrieved from

      9. Mitra, S., Posarac, A., & Vick, B. (2013). Disability and poverty in developing countries: A multi-dimensional study. World Development, 41, 1–18. doi:10.1016/j.worlddev.2012.05.024

      10. Lindsay, S., & McPherson, A. (2012). Experiences of social exclusion and bullying at school among children and youth with cerebral palsy. Disability & Rehabilitation, 34, 101–doi:10.3109/09638288.2011.587086.

      11. Lindsay, S. (2011). Discrimination and other barriers to employment for teens and young adults with disabilities. Disability & Rehabilitation, 33, 1340–1350. doi:10.3109/09638288.2010.531372

      12. United Nations General Assembly. (2006). Convention on the rights of persons with disabilities (A/RES/61/106). Retrieved from

      13. Burgstahler, S., & Cronheim, D. (2001). Supporting peer-peer and mentor-protege relationships on the internet. Journal of Research in Technology and Education, 34, 59–74. doi:10.1080/15391523.2001.10782334

      14. Lindsay, S., L, Hartman, R., & Fellin, M. (2016). A systematic review of mentorship programs to facilitate transition to postsecondary education and employment for youth and young adults with disabilities. Disability & Rehabilitation, 38, 1329–1349. doi:10.3109/09638288.2015.1092174

      15. Stumbo, N., Martin, J., Nordstrom, D., Rolfe, T., Burgstahler, S., Whitney, J., … & Misquez, E. (2010). Evidence-based practices in mentoring students with disabilities: Four case studies. Journal of Science Education for Students with Disabilities, 14, 33–54. doi:10.14448/jsesd.03.0003

      16. Powers, L., Sowers, J., & Stevens, T. (1995). An exploratory randomized study of the impact of mentoring on the self-efficacy and community-based knowledge of adolescents with severe physical challenges. Journal of Rehabilitation, 61, 33–41.

      17. Wilson, N., Cordier, R., Ceccarelli, M., MacCallum, J., Melbourne, B., Vaz, S., ... & Stancliffe, R. (2018). Intergenerational mentoring at Men’s Sheds: A feasibility study. Journal of Applied Research in Intellectual Disabilities, 31, 2105–e2117. doi:10.1111/jar.12338

      18. O'Mally, J., & Antonelli, K. (2016). The effect of career mentoring on employment outcomes for college students who are legally blind. Journal of Visual Impairment and Blindness, 110, 295–307.

      19. Shpigelman, C., & Gill, C. (2009). A conceptual framework for electronic socioemotional support for people with special needs. Journal of Rehabilitation Research, 32, 301–doi:10.1097/MRR.0b013e32831e4519.

      20. Sword, C., & Hill, K. (2003). Creating mentoring opportunities for youth with disabilities: Issues and suggested strategies. American Rehabilitation, 27, 14–18.

      21. Buckner, A. (1993). Mediating at-risk factors among seventh and eighth grade students with specific learning disabilities using a holistically based model. (Doctoral dissertation). Retrieved from ERIC Document Reproduction Service. (Accession No. 368122).

      22. Muscott, H., & O'Brien, S. (1999). Teaching character education to students with behavioral and learning disabilities through mentoring relationships. Education & Treatment of Children, 22, 373–390.

      23. West, M., Targett, P., Steininger, G., & Anglin, N. (2001). Project corporate support (CORPS): A model demonstration project on workplace supports. Journal of Vocational Rehabilitation, 16, 111–118.

      24. Barnard-Brak, L., Schmidt, M., Wei, T., Hodges, T., & Robinson, E. (2013). Providing postsecondary transition services to youth with disabilities: Results of a pilot program. Journal of Postsecondary Education and Disability, 26, 135–144.

      25. Gregg, N., Galyardt, A., Wolfe, G., Moon, N., & Todd, R. (2017). Virtual mentoring and persistence in STEM for students with disabilities. Career Development and Transition for Exceptional Individuals, 40, 205–214. doi:10.1177%2F2165143416651717

      26. Pham, Y., & Murray, C. (2016). Social relationships among adolescents with disabilities: Unique and cumulative associations with adjustment. Exceptional Children, 82, 234–250. doi:10.1177/0014402915585491

      27. Cawthon, S., Johnson, P., Garberolgio, C., & Schoffstall, S. (2016). Role models as facilitators of social capital for deaf individuals: A research synthesis. American Annals of the Deaf, 161, 115–127. doi:10.1353/aad.2016.0021

      28. Lindsay, S., McDougall, C., Sanford, R., Menna-Dack, D., Kingsnorth, S., & Adams, T. (2015). Exploring employment readiness through mock job interview and workplace role-play exercises: Comparing youth with physical disabilities to their typically developing peers. Disability & Rehabilitation, 37, 1651–1663. doi:10.3109/09638288.2014.973968

      29. Kohut, A., Stinson, J., van Wyk, M., Giosa, L., & Luca, S. (2014). Systematic review of peer support interventions for adolescents with chronic illness: A narrative synthesis. International Journal of Child and Adolescent Health, 7, 183–197.

      30. Gini, G., & Pozzoli, T. (2009). Association between bullying and psychosomatic problems: A meta-analysis. Pediatrics, 123, 1059–1065. doi:10.1542/peds.2008-1215

      31. Dixon, R. (2006). A framework for managing bullying that involves students who are deaf or hearing impaired. Dear Education International, 8, 11–32. doi:10.1002/dei.16

      32. Kolakowsky-Hayner, S., Wright, J., Shem, K., Medel, R., & Duong, T. (2012). An effective community-based mentoring program for return to work and school after brain and spinal cord injury. NeuroRehabilitation, 31, 63–73. doi:10.3233/NRE-2012-0775

      33. Bell, C. (2012). Mentoring transition-age youth with blindness. Journal of Special Education, 46, 170–179. doi:10.1177%2F0022466910374211

      34. Kim-Rupnow, W., & Burgstahler, S. (2004). Perceptions of students with disabilities regarding the value of technology-based support activities on postsecondary education and employment. Journal of Special Education Technology, 19, 43–56. doi:10.1177/016264340401900204

      35. Burgstahler, S., & Chang, C. (2007). A preliminary report of the AccessSTEM/DO-IT longitudinal transition study.

      36. Powers, L., Turner, A., & Westwood, D. (2001). Take charge for the future: A controlled field-test of a model to promote student involvement in transition planning. Career Development and Transition for Exceptional Individuals, 2, 89–104. doi:10.1177%2F088572880102400107

      37. Francis, G., Gross, J., Turnbull, A., & Turnbull, R. (2013). The Family Employment Awareness Training (FEAT): A mixed-method follow-up. Journal of Vocational Rehabilitation, 39, 167–181. doi:10.3233/JVR-130652.

      38. Francis, G., Gross, J., Turnbull, R., & Parent-Johnson, W. (2013). Evaluating the effectiveness of the family employment awareness training I'm Kansas: A pilot study. Research in Practice Persons Severe Disability, 38, 44–57.

      39. Shem, K., Medel, R., Wright, J., Kolakowsky-Hayner, S., & Duong, T. (2011). Return to work and school: A model mentoring program for youth and young adults with spinal cord injury. Spinal Cord, 49, 544–548. doi:10.1038/sc.2010.166

      40. Abery, B., Rudrud, L., & Arndt, K. (1995). Evaluating a multicomponent program for enhancing the self-determination of youth with disabilities. Intervention in School and Clinic, 30, 170–179. doi:10.1177/105345129503000307

      41. Jerson, B., D'Urso, C., Arnon, R., Miloh, T., Iyer, K., Kerkar, N., & Annuziato, R. (2013). Adolescent transplant recipients as peer mentors: A program to improve self-management and health-related quality of life. Pediatric Transplantation, 17, 612–620. doi:10.1111/petr.12127

      42. Foster, S., & MacLeod, J. (2004). The role of mentoring relationships in the career development of successful deaf persons. Journal of Deaf Studies and Deaf Education, 9, 442–458. doi:10.1093/deafed/enh053

      43. Scholl, L., & Mooney, M. (2003). Youth with disabilities in work-based learning programs: Factors that influence success. Journal for Vocational Special Needs Education, 26, 4–16.

      44. Grenwelge, C., & Zhang, D. (2013). The effects of the Texas youth leadership forum summer training on the self-advocacy abilities of high school students with disabilities. Journal of Disability Policy Studies, 24, 158–169. doi:10.1177/1044207312457415

      45. Ahrens, K., DuBois, D., Lozano, P., & Richardson, L. (2010). Naturally acquired mentoring relationships and young adult outcomes among adolescents with learning disabilities. Learning Disabilities in Research and Practice, 25, 207–216. doi:1111/j.1540-5826.2010.00318.x

      46. Curtin, C., Humphrey, K., Vronsky, K., & Mattern, K. (2016). Expanding horizons: A pilot mentoring program linking college/graduate students and teens with ASD. Clinical Pediatrics, 55, 150–156. doi:10.1177/0009922815588821

      47. Stewart, M., Barnfather, A., Magill-Evans, J., Ray, L., & Letourneau, N. (2011). Brief report: An online support intervention: Perceptions of adolescents with physical disabilities. Journal of Adolescence, 34, 795–800. doi:10.1016/j.adolescence.2010.04.007

      48. Gregg, N., Wolfe, G., Jones, S., Todd, R., Moon, N., & Langston, C. (2017). STEM e-mentoring and community college students with disabilities. Journal of Postsecondary Education and Disability, 29, 47–63.

      49. Stinson, J. S., Kohut, S., Forgeron P., Amaria, K., Bell, M., Kaufman, M., Luca, N., Luca, S., Harris, L., Victor, C., & Spiegel, L. (2016). The iPeer2Peer Program: A pilot randomized controlled trial in adolescents with juvenile idiopathic arthritis. Pediatric Rheumatology, 14, 48. doi:10.1186/s12969-016-0108-2

      50. Fraas, M., & Bellerose, A. (2010). Mentoring programme for adolescent survivors of acquired brain injury. Brain injury, 24, 50–61. doi:10.3109/02699050903446781

      51. Powers, L., Geenen, S., & Powers, J. (2012). My life: Effects of a longitudinal, randomized study of self-determination enhancement on the transition outcomes of youth in foster care and special education. Child and Youth Services Review, 34, 2179–2187. doi:10.1016/j.childyouth.2012.07.018

      52. Hillier, A., Fish, T., Cloppert, P., & Beversdorf, D. (2007). Outcomes of a social and vocational skills support group for adolescents and young adults on the autism spectrum. Focus on Autism and other Developmental Disorders, 22. doi:10.1177%2F10883576070220020201

      53. Ryan, C., Kramer, J., & Cohn, E. (2016). Exploring the self-disclosure process in peer mentoring relationships for transition-age youth with developmental disabilities. Intellectual and Developmental Disabilities, 54, 245–249. doi:10.1352/1934-9556-54.4.245.

      54. Kramer, J., Ryan, C., Moore, R., & Schwartz, A. (2018). Feasibility of electronic peer mentoring for transition‐age youth and young adults with intellectual and developmental disabilities: Project teens making environment and activity modifications. Journal of Applied Research in Intellectual Disabilities, 31, e118–e129. doi:10.1111/jar.12346

      55. Barnfather, A., Stewart, M., Magill-Evans, J., Ray, L., & Letourneau, N. (2011). Computer-mediated support for adolescents with cerebral palsy or spina bifida. Computers and Informatics in Nursing, 29, 24–33. doi:10.1097/NCN.0b013e3181f9db63.

      56. Bobroff, S., & Sax, C. (2010). The effects of peer tutoring interview skills training with transition-age youth with disabilities. Journal of Vocational Rehabilitation, 33, 143–157. doi:10.3233/JVR-2010-0523

      57. Jones, M., & Goble, Z. (2012). Creating effective mentoring partnerships for students with intellectual disabilities on campus. Journal of Policy and Practice on Intellectual Disabilities, 9, 270–278. doi:10.1111/jppi.12010

      58. Westerlund, D., Granucci, E., Gamache, P., & Clark, H. (2011). Effects of peer mentors on work-related performance of adolescents with behavioral and/or learning disabilities. Journal of Positive Behavioral Interventions, 8, 244–251. doi:10.1177%2F10983007060080040601

      59. Seeger, K. (2007). Mentoring youth with disabilites: The mentor’s lived experiences. (Master’s thesis). Retreived from Louisiana State University Digital Commons. (#2187).

      60. Schwartz, A., & Kramer, J. (2018). “I just had to be flexible and show good patience”: Management of interactional approaches to enact mentoring roles by peer mentors with developmental disabilities. Disability & Rehabilitation, 40, 2364–2371. doi:10.1080/09638288.2017.1334835.

      61. Burgstahler, S., Chang, C. (2007). Gender differences in perceived value of components of a program to promote academic and career success for students with disabilities. Journal of Science Education for Students with Disabilities, 12, 1–11.

      62. Burgstahler, S., Doyle, A. (2005). Gender differences in computer-mediated communication among adolescents with disabilities: Science, technology, engineering, and mathematics case study. Disability Studies Quarterly, 25.

      63. Stumbo, N., & Lindahl-Lewis, P. (2008). Two mentorship case studies of high school and university students with disabilities: Milestones and lessons. Journal of Rehabilitation, 74, 45–51.

      64. Shpigelman, C., & Gill, C. (2012). The characteristics of unsuccesful e-mentoring relationships for youth with disabilities. Qualitative Health Research, 23, 463–475. doi:10.1177/1049732312469115.

      65. Shpigelman, C., Weiss, P., & Reiter, S. (2009). E-mentoring for all. Computers in Human Behaviour, 25, 919–928. doi:10.1016/j.chb.2009.03.007

      66. Kramer, J., Hwang, I., Helfrich, C., & Carrellas, A. (2018). Problem-solving intervention to teach transition age youth with developmental disabilities to resolve environmental barriers. International Journal of Disability, Development and Education, 65, 57–75. doi:10.1080/1034912X.2017.1346237

      67. DuBois, D., & Karcher, M. (2014). Handbook of Youth Mentoring. Thousand Oaks, CA: SAGE.

      68. Grossman, J., & Rhodes, J. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219. doi:10.1023/A:1014680827552

      69. Sowers, J., Powers, L., Schmidt, J., Keller, T., Turner, A., Salazar, A., & Swank, P. (2017). A randomized trial of a science, technology, engineering, and math mentoring program. Career Development and Transition for Exceptional Individuals, 40, 196–204. doi:10.1177%2F2165143416633426

      70. Rhodes, J., Grossman, J., & Resch, N. (2000). Agents of change: Pathways through which mentoring relationships influence adolescents academic adjustment. Child Development, 71, 1662–1671. doi:1111/1467-8624.00256

      71. Parra, G., DuBois, D., Neville, H., Pugh-Lilly, A., & Povinelli, N. (2006). Mentoring relationships for youth: Investigation of a process-oriented model. Journal of Community Psychology, 30, 367–388. doi:10.1002/jcop.10016

      72. Powers, L., Schmidt, J., Sowers, J., & McCracken, K. (2015). Qualitative investigation of the influence of STEM members on youth with disabilities. Career Development and Transition for Exceptional Individuals, 38, 25–38. doi:10.1177/2165143413518234

      73. Froschl, M., Rousso, H., & Rubin, E. (2001). Nothing to do after school: More of an issue for girls. In H. Rousso & M. Wehmeyer (Eds.), Double jeopardy: Addressing gender equity in special education (pp. 313–336). Albany, NY: State University of New York Press.

      74. Bedell, G., Wade, S., Turkstra, L., Haarbauer-Krupa, J., & King, J. (2016). Informing design of an app-based coaching intervention to promote social participation of teenagers with traumatic brain injury. Developmental Rehabilitation, 20, 408–417. doi:10.1080/17518423.2016.1237584

      75. Hayes, E., & Balcazar, F. (2008). Peer-mentoring and disability: Current applications and future directions. In T. Kroll (Ed.), Focus on Disability: Trends in Research and Application (pp. 89–108). New York, NY: Nova Science Publishers.

      76. Daughtry, D., Gibson, J., & Abels, A. (2009). Mentoring students and professionals with disabilities. Professional Psychology: Research and Practice, 40, 201–205. doi:10.1037/a0012400

      77. Burgstahler, S., & Crawford, L. (2007). Managing an e-mentoring community to support students with disabilities: A case study. Association for Advancement of Computing in Education Journal, 15, 97–114.

      78. Luecking, R., & Wittenburg, D. (2009). Providing support to youth with disabilities transitioning to adulthood: Case descriptions from the youth transition demonstration. Journal of Vocational Rehabilitation, 30, 241–251. doi:10.3233/JVR-2009-0464

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.

PDF button Facebook button Twitter button

October 2017

This review examines research on digital and electronic forms of mentorship, or e-mentoring. The review is organized around four questions:

  1. What is the documented effectiveness of this approach to mentoring?

  2. What factors shape the effectiveness of e-mentoring among youth?

  3. What are the intervening processes (mediators) that are most important in linking e-mentoring to youth outcomes?

  4. To what extent have e-mentoring programs reached and sustained the engagement of intended youth, been implemented with high quality, and been adopted and sustained by organizations and settings?

To date, only a small number of empirical studies address these questions. With widespread use of digital and electronic communication being fairly new, however, it is not surprising that e-mentoring is a relatively under-investigated area. The research that does exist reveals the following preliminary findings:

  • Evidence on the effectiveness of e-mentoring for improving youth outcomes is mixed, as in some effects are good and some are null; the limited number of studies that utilize a comparison group not receiving e-mentoring further complicate the ability to draw conclusions about its effectiveness.
  • Although some e-mentoring formats, such as email interactions, have been successful in improving youth outcomes, it is not clear which formats work best for a given population of youth.
  • Although there are several potential factors that could moderate the effects of e-mentoring, including race and gender of youth, most studies to date have only explored level and quality of interpersonal communication.
  • Interaction frequency and relationship quality may be important mediators of youth outcomes in e-mentoring programs.
  • E-mentoring programs that have been implemented and sustained seem to benefit from clear guidelines, structure, and organizational tools.

Insights for practitioners are provided at the end of this review. This commentary recommends that programs wishing to adopt and utilize electronic communication in their mentoring programs should always consider factors that would enable mentors and mentees to use the associated tools (e.g., mobile devices, websites). More specifically, programs are advised to clearly articulate why and how e-mentoring formats can facilitate or enhance mentor-mentee interactions, anticipate potential challenges and how to overcome them, and determine how staff roles may change to facilitate and support electronic communications.

  • Introduction

    Electronic, online, or digital mentoring, often referred to collectively as e-mentoring, has grown in popularity over recent decades. This is particularly true as social media, text messaging, and online communications such as chat functions or emails have become popular (and sometimes primary) forms of communication, especially among young people. E-mentoring requires some form of information and communication technology (ICT), such as an Internet-connected computer, smartphone, or tablet. There has been a rapid growth in diverse technology-mediated forms of communication in both formal and informal contexts. For instance, computer-mediated communication, which refers to email, LISTSERVs, chat groups, and computer conferencing,1 is now a normal part of both professional and personal interactions around the globe. This term may feel a little dated, however, as many of these types of communications now occur with the use of smartphones and other mobile devices, not just computers. Another example of such communications are social media forums, which are online discussion environments that offer a safe space, friendship within the group, flexible help, and peer support for recovery and relapse prevention, such as for those with eating disorders.2 These interactions sometimes can be classified as examples of peer e-mentoring, especially when the communicators share characteristics such as illness, a social role such as mothers, or similar career types. Another type of digital platform is app-mediated communication, which is conducted through popular apps such as WhatsApp, Snapchat, Kik, or Viber.

    The Significance of E-Mentoring

    Why is e-mentoring so compelling? For one, in the past few decades, ICTs have transcended the geographical and psychological distance between people.3 This type of communication allows individuals who live at a far distance, and who perhaps will never meet, to communicate and build a relationship. In the context of mentoring, this means mentor pairs or groups are not bounded by geographical location.4 It also allows communicators to overcome constraints associated with in-person meetings, such as scheduling or travel.5 E-mentoring could allow for a mentoring pair (or group) to build cohesiveness at a distance, thereby building social capital for young people who do not have it in close physical proximity.6 Mentoring programs increasingly are designed to take advantage of these e-mentoring strategies for a variety of desirable purposes, such as mentoring for youth in rural areas, those with chronic illness, those interested in pursuing certain professions or higher education, and those with disabilities.

    Scope of Review

    For this review, e-mentoring refers to mentoring conducted entirely or in part using electronic communication, such as email, text, social media, messaging applications, or computer platforms. This also includes the use of technology to support and/or enhance in-person mentoring relationships (for example, using email communications to stay in touch between in-person meetings or to share resources). The focus of this review is on e-mentoring as a supported modality for mentor-mentee communication within formal mentoring programs for youth (up to ages 18). Research on e-mentoring occurring in informal mentoring relationships is considered, however, as background where available. For the purposes of this review, e-mentoring also does not include the use of web or mobile device-based resources intended for joint use by mentor and mentee, or web/device-based mentor or staff training modules. For example, if a mentoring pair uses a website to practice math skills while together in person, this is not included as e-mentoring, but if they use a website/mobile app to interact and keep in touch in between in-person visits, that is included.

    The present review examines the available research using the above definition to answer the following four questions:

    1. What is the documented effectiveness of e-mentoring for youth?

    2. What factors shape the effectiveness of e-mentoring among youth?

    3. What are the intervening processes (mediators) that are most important in linking e-mentoring to youth outcomes?

    4. To what extent have e-mentoring programs reached and sustained the engagement of intended youth, been implemented with high quality, and been adopted and sustained by organizations and settings?

    A literature search was conducted to identify journal articles, book chapters, and other types of reports and unpublished work that report findings pertinent to one or more of these questions. This search was conducted using library search engines, such as PsychINFO, Springer, and ScienceDirect, as well as Google Scholar. Keywords used in the searches included children, youth, e-mentoring, online mentoring, computer-mediated communication, and social media mentoring. Additionally, research referenced in relevant chapters and prior literature reviews in the area were reviewed for potential relevance. A total of 20 studies met criteria for inclusion in the review.

  • What Is the Documented Effectiveness of E-Mentoring for Youth?


      With the rise of digital and social media as important forms of modern communication, it is natural to think they could play a useful role in mentoring relationships. Social media interactions, for instance, have been shown to be a source of or vehicle for social support, especially for those with weaker in-person support.7 Online support has been found to be predictive of lower incidence of depressive thoughts, and there is evidence to suggest it can buffer the effects of peer victimization.8 E-mentoring is also of interest because many existing programs were developed as a means to compensate when naturally occurring mentoring relationships were unavailable and when opportunities to participate in traditional mentoring programs (such as in-person meetings) were not possible. E-mentoring programs can create educational and vocational opportunities for disadvantaged or underrepresented populations.9

      Findings from studies of mHealth (or mobile-device-led health interventions) for youth suggest that ICTs, particularly mobile phones, are an effective way to reach young people and to increase their knowledge and produce behavior change for health-related outcomes. mHealth interventions for youth have been shown to be effective in promoting behaviors such as adherence to medications,10 self-management of type 1 diabetes,11 and utilization of sexual and reproductive health services.12 Today’s adolescents have ubiquitous access to mobile technology, and this is true across social status and location. These successes with mHealth could be used as a foundation for applying technology to youth mentoring.

      The challenges of e-mentoring.

      E-mentoring comes with its own set of challenges, however. On the most practical level, it requires access to ICTs, including computers or mobile devices (e.g., smartphones or tablets) and technical support for the technology and digital platform. The chosen technology also must be accessible to all mentors and mentees, which may be particularly challenging when working with specific populations, such as youth with disabilities. Mentors and mentees participating in an e-mentoring program also must be technology literate. If a mentor is not familiar with social media platforms, for example, using them to build a mentoring relationship may not be productive without sufficient training. Mentors and mentees also need sufficient ICT communication skills, such as reading comprehension and the ability to sufficiently express oneself through text and/or emojis (digital images used to express an idea or emotion in digital communication).9

      Vast opportunities for e-mentoring practice.

      Despite these challenges, there is a tremendous opportunity to use ICTs in mentoring.13 E-mentoring programs utilized with some college-age populations (beyond the age range for this review) have found that personal and emotional interactions often develop between mentors and mentees, especially if the pair also meets face-to-face and student mentees do not have an alternative support network,14 or if the interaction via electronic means is a mandatory part of the relationship.5


      Seven studies included in this review reported findings that address the potential effectiveness of e-mentoring for improving youth outcomes. Five of these studies looked at the use of online interactions only, whereas the remaining two combined face-to-face meetings with online interactions.

      One of the earliest studies of e-mentoring looked at interaction between mentors and mentees via email only in a program called the Digital Heroes Campaign.15 In this program, youth were matched with online mentors over a two-year period. Using a mixed methods approach (surveys, interviews, focus group discussions, email transcripts), researchers assessed the nature, types, and quality of relationships that developed in the program. They found that youth and mentors both perceived a positive impact as a result of the program; however, deep connections between mentors and mentees were relatively rare.

      Another early pilot study examined whether or not e-mentoring had an academic and psychological impact on 32 high school students who were at risk of dropping out.16 Each student was matched randomly with a volunteer adult mentor recruited from the schools’ business and educational partners. The pairs never met face-to-face—all communication was through email. In a comparison of program participants to nonparticipants, there were no significant differences between the two groups on self-esteem, career indecision, attendance, or academic achievement. However, rich dialogue occurred between the students and mentors, which suggested the program merited further exploration.

      One of the more well-known e-mentoring programs for youth is the iMentor College Ready curriculum.17, 18 This program uses a “blended” approach to mentoring—ninth grade mentees communicate via email and meet face-to-face with college-educated mentors, as well as participate in weekly college preparatory classes. In a recent evaluation of tenth grade students in the program in New York City schools, programmatic and survey data showed that those students who were in the iMentor program scored higher than comparison students after one academic year on measures of interpersonal support, future planning, college aspirations, and career planning. The evaluation did not, however, show evidenced effects for five noncognitive outcomes, grade point average, chronic absenteeism, or the percent of students on track for graduation.

      Another study looked at the effectiveness of a one-year online mentoring program in Germany for girls ages 11 to 18 in science, technology, engineering, and math (STEM) called CyberMentor.19 Female mentors and mentees communicated via email, online chats, and forums. In the evaluation, girls were assigned randomly to either the treatment group or a wait-list control group. Girls in the treatment group showed greater levels of desirable short- and long-term gains in STEM-related outcomes compared to wait-list control participants after one academic year. These included STEM activity, knowledge of STEM topics, knowledge about university studies and jobs in STEM, confidence in one’s own STEM abilities, self-assessment of STEM competencies, and intentions for academic elective choices.

      A four-year mixed-method prospective cohort study at two pediatric hospitals in Canada used an online mentor as part of a way to empower youth with chronic health conditions in their transition to adult healthcare.20 All study participants had access to an online transition mentor and the Youth KIT, a tool that includes goal-setting activities. The study found that participants had modest perceptions about the utility of the Youth KIT and online mentor. Overall, it was concluded that these two transition interventions were insufficient for empowering this sample of youth.

      Another study sought to look at the practicality of and to develop an implementation model for an inquiry-based learning environment (IBLE) that included e-mentoring using videoconference.21 Inquiry-based learning is “an approach to learning that involves a process of exploring the natural or material world, and that leads to asking questions, making discoveries, and rigorously testing those discoveries in the search for new understanding.”22 This study tested IBLE in a rural environment using a mixed-method approach focused on affective and cognitive outcomes; a pre-/post-test quasi-experimental designi was nested in a case study of three eighth grade math classes in one rural school. Results showed that IBLE appeared to have enhanced students’ learning, most significantly their affective development, including increasing their engagement and motivation, broadening their understanding of the relevancy of math and science in students’ lives, and augmenting their awareness of roles and careers in math and science.

      Another study used a randomized controlled design to look at the impact of an e-mentoring program for secondary school students (tenth through twelfth grades) with learning disabilities on their ability to identify post-school interests and goals and to map out the steps necessary to achieve them (transition competency).23 Students from eight high schools were randomly assigned to a control or intervention group. The intervention participants were matched with a college mentor, with whom they corresponded via a virtual classroom and attended two college campus visits. In the control condition, students only received the college tour and a simulated college classroom visit. Results showed that students with disabilities in the mentored group outperformed students in the control group on four of five measures—transition competency (both self-reported and verified by the parents and student’s special education teacher), self-determination, and social and academic connectedness.

      iA quasi-experimental design often looks like an experimental design but does not include random assignment of participants.


      1. The available evidence on the effectiveness of e-mentoring is mixed and does not allow one to draw conclusions about which formats work for which types of youth.

      2. The evidence also does not permit even tentative conclusions about the effectiveness of e-mentoring for different types of youth outcomes.

  • What Factors Shape the Effectiveness of E-Mentoring among Youth?


      There are several factors we might expect to mediate the relationship between an e-mentoring program and youth outcomes, based on research on traditional mentoring relationships. These could include the youth and mentor’s interpersonal histories, social competencies, the youth’s developmental stage, mentoring relationship quality, program practices, and the youth’s family and community context.24 In an e-mentoring program, however, there may be several unique factors that could potentially moderate the effectiveness of a program for youth. These could include demographics, personal factors, interpersonal communication styles, accessibility issues, or program implementation factors.9


      Some youth, such as those in rural locations, may receive more benefit from e-mentoring compared to a face-to-face program with infrequent meetings.25 Socioeconomic status (which may be confounded with race/ethnicity) may also be an influencing demographic characteristic, as youth from resource-poor areas may not have access to ICTs. On the other hand, lower socioeconomic status can lead to situations where e-mentoring is an ideal intervention. For instance, a youth’s family, neighborhood, and school may be unstable sources of support,23 not allowing for a youth to readily access naturally occurring mentors, or low income may make it impossible for a youth to access transportation to meet with a mentor. Finally, gender equality in the use of technology is crucial,4 and gender differences in the way males and females communicate via ICTs may influence the strength of an e-mentoring relationship.25, 26

      Personal factors.

      Both a mentor and a mentee’s personal circumstances may impact the outcomes of an e-mentoring program. For instance, an individual’s computer skills or adoption of ICT devices in general has the potential to influence the effectiveness of an e-mentoring program.9, 27 Youth in frequent crisis, without an alternative support network,14 and with needs that go beyond less frequent face-to-face meetings may also benefit from an ICT component, as it would allow them to reach a mentor much more quickly than would scheduling an in-person meeting to discuss issues.

      Interpersonal communication styles.

      A mentor who is accustomed to communicating in-person as opposed to text or email may find an e-mentoring program limiting, whereas a young person may find it more appealing because ICTs are a primary mode of communication for today’s youth. The lack of body language and nonverbal communication in e-mentoring may lead to misinterpretations and misunderstandings.13 However, mentors or mentees with social anxiety might find e-mentoring a more palatable mode of connecting.26, 28 Emotional maturity in general may be more important for an e-mentoring relationship as compared to a traditional face-to-face relationship, as e-mentoring requires the ability to disclose and share emotions online and in writing.9, 13 Without this ability, communication via ICT may be superficial or informational only.

      Accessibility issues.

      The potential exists for e-mentoring to be more accessible for youth with intellectual and developmental disabilities who may have difficulty participating in face-to-face mentoring because of a lack of transportation and availability of direct support.28, 29 It may also allow for mentors and mentees with chronic health conditions or physical disabilities to still participate in a consistent and reliable mentoring relationship, a crucial component for mentoring relationship strength and quality,30 since the interactions could occur from their homes or schools, thus enhancing effectiveness.

      Program implementation.

      Access to a stable mobile network or Internet service is essential for an effective e-mentoring program.31 The ability to access technological support could also be important when using ICTs to facilitate mentoring.32 Questions regarding security and confidentiality of an e-mentoring program may also influence how much information a mentor or mentee reveals in their relationship,27 which can contribute to weaker or stronger relationship ties.15 The program materials must also be easy to use. For example, if all mentoring interactions occur in a virtual world, then familiarity with how the world works, or sufficient training on how to utilize all features, would be necessary. Finally, the use of e-mentoring models alone versus “blended” models where e-mentoring is combined with traditional face-to-face mentoring could each produce different outcomes for youth.


      Six research studies assess possible moderators that contribute to the effectiveness of e-mentoring programs. These studies focus largely on demographic factors, but also the quality of the electronic interaction and some characteristics of the mentors and mentees themselves.


      A previously mentioned study investigating the impact of an e-mentoring program on students with learning disabilities and their ability to identify post-school interests and goals and the steps necessary to achieve them found that estimated effects on student outcomes varied according to student race and socioeconomic status.23 Non-Caucasian students demonstrated more improvement on the measure of self-determination in relation to program participation than did Caucasian students, as did students from higher socioeconomic status backgrounds relative to those from lower socioeconomic backgrounds.

      One case study looked at gender as a potential moderator of the effects of computer-mediated communication between high school students with disabilities and adult mentors.33 The goal of the program was to promote the participation of these students in STEM fields. Differences in the content of the communication between mentors and mentees were found to be consistent with traditional gender roles; males were more likely to provide and seek information about the Internet and technology than were females, yet females communicated more frequently, shared more personal information, and had a personal tone in their interactions. This suggests that for e-mentoring programs focused on STEM, the programs may need to find a way to encourage female mentees to increase support-seeking related to the STEM topics themselves.

      Another longitudinal observational study of a nine-month e-mentoring program looked at social capital as a possible factor moderating success.34 In this program, mentees developed multiple relationships with a “network” of mentors and communicated with their network entirely through electronic media. Results showed that students with an educational role model at home saw greater improvements in general and career-based self-efficacy than did students with no educational role model.

      Personal factors.

      Only one study looked at personal factors as potential moderators of an e-mentoring program. This study looked at mentees’ preprogram attitudes and experience, specifically prior Internet use, previous experience with mentoring, motivation to participate in the program, and preprogram general self-efficacy as they related to outcomes focused on general, career, and fiscal self-efficacy.35 Survey data was collected from students over one year in 50 schools in which at least two-thirds of the student population fell below the U.S. federal poverty line. All students were required to participate in the mentoring program. Findings showed that general self-efficacy prior to the program and motivation to participate predicted the development of a more positive relationship between the mentor and mentee. Having mentors in the past was not found to be a predicting factor related to program outcomes. This suggests that a mentee’s background and personal context may need to be considered at the start of an e-mentoring program in order to maximize the development of the mentor/mentee relationship.

      Interpersonal communication styles.

      A qualitative study looked at primary communication via email, with three face-to-face meetings over an eight-month intervention period, between youth ages 15 and 20 with disabilities and mentors.36 This study compared the characteristics of “successful” (exchanged more than 50 percent of required messages; evaluated by mentors/students/program coordinator/teachers as most successful) and “unsuccessful” matches who completed the program. Match and mentor characteristics related to perceived success included mentoring style (instructional vs. relational), communication frequency, communication style, and the amount of face-to-face meetings.

      Accessibility issues.

      While several e-mentoring programs have been designed specifically to help youth overcome or cope with health issues,2, 20, 28, 33, 36, 37, 38, 39 none of the research looked at accessibility as a moderator of e-mentoring outcomes.

      Program implementation.

      No studies looked at technology as a moderator of e-mentoring outcomes for youth. One recent study did look at the program implementation practices of one-on-one versus group online mentoring in Germany as a way to encourage talented girls in STEM.40 Girls in gifted education were mentored online by female STEM academics for six months in either one-on-one or group mentoring. The study found that group mentoring was more effective than one-on-one mentoring for several youth outcomes, including the proposition of STEM communication, the extent of STEM-related networking, more growth in STEM elective intentions, and participants’ importance in their respective STEM networks. Overall, this study suggests group mentoring may be a more successful way to reach gifted girls in STEM than one-on-one mentoring. The study did not, however, look at girls being mentored versus those who were not.


      1. Some demographic characteristics, including gender, race, socioeconomic status, and having an educational role model at home, may influence the impact of e-mentoring, although currently available research does not suggest reasons why.

      2. General self-efficacy and motivation to participate might be related to the development of a positive relationship between a mentor and mentee in an e-mentoring program, but it is not clear if this truly moderates program outcomes for youth.

      3. Interpersonal characteristics such as mentoring style and communication style might moderate program outcomes, but there are no true tests of moderation to support this qualitative finding.

      4. How an e-mentoring program for youth is implemented—one-on-one versus group—may be an important moderator. For gifted girls with interests in STEM, group e-mentoring seems to be more effective than a one-on-one format.

      5. There are no known studies to date assessing how mentoring format—traditional, e-mentoring, or a blended model—affect youth outcomes.

  • What Are the Intervening Processes (Mediators) that Are Most Important in Linking E-Mentoring to Youth Outcomes?


      Factors we might expect to mediate the relationship between an e-mentoring program and youth outcomes, based on research on traditional mentoring relationships, include social-emotional development of the youth, cognitive development, and identity development, which in turn affect the quality of a youth’s parental and peer relationships, leading to better youth outcomes.24

      The level of engagement with e-mentoring materials, whether they are emails, a virtual reality, or a smartphone app, may, in part, determine the benefits of e-mentoring for a participating youth. For instance, if e-mentoring occurs in an interactive program, one activity may not be as engaging as a collection of activities. Furthermore, when using text-only communication, the recipient could be unaware of the sender’s emotions unless the sender explicitly expresses them in the text or via emoji. Multiple studies have shown that “social presence” is essential for e-mentoring relationships to develop successfully.9, 39, 41

      If those invested in an e-mentoring program, including the mentor, mentee, and—in cases where the activity is school-based—the teacher, are not satisfied with the program, this is quite likely to influence program outcomes. The type of support a mentee may gain from an e-mentor, such as informational support, tangible assistance, social support, or emotional support, could influence outcomes.9 In a study of an e-mentor program for college students, for example, relationships in which students received more vocational and psychosocial support were associated with better outcomes, including higher levels of career planning and intentions to continue the relationship.32 The same was also found to be true among mentors who reported providing more support. Relationship qualities, such as coordination of expectations, trust, self-disclosure, and empathy, could also be important links in pathways to positive youth outcomes.9 It is unknown how e-mentoring plays a role in these factors.


      Five studies focused on mediating factors that contribute to the efficacy of a youth e-mentoring program, each focusing on interaction frequency as a mediator of the ability of the e-mentoring to build self-efficacy in youth. In the first study,35 the researchers looked at a longitudinal sample of students in the iCouldBe program, which is a not-for-profit organization that creates and manages online adult-youth mentoring programs targeting lower-income middle and high school students. All contact between the students and mentors was exclusively online and anonymous. Results showed that interaction frequency fully mediated the relationship between program antecedents (the mentee’s previous Internet experience and initial participation motivation) and general self-efficacy. The same was true for motivation—interaction frequency was positively related to motivation, which was related to general self-efficacy and fiscal efficacy, as well as Internet use and program satisfaction. Mentee interactions partially mediated the relationship between career efficacy and Internet experience, career efficacy and motivation, and program satisfaction and mentee motivation. These results do not directly address the question of what processes mediate effects of e-mentoring program participation on youth outcomes. However, they could be informative in this regard based on the consideration given to differences in program experiences related to variation in outcomes for participating youth.

      DiRenzo and colleagues used longitudinal data from a nine-month e-mentoring program to look at the influence of formal e-mentor networks and family-based role models on psychosocial and career-related outcomes.34 They found that the relationship quality of the e-mentor network was positively associated with general and career-based self-efficacy, which, in turn, seemed to enhance career aspirations of the youth. It should be noted, however, that the relationship between e-mentor network relationship quality and general self-efficacy was significant only for those with an educational role model in the family, not for those without such a role model. Furthermore, even though the relationship between e-mentor network relationship quality and career-based self-efficacy was significant for both those with and without an educational role model, the relationship was stronger for those with a role model.

      The qualitative study of mentor-mentee interaction via email only, described previously, found a positive relationship between frequency of email contact and markers of relationship quality, including friendliness, mutual sharing, and deeper discussions of more personal issues.15

      A process evaluation of a virtual world for pediatric transplant recipients looked at factors such as time spent online, initiation of chat conversations, initiation of activities, and out-of-world contact as possible factors contributing to a successful e-mentoring relationship.38 The Camp Zora graphic virtual world was designed to create a community that offers psycho-educational support and the possibility of participating in virtual curriculum activities that address school transition and medical adherence. Quantitative data, which came from online chat transcriptions and out-of-world contact, included the number of logins, time spent online, and the number of objects, characters, and virtual spaces created during the study. Results showed that a successful relationship with an e-mentor was mediated through being a consistent presence online, initiating the majority of conversations and curricular activities, promoting relationships between other participants, and devoting attention to out-of-world communication.

      Question one of the iMentor study described above17 looked at variations in implementation of the program across schools and its association with the strength of the relationship between a mentor and mentee. Specifically, they looked at the number of mentoring matches made, the number of iMentor classes held, the frequency with which students and mentors emailed, and the number of students who attended at least six events. They found the two schools that implemented the program with the highest fidelity had the highest mentor/mentee email rates and event attendance rates. They also found that these two schools had the highest student school attendance rates.


      1. Studies assessing interaction frequency and relationship quality in e-mentoring show these factors have an influence on youth outcomes, such as self-efficacy and motivation.

      2. While some studies did not directly assess what processes mediate effects of e-mentoring program participation on youth outcomes, their results could be informative based on the consideration given to differences in program experiences related to variation in outcomes for participating youth.

  • To What Extent Have E-Mentoring Programs Reached and Sustained the Engagement of Youth, Been Implemented with High Quality, and Been Adopted and Sustained by Organizations and Settings?


      Challenges to overcome in e-mentoring.

      One key challenge to overcome in an e-mentoring program is skill deficiency, including reading comprehension and expressive written communication, as well as computer, Internet or ICT device skills, or typing ability. These limitations may lead to a lack of interest in the program or a misunderstanding of electronic correspondence if not properly addressed.9, 13, 35, 39 E-mentoring is naturally dependent on functioning technology, and time delays may lead to feelings of abandonment, panic, and/or frustration for both the mentor and the mentee.9, 13

      Best practices for implementing e-mentoring.

      A previous review9 provided a checklist for program developers who want to be successful in implementing an e-mentoring program for youth: establish program goals, recruit participants, match pairs or groups, provide ongoing support, and evaluate the program’s implementation and outcomes. Just like in traditional face-to-face mentoring, it may be important to consider each phase of the program.

      In addition to these typical phases for consideration, other factors that could be equally critical for e-mentoring, included:

      1. selecting appropriate and accessible ICT, including various communication channels;
      2. establishing frequency of contact and appropriate expectations for frequency;
      3. supplying tutorials and conducting retraining for both mentors and mentees;
      4. establishing expectations for the longevity of the relationship;
      5. considering ways to build relationship quality, emotional closeness, and interaction levels;
      6. engaging the participants through online task-based activities; and
      7. developing mechanisms for support and involvement of the parents/guardians.9, 42

      It may also be helpful to point out the possible advantages of using e-mentoring compared to traditional face-to-face methods to new mentors before they are matched with a mentee.27 This may be especially important for mentors who are older or are not as confident in their use of ICTs. In fact, some observers have suggested that programs focus on what e-mentoring can contribute rather than what it lacks in comparison to face-to-face program models.14 Finally, mentees could be encouraged to initiate contact in e-mentoring relationships in order to engage in a symmetrical communication process.27 This constant electronic feedback by mentees may counter hesitations that e-mentoring is less engaging than face-to-face interaction.


      Six of the studies included in this review were reflective of efforts to engage in e-mentoring in a high-quality programmatic way. The first example is a study that assessed 26 lengthy email relationships between students in seventh to twelfth grades and volunteer scientists who advised them on science projects.41 In this study, students were assigned to work in research teams. The teacher then matched the teams with a volunteer mentor with some expertise in the area. Student teams communicated with their mentors via email to seek expert advice when they faced a problem with their project. The mentor would then respond with suggestions and data sources. Factors that predicted the sustainability of a mentoring relationship in this study included relationships that have “productive utility” for students. In other words, instances where the mentors helped steer students toward more manageable approaches to thinking about their projects were seen as a benefit. Another factor that appeared to help sustain these relationships was “lightweight” interaction—i.e., requiring as little as 15 minutes of time per week.

      A case study was conducted for an Internet-based mentoring community for college-bound youth with disabilities called DO-IT.37 This e-mentoring community is made up of high school and college students, graduates of the program who are now in college or employed, and mentors who are volunteer college students and working professionals. DO-IT uses Internet discussion lists to allow mentors and mentees to talk about topics of mutual interest in a group style of mentorship. The program is designed to support students with disabilities to pursue STEM and business fields. The case study outlines insights from the program experiences that may facilitate successful implementation of an e-mentoring community: establish goals for the program; select appropriate technology for communication; develop the communication structure (such as initial messages, replies, and forwarded messages); develop procedural and behavioral guidelines for mentees, mentors, and parents; recruit and orient participants; introduce new mentors and mentees to the community; provide supervision and ongoing support of mentors; manage online discussions through question prompts for individual encouragement; and evaluate the program by seeking feedback from community members and updating the application and training materials as needed. All of these factors were believed to have led to the sustainability of the program, which has been ongoing for decades.

      One early study of “telementoring” looked at a program for students in seventh to twelfth grades who had lengthy email relationships with volunteer scientists who advised them on science projects.41 Classroom observations and preliminary interviews were used to collect data at three different time points. The telementoring served as a practical way to provide support for ambitious science learning, as indicated by students being able to think through their project approaches more productively and with continuing motivation. In order for it to be sustainable, however, the researchers recommended, based on their findings, that students’ work should be made more visible to the mentors than what email alone provides. Also, teachers were found to require organizational tools to help them manage the program.

      A qualitative study examined the use of e-mentoring to provide social and emotional support for Israeli youth ages 15 to 20 years who had socioemotional disabilities.39 The mentors also had disabilities. Each mentor was matched with two mentees, and they were expected to communicate by sending at least two email messages per week for about four months. Findings supported the potential of e-mentoring for personal development and empowerment of this population of youth; however, most of the mentors and mentees considered the e-mentoring process to be a barrier to developing a personal relationship. For instance, they reported a lack of “reciprocal self-disclosure,” where one partner’s disclosure is followed by the disclosure of others, which may happen more naturally in a face-to-face interaction. The irregular pace of email also frustrated the mentors, and participants expressed desire for more face-to-face contact in addition to the e-mentoring component. Despite some of these dissatisfactions, the majority of mentors and mentees rated the program as an enjoyable experience.

      Finally, the study that looked at developing an implementation model for IBLE21 also documented challenges in the process of creating an IBLE environment. These challenges included students’ lack of knowledge of current math and science use in existing professions, the technical quality of videoconferences, camera/monitor use and setup logistics, and the students’ physical comfort during the sessions.


      1. E-mentoring programs that have been implemented and sustained seem to benefit from clear guidelines, structure, and organizational tools.

  • Implications for Practice

    (Mike Garringer, MENTOR: The National Mentoring Partnership)

    As electronic communication seeps into more facets of modern life, especially for youth today who are growing up as what many have described as “digital natives” who will never know a world without instant communication at one’s fingertips, it is understandable that youth mentoring programs would be thinking about how to incorporate online communication into their services or even to make the entire mentoring experience “virtual.” But the adoption of virtual communication in the mentoring field has been far slower than it has been for society as a whole⎯a 2016 survey of youth mentoring programs in the United States found that only 1 percent of programs identified as primarily using an e-mentoring model, and only 3 percent indicated they have matches that meet online some of the time.43 So while both youth and adults may be glued to their screens, those devices are not being integrated into mentoring as much as one might expect.

    As noted in the review, there are meaningful challenges to implementing e-mentoring, ranging from issues of technology access and use, to philosophical concerns about the ability of participants to get what they need out of a relationship that does not meet in-person at least occasionally. But it does seem inevitable that electronic communication will be a meaningful part of mentoring relationships moving forward⎯the tools of modern communication are simply too ubiquitous to not be integrated in increasingly important ways. As youth-serving programs wade into these waters, they might find increased success by following the axiom of “people, not platform.” Programs that want to utilize electronic communication in mentoring often get hung up on the bells and whistles of the technology itself⎯the use of video, the ability to monitor interactions, the notifications and alerts, etc. But whether your program is integrating electronic communication into a primarily face-to-face program or using it for all of the mentor-mentee interactions, there should be considerable thought put into how and why people will use the tools provided, especially since the review here notes that there are mixed findings about the types of platforms that work best and the many considerations that might moderate the effectiveness of an e-mentoring approach. The following questions and topics can help mentoring programs integrate communication technology effectively.


      This review notes several ways in which research suggests technology can be an asset for mentor-mentee communication. It may alleviate geographic distance, allow participants with physical disabilities greater access to mentoring experience, overcome some of the awkwardness and shyness of initial in-person meetings, and allow youth to communicate with mentors on a more frequent and less time-bound basis. It may allow a child’s mentor to be just a few clicks away at all times. Notwithstanding the early stage of research in this area, it appears that there are ways in which communicating electronically may be preferable to face-to-face communication or, at the very least, offer beneficial opportunities for additional mentoring exposure and more real-time interactions. Informed by the findings of this review, program staff should map out the reasons they think electronic communication can help facilitate a better mentoring experience for youth and volunteers (and likely avoid proceeding if they are unable to come up with a convincing list of reasons as to why it could be an improvement).

      But even if there are compelling reasons to offer or allow online communication between mentors and mentees, programs will also need to anticipate some of the potential challenges noted in questions two and three of this review and plan accordingly. These challenges can include:

      • Access to technology, both for mentors and mentees. And if participants would be using their personal devices for this communication, is there a philosophical or ethical concern about offering something that not all mentees could participate in or take advantage of? Imagine being the only mentee in a program paired with a mentor who can’t text you or who can’t follow your accomplishments on Facebook. Some young people could have a radically different mentoring experience if technology is available inconsistently across the participants.

      • Use of the technology. Even if participants have access to the technology, are they equally familiar with it? Is it easy to use? How will it impact the relationship if one of them struggles to use the tools as intended?

      • Rules around the use of technology. During what hours can mentees write to their mentors? What communication types are allowed (email, text, social media, etc.)? What are mentors allowed to do or say via technology? What are the expectations around frequency and volume of interaction?

      • Changes in technology. If your program invests in one form of digital communication, will that approach still be viable in five years? In ten? The pace of change in communication technology is staggering, and it can be a struggle for resource-shy nonprofits to keep up. In one interview on e-mentoring, researcher Kevin O’Neill noted, “In recent years I have found high school students in my e-mentoring programs increasingly frustrated with having to post messages to their mentors in my secure forum and wait for replies. They ask why their mentors are not available on instant messaging or on Facebook.” So even when a program invests in a platform that is currently “state of the art”, as I’m sure this forum was at the time, participants can, within a few years, feel like they are using a platform that is not up to their expectations.

      Perhaps the last consideration around the use of electronic communication at the relationship level is thinking about whether this type of communication will make users feel more or less comfortable in forging a deep or otherwise beneficial relationship and whether they even have an ability to do that within an electronic medium. This review notes research suggesting that it takes a certain level of maturity to share personal information online and that some youth may not feel comfortable putting some things about their life in writing and hitting “send.” (Of course, it may be equally true that it could be easier for some youth to disclose things online rather than in person.) It may be challenging for mentors to bond with a youth without face-to-face interactions, or in a program that combines the two, the risk may be that electronic communication supersedes the in-person time in a way that is undesirable or negates the trust, empathy, and mutuality that is essential to mentoring relationships. Programs must think through these pros and cons and consider whether using electronic communication is a net gain for participants or perhaps not worth the challenges.


      Although not discussed in much detail in the review, it is important to consider the many ways in which staff will have to support the e-mentoring efforts. Just because mentors and youth can text each other furiously throughout the day does not mean they will not have the same types of issues and need the same amount of monitoring and support as would in-person matches. In fact, the research considered in this review points to several things staff should be prepared to address:

      • Ensuring matches are communicating frequently and meeting each other’s expectations. As noted in the review, the frequency and content of electronic communications seems to mediate the effectiveness of the experience. Without regularly scheduled in-person meetings, matches can become “out of sight, out of mind.” Programs will need to consider whether they want to provide discussion prompts, send reminders to write frequently, or otherwise reach out to participants to make sure they are talking frequently enough, as well as about the right things. It is also possible to have pairs that spend a lot of time communicating but just at a surface level or in ways that leave one or both of the participants dissatisfied. All of this prompting, monitoring, and spurring of proper participation will take a lot of effort on the part of the staff. Programs should expect and plan around this critical aspect of ensuring effective electronic communication.

      • Offering help in smoothing over communication hiccups. One of the challenges of electronic communication noted in the review is that there are no visual clues to help fill in the nuances of a conversation. All participants have are the words on the screen, which can easily lead to misunderstandings and disagreements about tone or intention. There can also be challenges related to the timing of responses⎯it is easy to think of scenarios where one person shares something personal and expects a reply that is delayed for a whole host of reasons that have nothing to do with the content of the message. It is entirely possible that, depending on the circumstances and persons involved, electronic communication can increase the stress and anxiety around forming a mentoring relationship rather than relieving it. Programs should think carefully about how and when staff can step in and keep the matches on track.

      • Preparing participants for e-mentoring. Needless to say, any program that wants to heavily use technology will need to make sure participants can use it effectively. In addition to all the things mentors and mentees must be trained on regarding how to do the work of a relationship, there is an added layer of training to get them comfortable with using the technology in a way that does not interfere with the relationship itself.

      It is worth noting that all of the six core Standards of The Elements of Effective Practice for Mentoring apply to e-mentoring programs as well. From recruitment all the way to match closure, programs must be prepared to think about how the technology will influence how staff meet those standards. In some cases, electronic communication between mentors and youth may make their jobs easier. But in others, programs may find that e-mentoring is actually more labor-intensive for staff, particularly if maximizing the frequency and quality of those interactions is a must.


      After thinking about all those “people” considerations of e-mentoring for mentors, youth, and staff, a program will eventually need to turn its attention back to the “platform” and figure out exactly how all of this will work. Programs may find it helpful to develop a formal technology implementation plan. One school district in Texas offers a nice guide to developing e-mentoring programs that suggests these simple components and concepts for effectively implementing technology in a mentoring context:

      Design a technology implementation plan that includes:

      • A communication system that will meet the needs of the program and its participants;
      • A communication system that is safe and reliable;
      • Clearly defined technology requirements of partner organizations;
      • The defining of technology-related roles and responsibilities among program participants;
      • Determining whether the mentor and youth participants need e-mail accounts or computers and whether your program will provide them;
      • Policies regarding privacy and security of participant data and communication; and
      • Policies regarding access to communication content, including privacy. Who will be allowed to view e-mails and under what circumstances will they be viewed?

      Hopefully answering questions such as these will allow programs to effectively integrate the use of technology platforms into their services and mentoring relationships. The fast-paced, connected, digital world humanity is now developing demands that mentors, mentees, and practitioners thoughtfully plan for an increased technological role in their work.

  • References

    1. Bierema, L., & Merriam, S. (2002). E-mentoring: Using computer-mediated communication to enhance the mentoring proceess. Innovative Higher Education, 26, 211–227.

    2. Kendal, S., Kirk, S., Elvey, R., Catchpole, R., & Pryjmachuk, S. (2017). How a moderated online discussion forum facilitates support for young people with eating disorders. Health Expect, 20, 98–111.

    3. Risquez, A. (2008). E-mentoring: An extended practice, an emerging discipline. In S. Kelsey & K. St. Amant (Eds.), Handbook of Research on Computer-Mediated Communication (Vol. 2, pp. 656–677). Hershey, PA: Information Science Reference.

    4. Bennett, D., Hupert, N., Tsilalas, K., Meade, T., & Honey, M. (1998). Critical issues in the design and implementation of telementoring environments. New York, NY: Education Development Center's Center for Children and Technology. Retrieved from

    5. Kasprisin, C., Single, P., Single, R., & Muller, C. (2003). Building a better bridge: Testing e-training to improve e-mentoring programmes in higher education. Mentoring & Tutoring, 11, 67–78.

    6. O'Neill, D. (2011). Local social capital through e-mentoring: An agenda for new research. Learning, Media and Technology, 36, 315–321.

    7. Kim, H. (2014). Enacted social support on social media and subjective well-being. International Journal of Communication, 8, 2340–2342.

    8. Cole, D., Nick, E., Zelkowitz, R., Roeder, K., & Spinelli, T. (2017). Online social support for young people: Does it recapitulate in-person social support; can it help? Computers in Human Behavior, 68, 456–464.

    9. Shpigelman, C. (2014). Electronic mentoring and media. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp. 259–272). Thousand Oaks, CA: SAGE.

    10. Graves, M., Roberts, M., Rapoff, M., & Boyer, A. (2010). The efficacy of adherence interventions for chronically ill children: A meta-analytic review. Journal of Pediatric Psychology, 35, 368–382.

    11. Cafazzo, J. A., Casselman, M., Hamming, N., Katzman, D. K., & Palmert, M. R. (2012). Design of an mHealth app for the self-management of adolescent type 1 diabetes: a pilot study. J Med Internet Res, 14, e70.

    12. Ippoliti, N. B., & L'Engle, K. (2017). Meet us on the phone: Mobile phone programs for adolescent sexual and reproductive health in low-to-middle income countries. BMC Reproductive Health, 14, 1–8.

    13. Ensher, E., Heun, C., & Blanchard, A. (2003). Online mentoring and computer-mediated communication: New directions in research. Journal of Vocational Behavior, 63, 264–288.

    14. Risquez, A., & Sanchez-Garcia, M. (2012). The jury is still out: Psychoemotional support in peer e-mentoring for transition to university. Internet and Higher Education, 15, 213–221.

    15. Rhodes, J., Spencer, R., Salto, R., & Sipe, C. (2006). Online mentoring: The promise and challenges of an emerging approach to youth development. Journal of Primary Prevention, 27, 497–513.

    16. Culpepper, D. (2008). Determining the quality and impact of an e-mentoring model on at-risk youth. [Doctoral dissertation]. Retrieved from University of South Florida Scholar Commons.

    17. Merrill, L., Kang, D., Siman, N., & Soltani, J. (2016). Focus on mentee-mentor relationships: The 10th grade implementation of iMentor College Ready Program. New York, NY: The Research Alliance for New York City Schools. Retrieved from

    18. Merrill, L., Siman, N., Wulach, S., & Kang, D. (2015). Bringing Together Mentoring, Technology, and Whole-School Reform: A First Look at the iMentor College Ready Program. New York, NY: The Research Alliance for New York City Schools. Retrieved from

    19. Stoeger, H., Duan, X., Schirner, S., Griendl, T., & Ziegler, A. (2013). The effectiveness of a one-year online mentoring program for girls in STEM. Computers & Education, 69, 408–418.

    20. Gorter, J. W., Stewart, D., Cohen, E., Hlyva, O., Morrison, A., Galuppi, B., . . . Trace Study Group. (2015). Are two youth-focused interventions sufficient to empower youth with chronic health conditions in their transition to adult healthcare: a mixed-methods longitudinal prospective cohort study. BMJ Open, 5, e007553.

    21. Li, Q., Moorman, L., & Dyjur, O. (2010). Inquiry-based learning and e-mentoring via videoconference: A study of mathematics and science learning of Canadian rural students. Education Technology Research and Development, 58, 729–753.

    22. National Science Foundation. (2005). Inquiry: Thoughts, views, and strategies for the K-5 classroom. Arlington, VA: National Science Foundation.

    23. Collier, M. (2009). An investigation study of the effects of e-mentoring on transition planning for secondary students with learning disabilities. [Doctoral dissertation.] Salt Lake City, Utah: University of Utah.

    24. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, 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.

    25. Wilson Jr., E. (2013). Factors that influence communication in e-mentoring with urban alternative high school students. [Doctoral dissertation]. KU ScholarWorks.

    26. Pierce, T. (2009). Social anxiety and technology: Face-to-face communication versus technological communication among teens. Computers in Human Behavior, 25, 1367–1372.

    27. Panopoulos, A., & Sarri, K. (2013). E-mentoring: The adoption process and innovation change. International Journal of Information Management, 33, 217–226.

    28. Durkin, K., Conti-Ramsden, G., & Walker, A. (2010). Computer-mediated communication in adolescents with and without a history of specific language impairment (SLI). Computers in Human Behavior, 26, 176–185.

    29. McDonald, K., Balcazar, F., & Keys, C. (2005). Youth with disabilities. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 493–507). Thousand Oaks, CA: SAGE.

    30. Rhodes, J., Reddy, S., Roffman, J., & Grossman, J. (2005). Promoting successful youth mentoring relationships: A preliminary screening questionnaire. Journal of Primary Prevention, 26, 147–167.

    31. Single, P., & Single, R. (2005). E-mentoring for social equity: Review of research to inform program development. Mentoring & Tutoring: Partnership in Learning, 13, 301–320.

    32. Murphy, W. (2011). From e-mentoring to blended mentoring: increasing students' developmental initiation and mentors' satisfaction. Academy of Management Learning & Education, 10, 606–622.

    33. Burgstahler, S., & Doyle, A. (2005). Gender differences in computer-mediated communication among adolescents with disabilities: A case study. Disability Studies Quarterly, 25.

    34. DiRenzo, M., Weer, C., & Linnehan, F. (2013). Protégé career aspirations: The influence of formal e-mentor networks and family-based role models. Journal of Vocational Behavior, 83, 41–50.

    35. DiRenzo, M., Linnehan, F., Shao, P., & Rosenberg, W. (2010). A moderated mediation model of e-mentoring. Journal of Vocational Behavior, 76, 292–305.

    36. Shpigelman, C. N., & Gill, C. J. (2013). The characteristics of unsuccessful e-mentoring relationships for youth with disabilities. Qualitative Health Research, 23, 463–475.

    37. Burgstahler, S., & Crawford, L. (2007). Managing an e-mentoring community to support students with disabilities: A case study. AACE Journal, 15, 97–114.

    38. Cantrell, K., Fischer, A., Bouzaher, A., & Bers, M. (2010). The role of e-mentorship in a virtual world for youth transplant recipients. Journal of Pediatric Oncology Nursing, 27, 344–355.

    39. Shpigelman, C., Weiss, P., & Reiter, S. (2009). E-mentoring for all. Computers in Human Behavior, 25, 919–928.

    40. Stoeger, H., Hopp, M., & Ziegler, A. (2017). Online mentoring as an extracurricular measure to encourage talented girls in STEM (Science, Technology, Engineering, and Mathematics): An empirical study on one-on-one versus group mentoring. Gifted Child Quarterly, 1, 1–11.

    41. O'Neill, D. K., & Gomez, L. M. (1998). Sustaining mentoring relationships on-line: Proceedings of the ACM 1998 Conference on Computer Supported Cooperative Work (pp. 325–334). Seattle, WA: Association of Computing Machinery.

    42. Miller, H., & Griffiths, M. (2005). E-Mentoring. In D. DuBois & M. Karcher (Eds.), Handbook of youth mentoring (1st ed., pp. 300–313). Thousand Oaks, CA: SAGE.

    43. Garringer, M., McQuillin, S., & McDaniel, H. (2017). Examining youth mentoring services across America: Findings from the 2016 Mentoring Program Survey. Boston, MA: MENTOR: The National Mentoring Partnership.

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.

One-to-One Cross-Age Peer Mentoring
PDF button Facebook button Twitter button

September 2017

This review addresses four topics related to one-on-one cross-age peer mentoring for children and adolescents, including:

  1. Its documented effectiveness for mentees and mentors,

  2. The extent to which effectiveness depends on characteristics of mentors, mentees, or program practices,

  3. Intervening processes likely to link cross-age peer mentoring to youth outcomes, and

  4. The success of efforts to reach and engage targeted youth and achieve high quality implementation.

Extending a 2007 MENTOR Research in Action monograph definition of cross-age peer mentoring,1 which also was used in other literature reviews on cross-age peer mentoring,2, 3 this review sharply differentiates cross-age one-to-one peer mentoring programs from cross-age peer group mentoring, peer-led education or targeted preventative interventions, and peer mentoring as an informal practice within larger programs.

Overall, evidence is beginning to accumulate that supports at least the short-term effectiveness of formal cross-age peer mentoring programs. But this literature is growing at a very slow pace, mainly, it seems, because most of the literature on “peer mentoring,” old and new, combines one-to-one cross-age peer mentoring with group peer mentoring programs and peer education led by older youth.

The limited evidence of effectiveness of cross-age peer mentoring, specifically as defined in this review, reveals benefits accrued by both children (mentees) and their teenage mentors. However, benefits to mentors are not the focus of this review. The strongest effects for mentees appear to be increases in school attitudes (e.g., connectedness), relationships with adults (both teachers and parents) and peers, and improvements in internal affective states (e.g., self-esteem).

The most significant moderators of program effectiveness appear to be the mentors’ attitudes and motivations, and the degree of clear programmatic infrastructure and fidelity of its implementation. Involvement of parents in programs also seems to yield larger benefits, and securing support from school administrators and teachers can directly influence effectiveness.

The means by which programs have positive effects on mentees appears to be largely through the consistent and affirming presence of mentors, and the clarity and predictability resulting from a clear program structure. These assist mentors in establishing what Rhodes4 describes as the building blocks of successful mentoring relationships—empathy, trust, mutuality—despite variability in the maturity and social distractibility of the teenage mentors.

  • Introduction

    This review examines what we know about the effectiveness of mentoring on youth outcomes when the mentors are not adults, but rather youth who are at least two to three years older than their mentees. We differentiate cross-age peer mentoring from other types of peer-led programs in order to underscore the extent to which mentoring, and not other types of activities, contributes to mentee outcomes. This review draws primarily on evidence from seven published outcome studies that meet a specific definition of cross-age peer mentoring.

    It is possible that cross-age peer mentoring is so intuitively appealing because it can fill a void in the natural mentoring of children by older peers, which occurred in schools until the twentieth century. Well into the twentieth century, and until much later in many rural communities, students of all ages were taught by one teacher in the same room. This single-room school setting allowed older youth to interact with, befriend, support, encourage, and serve as role models for the children in their class. When classrooms were stratified by grade, those natural mentoring relationships between children and youth became less common. Today, formal cross-age peer mentoring programs provide one way of restoring this vital developmental experience for children.

    Cross-Age Peer Mentoring Defined

    A narrow and clearly defined definition of cross-age peer mentoring is essential for a strong scientific foundation for cross-age peer mentoring to develop by making clear distinctions among the different roles that peers can play in each other’s lives. The roles of an adult coach, educator, tutor, counselor, mediator, and mentor are well defined, but this is not so for youth-helping-youth relationships. Best known are the social roles of classmate, teammate, or friend. Less clear are the distinctions between the multiple helping roles youth play with children. Youth may coach, teach, tutor, and counsel children to resolve specific problems. Or youth can be trained to befriend and interact supportively with younger youth—to mentor. It seems apparent that without clearer definitions, research on cross-age peer mentoring will not advance.

    This review restricts the term cross-age peer mentoring to the matching of an older youth (the mentor) with a younger youth (the mentee), in which there is a difference of two or more years in age between mentor and mentee. An age difference likely permits the older youth mentor to fulfill several roles similar to that seen in adult-youth mentoring that same-age peer relationships might not. Being older allows the peer mentor to serve more effectively as a role model, to provide support, and to offer the mentee guidance. Typically, this happens in school or community programs wherein secondary (high school) students are matched with primary or middle school students.

    Teenage mentors’ general maturity level necessitates that cross-age peer mentoring programs provide the mentors considerable structure and supervision through the use of planned activities and group interactions.5 The developmental scaffolding of the mentoring relationship through the use of mentor-mentee activities that occur within a larger peer group context can make these programs responsive to the teen mentors’ own social needs. For example, the little research on cross-age peer mentoring programs enlisting preteen mentors suggests it is less effective than programs involving teenage mentors.6 This may happen because, developmentally, preteens lack necessary relationship building skills, and even high levels of program structure and supervision may not be enough to scaffold preteens to effectively mentor. These matches may become overly playful or very task- or problem-focused, neither of which allows the mentors to be both caring and growth-focused in ways needed to serve as role models, friends, and confidants in the face of challenges.

    The second criterion for program and study exclusion was whether the program description provided clear evidence that the mentoring relationship is given primacy over learning the curriculum. Studies in which program descriptions primarily detailed curriculum and gave little evidence of programmatic efforts to foster one-on-one relationships (e.g., in mentor training or designated time for un- or semi-structured interactions between a mentor and mentee) were excluded. At its core, cross-age peer mentoring emphasizes the relationship between mentor and mentee, and thus is differentiated from goal-oriented models or programs with more narrowly prescribed goals, such as improving academic skills (peer tutoring), imparting knowledge or raising awareness (peer education), resolving or preventing interpersonal problems (peer mediation or assistance), or addressing personal problems (peer counseling).1 Peer mentoring relationships may incorporate such activities or roles temporarily, but are neither prescribed nor prioritized.

    Excluded from this review were several studies of programs that restricted evaluation outcomes to the mastery of curricular content. Impressive work7 using older youth to deliver health education curriculum one-to-one to younger children, for example, was excluded because the program focus and primary outcomes of the study were the mastery of curricular content.

    Group peer mentoring programs were excluded from this review. Peer mentoring (in contrast with peer education, mediation, counseling, and tutoring) requires (by definition) developing a close, personal, but also growth-focused relationship. Consequently, most formal peer mentoring programs use a one-to-one model. Some programs use a group format or structure that provides opportunities for relationships to form between the older peer mentors and the mentees in their group. In our review, however, we found no studies of group peer mentoring programs that explicitly described programmatic efforts to cultivate these one-to-one relationships or that attempted to assess the presence or quality of the mentoring relationships in the program. Peer Group Connection,8 for example, is an impressive intervention, but studies of its effectiveness describe it more as a peer leadership program, rather than a one-to-one peer mentoring program.

    Finally, less researched, cross-age peer mentoring can happen informally, such as in summer camps, when a teen’s formal role (e.g., as a child’s camp counselor) with a child expands to incorporate friendship-like qualities, or informally as in the one-room schoolhouse described earlier. Studies were located of programs in which teens were trained to provide informal support for same-age or younger peers (e.g., to facilitate transitions to a new school). But none of these studies provided evidence of whether any mentoring relationships were formed or prioritized and therefore fall outside the scope of this review.

    The Scope of this Review

    This review considers only cross-age one-to-one peer mentoring programs that meet the previously mentioned definition1 and on which relatively rigorous research has been conducted, either experimental or quasi-experimental examinations of outcomes or on practices that may explain these outcomes. For example, the search described below identified several dissertations and published outcome studies of programs that did fit this definition, but lacked rigor, such as those having too few participants, inappropriate comparison groups, or excessively high attrition.

    A systematic literature search of the ProQuest, PsycInfo, ERIC, and Google Scholar databases was conducted starting with the terms “peer mentoring” and yielded over 15,000 documents. The search was restricted to all reports available, in any year, in English. Of those, 1,177 were peer reviewed and involved youth, adolescents, children, or teens as mentees. After excluding articles using teachers, adults (mentors 19 years of age and older), and college students as the mentors, 225 articles remained. After eliminating duplicates and applying inclusion/exclusion criteria above, 42 articles, book chapters, and evaluations were retained, but only seven of these studies were deemed of sufficient rigor to provide reliable validity evidence for program outcomes.

    To identify populations and practices that explain or moderate program outcomes, the next three sections of this review only consider practices reported in studies of the programs for which reliable outcomes were found. However, to incorporate the wealth of practice expertise and wisdom in the field, the fourth section of the review summarizes the perceptions on the key peer mentoring program practices reported by several hundred program coordinators.

  • What Are the Demonstrated Effects of Cross-Age Peer Mentoring on the Development of Children and Adolescents?


      This section presents evidence of benefits for children who participate as mentees across three domains. In their 2012 meta-analysis,9 DuBois and colleagues found no differences in the overall effectiveness of peer mentoring programs compared to adult mentoring programs. This equivalence might, but also might not, have been found had their comparison excluded the studies not considered in this review.


      This section presents outcome information from only seven studies of dyadic cross-age peer mentoring outcomes. Each has significant research design or program implementation flaws, but viewed collectively in terms of type of outcomes, they provide considerable corroboration.

      Having a large sample size provides one of the best ways to minimize chance findings. There were two large-sample studies of cross-age peer mentoring. The first is the Big Brothers Big Sisters of America’s High School Bigs program evaluation.10 This experimental study had a large sample, rigorous data analyses, and a control group, but the program’s implementation varied across settings yielding quasi-experimental estimates of how impacts may differ by conditions and participant characteristics. A second study11 with an even larger sample size is a secondary data analysis of the Institute for Educational Science’s (IES) study of the Student Mentoring Program.12 Although mentees were randomly assigned to the treatment (mentoring) or control group in both studies, they were not randomly assigned to have a teen or an adult mentor.

      The two randomized, controlled studies of the cross-age peer mentoring called the Cross-Age Mentoring Program (CAMP) were included.13 Although somewhat statistically underpowered, together they provide corroborating evidence of impact on specific outcomes and information about for whom and under what conditions program effectiveness varies.

      Three other studies utilized an alternative treatment or a matched comparison group to estimate variations in changes observed between youth in different program delivery conditions. The outcomes from these studies, which varied across participant groups or program delivery approaches, are included below even though they do not provide experimental evidence of program impact. The first of these used cross-age peer mentoring to prevent delinquency.14 In this two-year study, multiple ratings of behaviors and attitudes (reported by youth and adults) were compared across two program participation durations (one-semester and one-year [eight months]) and, across three contrasting curriculum delivery approach groups into which youth were randomly assigned. The second study looked at differences in pre-post changes for children matched with high school mentors as part of a Big Brothers Big Sisters agency in Edmonton, Canada (BBBS Edmonton).15 This study did not include a no-mentoring comparison group. Nor were youth randomly assigned to the different delivery approaches; but the strengths of this study are its attempt to replicate prior findings and test theory-based hypotheses, and its being sufficiently statistically powered to provide reliable estimates of between-condition differences. The final study was of a gang participation prevention program that used a matched comparison group from the community.16


      Half of the studies that included misbehavior and misconduct as an outcome variable reported declines in or lower rates of misconduct and misbehavior after participation in a cross-age peer mentoring program. The two studies that found reductions were serving youth at risk of engaging in delinquent behavior or entering a gang. Both were statistically underpowered and lacked a randomized control group. The delinquency prevention peer mentoring program found reductions in misbehavior from pre- to post-test that were replicated in both program duration conditions and were observed by multiple reporters. But the comparison groups were two alternate treatment conditions, such that the reductions were found only among youth in matches randomly assigned to focus first on friendship development. For those in this condition and who participated in the shorter program duration, the reductions were still present six months later.14 In the gang prevention program, which used 12 sessions of interactive violence prevention activities to structure matches’ time together, the mentees (n = 25) reported fewer positive attitudes toward violence at post-test than children in the matched comparison group (n = 35). However, outcome tests included only the children who completed a post-survey, which was half of the original participant sample.16

      Findings from the two large sample studies of school-based mentoring by teens and adults were mixed. The High School Bigs study reported no benefits of having a teenage mentor on misconduct (though there were reductions for mentees who had adult mentors).10 In the secondary analysis of data from the IES Student Mentoring Program study, those mentees with teenage mentors reported less misconduct at post-test compared to both mentees with adult mentors and the control group students, but the difference was not statistically significant.11


      The evidence of improvements in academic skills and attitudes resulting from participation in cross-age peer mentoring was found in one of two large-sample studies (IES), both of the smaller randomized controlled studies (CAMP), and one of the two quasi-experimental studies (BBBS Edmonton). The IES Student Mentoring Program study found improvements in scholastic efficacy but not in grades for children mentored by teens.

      The findings regarding the effectiveness of the BBBS High School Bigs program came from contrasting outcomes of adult and teen mentors within a larger study of the effectiveness of school-based mentoring in which nearly half (48 percent) of all mentors were teens.17 Considered separately, the matches with adults and with High School Bigs (teenage mentors) revealed that the magnitude of the program effects differed for children having an adult mentor versus a teen mentor. Children with teen mentors benefited less (or did not improve at all) on classroom effort, difficulty in class, self-reported GPA, and intentions to go to college. Thus, the improvements originally reported for youth in the larger study18 on academic behaviors and attitudes held only for children mentored by adults, not those mentored by teens.

      The randomized, small-sample study of the all-day, monthly cross-age peer mentoring program (CAMP), in which interactive academic instruction was provided for part of the day, demonstrated both improvements in spelling achievement scores and connectedness to school.20 The study of CAMP (implemented after school) found benefits on connectedness to school, but did not include grade or achievement outcomes measures. The BBBS Edmonton study found pre-post increases in teacher-reported academic performance.15

      Socioemotional well-being and skills/attitudes.

      The High School Bigs study casts light on the influence of peer mentoring on social outcomes. Mentees of High School Bigs benefitted only on 1 of 30 outcomes (peer/social acceptance), while youth with adult mentors benefitted on 12 outcomes, when compared to the control group. However, children with teen mentors scored higher than adult-mentored children on assertiveness, parent relationship quality, and social acceptance. The BBBS Edmonton study also found improvements over time among mentees on peer acceptance, connectedness to peers, and self-esteem, as did the evaluations of cross-age peer mentoring for youth at risk of delinquency and gang membership.14,16

      Multiple studies, including the High School Bigs study, have reported increased family connectedness or family life satisfaction among mentees. In the delinquency prevention program, which did not appear to have family engagement events, improvements in family functioning and relationships reported at post-test were also present at the six-month follow-up and in both program durations.14 Increases in connectedness to parents were observed in both studies of CAMP, in the study of the “nearby” program approach, wherein cross-age peer mentoring sessions were held weekly after school,19 and in the “faraway” program approach, in which buses delivered mentees one Saturday a month to the mentors’ schools for an all-day mentoring event.20 Both included biannual or quarterly family events for parents.


      1. Evidence of the effectiveness of peer mentoring programs is very limited, both because there are few studies of programs meeting the criteria for this review and because only seven of these programs had been tested with rigorous research designs.

      2. Multiple studies report evidence of increasing connectedness to family and peers, as well as peer acceptance and self-esteem.

      3. Consistent evidence was found regarding the benefits of school-based cross-age peer mentoring programs on school connectedness (or related outcomes like school bonding).

      4. There is conflicting evidence of cross-age peer mentoring effects on grades, class performance, or achievement, as well as on misbehavior and misconduct.

  • Under What Conditions Does the Effectiveness of One-to-One Cross-Age Peer Mentoring Vary?


      What is clear from the literature on cross-age peer mentoring is the wide variability in program infrastructure (staff support to mentors, mentor training, and activities) and program participants. Just among the seven studies used in this and the next section to describe conditions under which program effects vary and the pathways by which program effects may operate, the characteristics of the programs, participants and activities varied greatly. To reveal what can be learned about what moderates program impacts, this section highlights several program components and mentor characteristics that seem to influence program effectiveness for only those seven studies previously described for which reliable outcome estimates could be made.


      Group interaction opportunities.

      The High School Bigs study identified several moderators of program effectiveness. Subsequent reanalysis of the same data has revealed others. A key initial finding is that matches lasted longer when the program included both dyadic mentor-mentee time and participation in a larger group,10 which may be because this combination of activities helps meet the socialization needs of the teen mentors. However, it should be noted that when matches meet in the context of other matches, the mentees reported less positive relationships with their mentors. Teens may be less easily distracted by their own peers if some time is structured for teen mentors to interact with their own peers.

      Minimizing compensation and compulsory participation.

      In both the High School Bigs10 and the Big Brothers Big Sisters Edmonton teen mentoring studies,15 approximately 40 percent of the mentors received course credit or were otherwise required to participate as mentors. In both studies, fewer benefits were found in matches with compensated mentors.

      Staff contact, support, and communication.

      The value of staff support was revealed in the High School Bigs study as well, in that it was positively associated with mentors’ views of relationship quality and satisfaction with the program.21 Yet, providing high levels of staff support (defined as mentors’ perception that the program coordinator is available to talk, concerned about their experience, and interested in how the match is going), is especially difficult when many matches are meeting at the same time, and when the program leader also coordinates planned activities.22 This is the one of the main reasons the Cross-Age Mentoring Program13 and the revised High School Bigs Demonstration Model23 involve more experienced mentors as teen leaders to guide activities for the matches. This allows the program coordinator to observe matches, talk with matches, and deal with inevitable crises that arise, such as dealing with an angry parent or bus driver, or helping a student who is in crisis.


      Herrera and colleagues found that mentees reported more satisfying relationships when their mentors received more training.10 However, when Karcher and colleagues examined the contribution of hours of training on match quality,21 they found that staff support during the match was more directly influential and that additional training affected mentor-mentee relationship quality in large part by influencing the nature of the conversations and activities through which matches chose to engage. They also found that hours of training may be a double-edged sword, in that more hours of training predicted satisfaction with the program, goals, and supervision, but it also was negatively associated with the likelihood mentors would elect to mentor in the future. Therefore, the effect of additional training may be a function of the type of training, as teen mentors felt most positively about the program when they received training on youth and relationship development.

      Help structuring interactions.

      A discussion of program activities and the use of curricula as a moderator of program effectiveness must start by acknowledging that a primary criterion used for excluding studies from this review was whether the program description assigned significantly more weight or importance to curricular content than to relationship development as the main mechanism of change. Karcher describes how challenging yet important it is to train mentors to be comfortable using curricular activities, while also recognizing when to abandon learning in service of opportunities to deepen the relationship.24

      Teen mentors seem less able than adults to understand the power of a strong mentor-mentee relationship on their mentees’ lives. This is a developmentally appropriate constraint in their thinking.5 But it explains why reanalysis of the High School Bigs found that, even though relationship-focused and casual conversations predicted stronger relationships, mentors in matches engaging in more of these nondirective conversations viewed the quality of the program less positively.21 Conversely, teen mentors in matches engaging in more goal-directed conversation and structured activities held more positive views of the program (but did not seem to influence the quality of the relationships). The associations were the opposite of what predicted positive views of the program for adult mentors, suggesting that giving teen mentors guidance in how to structure their time (e.g., using structured or “inert” curricular activities) may be a uniquely important predictor of having a positive experience for teen mentors.

      Characteristics of mentees.

      Many teens are not prepared to develop relationships or even guide the interactions of younger youth. Relatively more mature teens can effectively cultivate relationships with younger children, but even they need consistent staff support, clear program structure, and training to engage in prosocial, positive, and mentee-focused interactions with behaviorally difficulty children. In both the Cross-Age Mentoring Program and the High School Bigs studies, having a more behaviorally difficult mentee predicted lower quality relationships (mentor-reported),25 inconsistent mentor attendance,26 and the mentor was less likely to continue in the program.10,21 Additionally, not all children may need or benefit from teen mentoring, especially when provided en masse. In the BBBS Edmonton program study, gains over time on several outcomes among mentees were greater when the mentees were identified individually rather than enrolled in whole groups (e.g., a whole classroom).15

      Characteristics of mentors.

      Some teens are better suited than others to serve as mentors to children. Although there has been little research on personality characteristics of teen mentors (e.g., outgoingness) and their effect on program outcomes, several studies have looked at attitudinal and motivational characteristics beyond whether or not their participation is compulsory or compensated. Research on BBBS High School Bigs, the BBBS Edmonton program, and CAMP has identified several attitudinal and motivational characteristics of mentors that explain variability in outcomes. These are the positive associations between program outcomes and teen’s attitudes toward youth in general and their degree of other-centeredness or social interest. Secondary analyses of the High School Bigs program revealed that teen mentors who reported more positive attitudes toward youth were particularly effective with more academically disconnected mentees (there were negative effects for academically connected mentees matched with teen mentors holding more negative views of youth).27 Similarly, a social interest scale was included as a screening and training28 tool in one program because of research linking mentors’ social interest to program outcomes.26, 24 These studies suggest that recruiting a specific type of mentor may be critical to program effectiveness, and that motivation for self-enhancement25 and compulsory participation15 may undermine relationship quality.

      Parent involvement.

      Holding family events seems to be important to program outcomes. It has been found to predict greater satisfaction with the match among mentors.25 The use of these events may explain why increases in mentees’ connectedness to parents and family functioning are more common in the cross-age peer than adult-youth mentoring studies.10, 20, 14


      1. Teen mentors may need and benefit more than adult mentors from staff support, program structure (e.g., planned activities), and ongoing training.

      2. Mentors should not be coerced or lured into mentoring because of the potential negative consequences for the mentees to whom these disinterested mentors are assigned.

      3. Mentors who hold more positive attitudes toward youth in their community, who are motivated to help, and who report greater social (rather than self-) interest should be selected.

      4. Although parent involvement has not been the specific focus of research in cross-age peer mentoring programs, evidence exists that engaging parents in these programs through family events may be useful in facilitating improvements in mentees’ connectedness to their parents.

  • What Intervening Processes Are Most Important in Linking One-to-One Cross-Age Peer Mentoring to Youth Outcomes?


      Similar to other mentoring approaches, the interpersonal relationships formed between mentors and mentees have been posited to be the primary link between cross-age peer mentoring and youth outcomes. Whereas mediators of program effectiveness for group mentoring include improvement in relationships with same-age peers as well as the mentoring relationship, the mechanisms of change most commonly presented for cross-age peer mentoring programs are similar to models explaining how adult-youth mentoring relationships influence program outcomes. Most cross-age mentoring logic models resemble or reference Rhodes’ model, suggesting the trusting, empathic, and reciprocal relationship with a mentor is what leverages changes in social, cognitive, and identity development. Notwithstanding are the studies under the moniker of “peer mentoring” that describe only the activities inherent in their youth-delivered curriculum that are presumed to influence changes in directly paralleled skill or attitudinal changes, and which lack any reference to the quality of the relationship or to the essential role of the mentoring relationship in achieving program outcomes.i In this section, two primary models used to explain the effectiveness of adult-youth mentoring relationships are used to frame research in cross-age peer mentoring programs.

      Although an important moderator of program effectiveness for cross-age peer mentoring programs (more than adult-youth mentoring programs) may be the effective use of some organized activities or loosely adhered to curriculum, the nature of how specific curricular focus influences outcomes is a question about the mechanisms of change. Consider the phenomenon reported earlier in which teens reported stronger relationships but viewed the program less positively when they engaged in more casual, verbal interactions. Clearly the degree of program structure (i.e., the provision of activities, clarity of program goals and focus on relationship, and availability of staff to guide matches) influences program effectiveness, but it is the experience of these interactions that explains when and how positive relationships form and effect better program outcomes. This section describes research on how activity type and activity decision-making seem to contribute to influence program outcomes.

      One framework29 that is useful for understanding how mentoring interactions influence program outcomes contrasts two approaches found in the adult mentoring literature. In that framework, the approach that focuses first on befriending the mentee (the developmental style) was found most helpful with younger mentees, and the more goal-directed (instrumental) style more consistent with the expectations and needs of teenage mentees. For this reason, the Cross-Age Mentoring Program is structured to provide the developmental style of mentoring, in which more relational structured activities are used to guide match development before shifting to activities related more to school and social skills building activities. This is consistent with the view4 that trusting and empathic relationships are the nexus of effective mentoring relationships.

      i Furthermore, “peer mentoring” studies of programs utilizing both a heavily curricular focus in which 1 or 2 youth teach groups of 10 to 12 “mentees” commonly proffer that the program influences outcomes by influencing relationships among peers more so than by the formation of close mentor-mentee relationships. This may be, in part, because the formation of strong dyadic mentor-mentee relationships is viewed as less frequently occurring at the same degree of intensity in group mentoring as in one-to-one mentoring (see Kuperminc’s Group Mentoring NMRC review). The mentor-mentee relationship becomes an additional process contributing to positive program outcomes in cross-age peer-group mentoring, rather than the primary, necessary and sufficient mechanism of change as in one-to-one cross-age peer mentoring.


      Program interaction approach.

      To test the benefit of this developmental approach, one study contrasted it against two other approaches (or alternative treatments).14 One reversed the order of activities to start with skill and goal-focused activities. The second used delinquency prevention activities throughout. They observed the positive changes among mentees after peer mentoring described earlier, but only in the program formats that started the program with relationship-building activities and later progressed to more structured, goal-directed activities. Research on BBBS of Canada’s teen mentoring programs found that the least goal-directed approach yielded the most positive changes over time.15

      Collaborative decision-making.

      The second element of the activities framework introduced earlier is the hypothesis that the degree of collaboration that occurs in matches is directly proportional to the magnitude of program outcomes. The process of joint ownership, of collaborating in making decisions about what to do, allows reciprocal, back-and-forth negotiations to take place; expressions of empathy; and the formation of trust. Yet, if cross-age peer mentoring programs must help teens structure their interactions, then the use of a pre-planned, nonnegotiable set of curricular activities may pose a problem. It reflects the staff unilaterally imposing or providing a structure rather than the participants having co-ownership of what they do. Of course, matches also could be relatively unstructured and still dominated by one participant, and thereby not reciprocal or mutually agreed upon or negotiated.

      To test this hypothesis about the need for mutuality, reciprocal negotiation, and collaboration, secondary analyses of the High School Bigs study data were conducted that contrasted the ways in which decisions about what to do were made.30 Based on mentors’ reports, match activities were determined primarily by staff, decided by either the mentor or mentee, or negotiated (“decided together”). The focus of these analyses was not only to see if one decision-making approach was superior, but also to see what types of interactions took place when matches decided together. Did their feelings about the relationships differ accordingly?

      What the researchers found was that in collaborative matches, the more “developmental” approach took place.30 The matches spent more time talking about friends and family, and played more sports and indoor games. The mentors reported higher relationship quality, felt more efficacious as mentors, and saw their mentees asking for their help more frequently. The mentees were more engaged, satisfied, and happy, and perceived the mentor to be more youth centered.

      Unilateral decision-making predicted fewer positive experiences.30 When Littles chose, they avoided goal-directed activities, while when Bigs chose they avoided more relational, playful activities. In both cases, the mentors and mentees found their matches less satisfying, engaging, and purposeful. When staff determined what matches would do, neither the mentor nor the mentee experienced the match positively. Overall, collaborative matches were most successful. The takeaway is that although having some prescribed curricular activities seems important to frame the relationship, the program structures in which mentoring really happens—where both participants feel engaged, the mentees feel safe to ask for support, and the mentors feel confident and valued—flex to the unique needs and experiences of the match. This adds a level of complexity to program development, coordination, mentor training, and delivery.

      These findings suggest the mechanisms of change in cross-age peer mentoring seem not unlike what research shows affords positive outcomes from adult-youth mentoring. The conditions in which the relationships develop, however, and specifically the nature of the structured support needed by teenage mentors, differs from that of adults. But too much structure, which is too rigidly adhered to, can cripple a match’s development and should not be considered peer mentoring. Program structure is just that—a structure which can hold the relationships and allow them to grow. Like a plant in a pot, program structure is a container in which the relationship can develop in its own way. Structuring this balance is not easy, however.


      1. Establishing a relationship first seems critical to generating the experiential building blocks of a mentoring relationship—empathy, trust, mutuality, and reciprocity.

      2. For teenage mentors, more structure is typically needed to create the conditions for befriending to occur between cross-age peers.

      3. Some guidance and activity advice (“interaction structure”) may be needed for mentors to feel competent and efficacious, but too much could feel stifling and deflating to the youth.

      4. Teenage mentors may need help to become flexibly reliant on prescribed or curricular activities, and require training in how to grow the relationship by strategically diverting into personal discussions instead of the provided task.

  • Have One-to-One Cross-Age Peer Mentoring Programs and Supports Reached and Engaged Targeted Youth and Been Implemented with High Quality?


      Beyond the moderating factors described above and the mediating role of mentoring interactions described earlier, little is known about what is unique to setting up a sustainable, efficacious cross-age peer mentoring program other than what can be gleaned by comparing programs with varying levels of best practices in terms of their relative effectiveness. Yet even this is not easy to do because of the dearth of rigorous studies that would allow true comparisons across programs and the tendency to not measure the specific contributions of their practices.

      Two anecdotes shed some light on the need for attention to what is missing in most cross-age peer mentoring programs. When the developer of the Cross-Age Mentoring Program worked with Michael Garringer and others at Education Northwest to package the program for dissemination,24 it took two years to supplement the materials used in those studies with planning guides, forms, and other procedural descriptions necessary to allow it to be replicated with fidelity in other contexts, such as in studies by independent researchers and by practitioners14, 7 in more than two dozen other school contexts.31 A second example is how the Big Brothers Big Sisters of America organization took the findings from the High School Bigs program to redesign it. Seeing the effects of this program only evident under some of the conditions described earlier, they developed a revised model,23 which may now be ready for an effectiveness study.


      To expand beyond what can be learned from the study of practices reported in outcome research, this section presents findings from a Delphi study including interviews with thirteen experts in the field of cross-age peer mentoring and survey data from 623 program coordinators.32 All participants, through interview and survey questions were asked what strategies effectively are most essential and pose the greatest challenges to the successful implementation delivery of cross-age peer mentoring programs. Their responses revealed four primary categories pertaining to the quality of the program coordinator; recruitment and preparation of mentors, support from school administration, and the need for program structure. The findings provide a unique, high-level (not specific to any one program) look at the challenges and strategies for overcoming barriers to successful program implementation.

      Program coordinators.


      One of the biggest challenges the experts referenced was “selecting unsuccessful program coordinators.” This occurred particularly in instances when coordinators did not express prior interest in leading a program but were instead assigned by a district office or school administrator. In these instances, coordinators were depicted as lacking the necessary motivation and sufficient time to focus on the many implementation logistics or match supervision. Turnover of program coordinators was also identified as a noteworthy challenge.


      In describing the characteristics of quality adult leaders, experts discussed the importance of coordinators’ commitment level, organizational and communication skills, approachability, and experience with that or similar programs. To compensate for a leader not having some of these qualities, several experts suggested creating stakeholder teams (comprised of administrators, faculty, counselors, etc.) to oversee some aspects of the program and serve as an advisory board. Those individuals also retain institutional knowledge that is critically important when there are changes in program coordinators or school administrators. Similarly, having a backup or co-coordinator also helps when a primary coordinator leaves the program.

      Recruiting reliable program peer mentors.


      Competition for students’ interest can dilute the applicant pool and/or weaken the commitment level of student mentors. Experts also warned against administrators forcing program coordinators to pick students for ulterior motives, even when those motives are benevolent, such as desiring a growth opportunity for a potential student leader. Many respondents felt that when students were not intrinsically motivated or fully aware of the demands of the program, mentors’ poor attendance and high attrition negatively affected both the program and mentees.


      Experts and coordinators emphasized the need for multiple promotional meetings to explain the program and its expectations to potential mentors. At these events, current mentors could answer questions based on their own experiences. Many coordinators also recommended advertising the program in classes, at assemblies, and through flyers and social media.

      In terms of the selection process, experts described the importance of selecting empathetic peer mentors who are motivated to help others (e.g., high in social interest26) and who have the time and ability to work closely with both peers and adults. They recommended using teacher and current peer mentor references in addition to interviews and behavior checks.

      Experts also emphasized the value of selecting a pool of students who mirror the mentee population in terms of background, interests, activities, academic records, personalities, and personal experiences. They felt this builds a sense of credibility among potential mentees and helps them feel connected with at least some of the mentors.

      Student training and preparing for the worst.


      Experts noted that programs providing poor or insufficient mentor training could negatively affect mentees. To address this potential, both experts and coordinators suggested a mandatory, multiday intensive training retreat at the beginning of the program along with ongoing training in necessary skills such as active listening and perspective taking, ethical conduct, confidentiality, and cultural factors. A somewhat unique perspective raised by many program coordinators in this study was the need to prepare mentors and administrators alike for what could go wrong.


      Prior to program launch, administrators and program coordinators should discuss how they want to handle the mistakes student leaders will inevitably make, such as when mentors do not follow through on their responsibilities. Experts and coordinators recommend creating some form of contract or code of ethics to clarify the frequency of absences that will be tolerated, and ensure all mentors receive training in these standards.

      Importance of program structure.


      Experts and coordinators described challenges with structuring consistent meeting times between mentors and mentees, and between program coordinators and mentors. They also emphasized an essential balance of structuring a curricular plan while still flexibly responding to match-specific needs and to relationship development more generally. Several described their curricula as at times feeling “scripted,” and the importance of adapting it to the needs of individual communities or matches.


      Experts discussed the importance of empowering teens to help create their own curriculum. Encouraging students to take ownership over portions of curriculum development was described as fostering responsibility and addressing local needs so that match activities do not feel “imposed from outside.” As one example of this from the research literature, the Cross-Age Mentoring Program: Connectedness Curriculum gives detailed instructions about creating a curriculum for one’s specific program that is consistent with the program logic model.33

      Ultimately it is important to note that the points made above, and on which the conclusions below are based, are derived from interviews and surveys from experts and program coordinators, but not from research on program practices that were specifically found to influence program outcomes. In addition, it is not clear how many of the 623 program coordinators who responded to the survey ran programs that align with the definition of cross-age peer mentoring used in this review. The following recommendations draw on considerable clinical wisdom and direct attention to practices that may be very important and provide hypotheses which subsequent research may be able to test empirically. Further, the degree to which such practices are implemented with fidelity—consistently, completely, over time—is critical and depends on documentation of program procedures and checklists used to ensure implementation fidelity. Only one of the seven programs in this review has made its implementation and fidelity checklists publicly available.33


      1. Staffing is critical to successful program implementation and sustainability. Effective coordinators are interested in leading the program, are well trained, possess the necessary organizational and leadership skills, and are effective at securing the resources they need.

      2. Program coordinators (in many ways like peer mentors) need support, co-coordinators, and an active advisory board involved in program operations. They should also work continuously to secure and maintain buy-in from school administrations.

      3. Teen mentors, like all mentors, need extensive initial and sufficient ongoing training. This should include information on program parameters and training on all the necessary skills to be an effective mentor. Training for teen mentors should also prepare them for responding to potential worst-case scenarios. Similarly, administrators and coordinators should develop a response protocol for those occasions in which mentors make mistakes.

      4. When choosing program curriculum or other activities to organize the matches, be sure relationship development can be prioritized, and consider allowing students to guide activity development to make curricula relevant to local needs.

  • Implications for Practice

    (Mike Garringer, MENTOR: The National Mentoring Partnership)

    While the evidence on the effectiveness and mechanisms of cross-age peer mentoring as defined in this review is fairly limited, there is still a lot we can learn here about how to design, structure, staff, and lead a cross-age peer mentoring program from the research we do have. Like all good “evidence-based” practice, these tips for practitioners represent a blend of findings from rigorous research and the invaluable insights of those that have run programs of this type (from the Delphi study32 of those 623 program coordinators) and learned many lessons about what facilitates success (or perhaps what dooms a program to ongoing struggles).

    In considering the evidence presented in this review, there are four main principles that we would encourage mentoring programs to follow. These are applicable to both existing efforts that want to improve services and brand new start-up efforts, which can bake these principles and philosophies into the program from an early stage.


      The review notes in several places that having the right person, or people, leading the program is critical to its success. While that seems like an obvious truism, the reality is that peer mentoring programs are often led by a teacher or other school staff member for whom running the program was presented as a required task, not a labor of love. Such individuals may not have the requisite enthusiasm to champion the program, engage administrators, lobby for access to school or budget resources, engage parents and mentors, and set a tone of positive support. The study noted in the review stressed the need to have someone in this lead role with the proper organizational skills and the drive to make the program succeed.

      If a school or program finds that one individual hard to come by, one strategy noted in the review is to manage the program through a “stakeholder team” that can distribute some of the tasks associated with the program, bring institutional knowledge and access, and serve as a back-up when one supervisor of matches is not available for some reason. This can also help build deeper buy-in for the program by bringing more champions to the table. Any mentoring program is only as good as its coordinator, but peer programs seem to be especially susceptible to the negative impact of insufficient leadership.

      Closely tied to the leadership of the program is ensuring that the mentors themselves are also the right individuals to do the task at hand. Research cited in this review emphasized the need to create a diverse pool of peer mentors that reflects the general makeup of the school or community. Often, peer programs overemphasize recruiting students who are considered leaders in the school or who are generally “popular” and engaging. But these students are often pressed for time, might not have enthusiasm for adding another task to their plate, and may be too homogenous to bring the level of diversity and difference needed to properly serve as a role model for diverse mentees.

      The review also cautions against incentivizing the mentor role through course credits or other rewards, especially if those incentives can be earned before the matches are set to close. Ideally, programs would want to engage mentors who want to serve in this role primarily for legitimate reasons of caring and altruism, rather than those that are only motivated by the incentive or “prize” offered by the program. So think carefully about which students or older youth would mirror the diversity of your mentees and would bring the proper motivation and skills as mentors.


      Peer mentoring programs often conflate the program activities with the purpose and value of the program as a whole. In fact, many such programs were excluded from this review due to the lack of clarity in the studies on those programs about the role that mentoring played in the program. Because older youth and their mentees need something to do together when they meet, and because just having discussions about life in general as well as sensitive topics can be challenging to mentors who are young people themselves, it can be tempting for program staff to stuff the program full of games and skill-building exercises and homework help and a thousand other activities so that matches are always “on the go” and doing some task. While this may make it seem like matches are busy and working on things, this approach tends to substitute activity for action and somewhat misses the point of peer mentoring, which is the relationship itself.

      The review notes that while activities can be purposeful in nature and focused on some aspect of the youths’ development or skill-building, programs also need to provide opportunities to be relational within activities. That is, they must provide opportunities for the mentor and mentee to work collaboratively, share perspectives and engage in discussion, build trust and mutuality, and enjoy being with each other. This can be a difficult balance to achieve, especially for programs that have promised administrators or funders big impacts on key areas of interest (grades, student behavior, etc.) as a result of the mentoring.

      But as the review authors note, a program that provides too much structure, and activities that overly restrict the mentor-mentee pairs’ interactions, is almost doomed to fail if the goal is primarily teaching information or learning specific skills. They note that all activities must be designed to help participants feel engaged with one another, help mentees feel safe and supported, and let the older mentors feel valued and confident in what they are doing. So regardless of the adult-defined outcomes of a peer mentoring program, chances are that a review of the program’s match activities, with an eye on whether they build or restrict relationship growth, will likely result in developing stronger matches and better results. So, make sure your peer mentoring program actually has an emphasis on mentoring!

      The review authors also note a related concern around activities that is often overlooked in programs: when a pair should deviate from an activity. As they note, forcing matches to stick it out through tasks or activities that are not working for one or both of the participants is unlikely to yield anything positive. Mentors should be empowered to break away from an activity if their mentees are feeling frustrated or have some other issue going on that is preventing them from engaging in the activity. Peer mentors should be trained on how and when to pause an activity and how to involve the supervising adults in getting them back on track or providing additional support to a mentee that is struggling to participate for some reason. Teen mentors shouldn’t be left on their own to navigate complicated situations or work with a mentee in crisis. Having a program coordinator positioned to provide support when an activity goes awry or when a mentor or mentee is struggling relationally is really important to the success of cross-age peer programs like the ones emphasized in this review.


      Regardless of the activities that you ultimately select for the program, peer mentors will need to be trained on not only how to do the activity, but also how to be relational while doing it. As adults, we often have considerable experience in interacting with unfamiliar people and building rapport in myriad of social situations and settings. Those things do not come easily or naturally to most teenagers, including those in a mentoring role for a younger child. Spend time training peer mentors to engage fully with their mentees, to stay on task (teens are easily distracted), and to prepare for all of the things that could go wrong during a particular activity (or in general in the relationship). The review notes that peer mentors can often feel unsupported or frustrated if their mentee is not engaging with them or if an activity isn’t going as planned. Once again, adequate supervision by a coordinator or other supportive adult can be critical in mitigating this frustration, but mentors should also be trained on strategies for handling such situations on their own if that adult support isn’t immediately available.

      One of the many delicate balancing acts that adults in these programs must do, is to not be too assertive into these relationships, while also being an asset when they do experience challenges. If adults are too directive or controlling of what is happening between mentors and mentees, the youth participants don’t get a chance to bond and build the relational skills that are the driving force behind personal growth in these programs. On the other hand, if matches are under-supervised, it can lead to a host of problems: frustrated mentors, mentees who stop engaging, and even negative role modeling in situations where the mentors start interacting with each other (often in inappropriate ways) instead of with their mentees. If your program has a stakeholder team, spend some time discussing how to be an asset to matches while resisting the urge to step in and simply “run things.”


      Most cross-age peer mentoring programs will get developed because of adult needs and desires⎯improved grades, better attendance, increased positive behaviors and a less disruptive school climate, improved attitudes about school, and so on. But while those motivations may lead to the initial development of a program, a strong peer mentoring model will eventually allow the young people themselves to plan, develop, and implement the program over time as much as possible. Although this is only touched on briefly in the review, successful peer mentoring programs like CAMP33 let the young people themselves adapt and improve the program over time.

      Programs that want to increase youth leadership in the implementation and growth of the model should provide opportunities for:

      • Former mentees in the program to join later as mentors. Having been on the other side of the relationship gives them a unique perspective and understanding of how to be successful in the mentor role.

      • Mentors to change and adapt the activity curriculum from year to year. No one is probably more aware of which activities worked best (or did not) and how they were best presented than the mentors themselves. Give them the chance to refine the curriculum at the end of every year and truly make the program something that they “own” and pass on to future classes. This is also a great way to make the program responsive to issues in the school or program setting that adults may be unaware of, but that the young people themselves feel need to be addressed.

      • Former mentors to recruit the next class of mentors. As noted in the review, former mentors can be tremendous assets in recruiting the next batch and helping to get a diverse and representative pool of qualified and motivated mentors for the next year.

  • References

    1. Karcher, M. J. (2007). Cross-Age Peer Mentoring. In MENTOR’s Youth Mentoring: Research in Action series. Alexandria, VA: National Mentoring Partnership.

    2. Karcher, M. J. (2005a). Cross-age peer mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 266–285). Thousand Oaks, CA: Sage Publications.

    3. Karcher, M. J. (2013). Cross-age peer mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp. 233–258). Thousand Oaks, CA: SAGE.

    4. Rhodes, J. E. (2002). Stand by me: The risks and rewards of youth mentoring. Cambridge, MA: Harvard University Press.

    5. Noam. G. G., Malti, T. K., & Karcher, M. J. (2013). Mentoring in developmental perspective. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp. 99–116). Thousand Oaks, CA: SAGE.

    6. Akos, P. T. (2000). Mentoring in the middle: The effectiveness of a school-based peer mentoring program. Unpublished Dissertation. Charlottesville, VA: University of Virginia.

    7. Smith, L. H. (2011). Piloting the use of teen mentors to promote a healthy diet and physical activity among children in Appalachia. Journal for Specialists in Pediatric Nursing, 16, 16–26. doi: 10.1037/0278-6133.24.3.235

    8. Johnson, V. L., Simon, P. P., & Mun, E-Y. (2014). A Peer-Led High School Transition Program Increases Graduation Rates Among Latino Males. The Journal of Educational Research, 107, 186–196. doi:10.1080/00220671.2013.788991

    9. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, 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. doi:10.1177/1529100611414806

    10. Herrera, C., Kauh, T. J., Cooney, S. M., Grossman, J. B., & McMaken, J. (2008). High School Students as Mentors: Findings from the Big Brothers Big Sisters School-Based Mentoring Impact Study. Philadelphia, PA: Public/Private Ventures.

    11. Hwang, N. (2015). Mentor age and youth development outcomes in school-based mentoring program. Presentation at the annual conference of the Society for Research on Educational Effectiveness. ERIC document ED562351.

    12. Bernstein, L., Rappaport, C. D., Olsho, L., Hunt, D., & Levin, M. (2009). Impact Evaluation of the US Department of Education’s Student Mentoring Program. Final Report. NCEE 2009–4047. Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education

    13. Karcher, M. J. (2008). The Cross-Age Mentoring Program (CAMP): A developmental intervention for promoting students’ connectedness across grade levels. Professional School Counseling, 12(2), 137–143.

    14. Sar, B. K., & Sterrett, E. M. (2011). Investigation of the effectiveness of a developmental mentoring model as an intervention/prevention strategy for juveniles of varying levels of risk among middle school youth in Metro Louisville. Final Report for Award 2011-JU-FX-018 funded by the Office of Juvenile Justice and Delinquency Prevention. NCJRS document 244758.

    15. Cavell, T. A., Gregus, S. J., Craig, J. T., Pastrana, F. A., & Rodriguez, J. H. (2017). Program-Specific Practices and Outcomes for High School Mentors and Mentees. Unpublished manuscript. Fayetteville, AK: University of Arkansas. Also, Cavell, T. A. (2013). Teen mentoring in Canada: Are outcomes related to program structure? Banff, Alberta, Canada: National Mentoring Symposium.

    16. Sheehan, K., DiCara, J. A., LeBailly, S., & Christoffel, K. K. (1999). Adapting the gang model: Peer mentoring for violence prevention. Pediatrics, 104(1), 50–54. doi:10.1542/peds.104.1.50

    17. Herrera, C., Grossman, J., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Making a Difference in School: The Big Brothers Big Sisters School-Based Mentoring Impact Study. Philadelphia, PA: Public/Private Ventures.

    18. Herrera, C., Grossman, J., Kauh, T. J., & McMaken, J. (2011). Mentoring in Schools: Impact Study of Big Brothers Big Sisters School-Based Mentoring. Child Development, 82, 346–361. doi: 10.1111/j.1467-8624.2010.01559.x

    19. Karcher, M. J. (2005). The effects of school-based developmental mentoring and mentors’ attendance on mentees’ self-esteem, behavior, and connectedness. Psychology in the Schools, 42, 65–77. doi:10.1002/pits.20025

    20. Karcher, M. J., Davis, C., & Powell, B. (2002). Developmental mentoring in the schools: Testing connectedness as a mediating variable in the promotion of academic achievement. The School Community Journal, 12, 36–52.

    21. Karcher, M. J., Hansen, K., Herrera, C., & Crisp, G. (2011). The right focus: How mentor, mentee and program characteristics contribute to program and relationship quality. Poster presented at the 2011 Biennial Meeting of the Society for Research in Child Development. Montreal, Quebec, Canada.

    22. Karcher, M. J., McClatchy, K., Wong, M., Zholu, Y., & Avera, J. (2013). InspireU Program Formative Evaluation: Final Report for the Office of the Mayor of San Antonio. Unpublished report. San Antonio, TX: University of Texas at San Antonio.

    23. Hansen, K. (2010). High School Bigs Model: Draft first year (with research evidence and essential elements). Philadelphia, PA: Big Brothers Big Sisters of America.

    24. Karcher, M. J. (2012a). The Cross-Age Mentoring Program (CAMP) for children with adolescent mentors: Training Guide. San Antonio, TX: Developmental Press.

    25. Karcher, M. J., Nakkula, M. J., Harris, J. (2005). Developmental mentoring match characteristics: The effects of mentors’ efficacy and mentees’ emotional support seeking on the perceived quality of mentoring relationships. Journal of Primary Prevention, 26, 93–110. doi:10.1007/s10935-005-1847-x

    26. Karcher, M. J., & Lindwall, J. (2003). Social interest, connectedness, and challenging experiences. What makes high school mentors persist? Journal of Individual Psychology, 59, 293–315.

    27. Karcher, M. J., Davidson, A., Rhodes, J. E., Herrera, C. (2010). Pygmalion in the program: The role of teenage mentors’ attitudes in shaping their mentees’ outcomes. Applied Developmental Science, 14, 212–227. doi:10.1080/10888691.2010.516188

    28. See free 1-hour online training at

    29. Karcher, M. J., & Nakkula, M. J. (2010). Youth mentoring with a balanced focus, shared purpose, and collaborative interactions. New Directions in Youth Development, 126 (Summer), 13–32. doi:10.1002/yd.347

    30. Karcher, M. J., Herrera, C., & Hansen, K. (2010). “I dunno, what do you wanna do?”: Testing a framework to guide mentor training and activity selection. New Directions for Youth Development, 126, 51–69. doi:10.1002/yd.349

    31. Boy with a Ball (2012). Fixing academic failure: The Boy with a Ball Harlandale Cross-Age Mentoring Program. Retrieved from

    32. Berger, J. R. M. (2016). The Implementation of School-Based Peer Programs: Successes, Challenges, and Solutions. Los Angeles, CA: UCLA: Education. Retrieved from:

    33. Karcher, M. J. (2012b). The Cross-age Mentoring Program (CAMP) for children with adolescent mentors: Connectedness Curriculum. San Antonio, TX: Developmental Press.

  • Tables

The National Mentoring Resource Center (NMRC) website offers a few guidebooks and resources that will be helpful to those managing a cross-age peer mentoring program:

Other online resources on peer mentoring

You can also find several tools and activity guides that can support group 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 group mentoring program.

Mentoring for Youth in Foster Care
PDF button Facebook button Twitter button

September 2017

This review examines research on mentoring youth in foster care. The review is organized around four questions:

  1. What is the effectiveness of mentoring for youth in foster care?

  2. What factors influence the effectiveness of mentoring for youth in foster care?

  3. What pathways are most important in linking mentoring to outcomes for youth in foster care?

  4. To what extent have mentoring initiatives for youth in foster care reached and engaged these youth, been implemented with high quality, and been adopted and sustained?

Research on mentoring youth in foster care is emerging. Several studies of program-based mentoring have employed rigorous designs, and studies of both program and natural mentoring are beginning to shed light on the conditions and processes that may be required to optimize benefits to youth. Because of the interpersonal vulnerability and high potential for adverse outcomes among this population, great care and coordination is required for implementing mentoring programs and supporting natural mentoring relationships. If done well, the benefits of mentoring may outweigh the potential risks of mentoring and foster youth may experience a range of positive outcomes. The existing evidence points toward several conclusions:

  • Both natural and program-based mentoring appear to be highly acceptable to youth in foster care, and mentees generally report high satisfaction with their mentoring experiences.
  • Available research suggests that mentoring for children in foster care (across a range of ages and mentoring formats) can have positive impacts on many, but not all, targeted outcomes, including mental health, educational functioning and attainment, peer relationships, placement outcomes, and life satisfaction.
  • Most formal mentoring programs that have been evaluated to date are multicomponent (that is, they include components other than one-to-one mentoring, such as skills groups) and utilize mentors who are agency staff members or university students.
  • The impact of mentoring may differ based on demographic, and placement characteristics and key processes, such as improvements in self-determination and prosocial skills, may be the mechanisms through which mentoring outcomes are realized for this population.
  • Finally, although there are many conceptual reasons why mentoring is an excellent fit for youth in foster care, there are pragmatic challenges that make widespread implementation difficult and no studies have examined program expansion or adaptation.

The review concludes with insights and recommendations for practice based on currently available knowledge. These insights highlight a number of factors to consider when developing and implementing mentoring programs for youth in foster care. Practitioners are encouraged to keep in mind that these youth may have challenges in engaging in mentoring relationships as a result of adverse experiences. Therefore, mentoring programs wishing to recruit, engage, and retain youth in foster care may need to access clinical expertise and develop collaborative relationships with agencies and professionals serving these youth. Programs should train and support their mentors to understand the critical importance of consistency, patience, and building and maintaining trust when working with these youth. Additionally, programs should consider incorporating activities that promote self-determination and goal setting and prepare youth for independent living, including the ability to build their social network and reconnect with significant adult supports.

  • Introduction

    Estimates of the number of youth in foster care in the United States have held steady at approximately 400,000 on any given day in the past five years.1 Nearly half live in non-relative foster homes, 30 percent in relative foster homes (referred to as “kinship care”), 8 percent in institutions, 6 percent in group homes, and about 5 percent live in other placements (e.g., pre-adoptive homes) or have run away. Many youth experience multiple placements while in care, and some move in and out of the system throughout their childhood. About 10 percent of youth who exit foster care “age out” or emancipate, defined as reaching age 18 without achieving permanency, such as adoption or reunification with their biological families.1

    The experience of substantial and traumatic adversity (e.g., abuse and neglect, exposure to substance use and violence, chronic disruptions in school and living situations, abandonment) is unsurprisingly linked with diminished physical and mental health, academic underachievement and school dropout, problematic substance use, poverty and homelessness, and incarceration.2, 3 Despite all odds, however, some youth avoid this negative trajectory.4, 5, 6 The presence of at least one supportive adult may help create the context through which resilience (i.e., the maintenance of positive adaptation despite experiences of significant adversity)7 is possible even in the face of maltreatment and foster care placement. Resilience is a dynamic process that involves more than individual strengths; external resources and the presence of larger support systems are necessary for children to overcome adversity.7 Increasing the number and quality of significant figures of support (e.g., mentors) available to youth increases their chances of healthy development.8, 9, 10

    Attachment theory posits that early relationship experiences with primary caregivers set the stage for future close relationships.11 When children have positive, secure attachments with caregivers, they develop appropriate “working models” (i.e., a set of expectations and beliefs about oneself, others, and the relationship between self and others) and glean benefits in the form of healthy relationships and positive youth outcomes. When children lack secure attachments due, for example, to abandonment, maltreatment, or placement in foster care, their working models are negatively distorted.12 Youth in foster care may believe they are unworthy of love, see hostility when others’ behavior is neutral, and be fearful of trusting people, which is perpetuated by additional negative relationship experiences. Fortunately, these models are amenable to change.13 “Corrective experiences” can shift working models and a healthy mentoring relationship can be one such experience. Although repairing attachment injuries may also require professional therapeutic intervention, mentors can buffer the impact of early and persistent exposure to adversity by filling an important relational void in the lives of youth in foster care through consistent, meaningful interactions.12 Over time, youth in foster care who experience positive relationships with mentors and others can alter their working models of relationships to enable them to form healthy relationships.

    Developmental systems and ecological theories also emphasize the role adults play in role modeling and facilitating social bonding across contexts, which increases social capital and the capacity for closeness.14, 15, 16, 17 Social scaffolding (i.e., the process through which adults provide guidance for youth in developing relationships and support networks) is particularly critical for understanding mentoring of youth in foster care. When social scaffolding is absent, youth in foster care come to rely only on themselves and may view dependence on others as a personal weakness or failure.18 Youth who age out of foster care without strong social scaffolding may experience what is referred to as psychological homelessness—a longing for “home” and enduring connections.19

    There is no doubt that youth in foster care, and those aging out of care, need support. Mentors are one of many essential resources. Via consistent, repetitive, and positive relational experiences, mentoring can foster resilience. Youth in foster care may be particularly responsive to the opportunity of a new relational experience but, because of their interpersonal vulnerabilities and complex needs, a thoughtful and cautious approach to mentoring this population is warranted.20, 21 The focus of this review is to examine whether natural and/or program-based mentoring can provide supportive relationships and achieve the intended benefits without producing any unintended negative consequences for youth in foster care. More specifically, this review addresses the following questions:

    1. What is the effectiveness of mentoring for youth in foster care?

    2. What factors influence the effectiveness of mentoring for youth in foster care?

    3. What pathways are most important in linking mentoring to outcomes for youth in foster care?

    4. To what extent have mentoring initiatives for youth in foster care reached and engaged these youth, been implemented with high quality, and been adopted and sustained?

    This review examines studies of individual and group mentoring (with or without other program components) in a range of potential contexts (e.g., site-based, community-based, e-mentoring) for children and adolescents in any type of court-ordered out-of-home care (i.e., non-relative foster care, kinship foster care, or congregate care) due to maltreatment. The review includes studies of transition-age youth (16 to 25 years old) as long as the studies included some participants under the age of 18.

    For this review, mentoring is defined by the National Mentoring Resource Center as “relationships and activities that take place between youth (i.e., mentees) and older or more experienced persons (i.e., mentors) who are acting in a nonprofessional helping capacity, whether through a program or more informally, to provide support that benefits one or more areas of the young person’s development.” (For further details, see What is Mentoring?) This definition typically excludes services and supports that are offered in formal professional roles by those with advanced education or training (e.g., social work, counseling). However, for the purposes of this review, these requirements were relaxed to include studies of programs in which mentors were required to hold professional degrees.

    A literature search was conducted to identify journal articles, book chapters, and other types of reports pertinent to one or more of the central questions for this review, including searches of PubMed, ProQuest Dissertations & Theses, PsycINFO, and Google Scholar, using an established set of keywords. This search identified a total of 30 articles/reports that met criteria for inclusion in this review.

  • What Is the Effectiveness of Mentoring for Youth in Foster Care?


      There are many reasons to believe that mentoring may be a contextually-sensitive and efficacious intervention for youth in foster care. Many young people who have been in foster care for a long time are resistant to more professional help, but they are open to a mentoring relationship. From a pragmatic perspective, mentoring does not require a stable caregiver to participate (as many youth interventions do), which enables the intervention to continue even after a child changes placements. Mentoring also may provide the young person opportunities to engage in activities that are restricted or logistically difficult to engage in due to issues with transportation, time, or financial resources. As most mentoring relationships are individualized, mentoring may work well as a strategy to intervene with youth in foster care who have heterogeneous challenges, including cognitive and learning disabilities as well as mental and physical health problems. Young people in care often have gaping holes in the developmental assets needed for attaining success in adulthood and seek support from individuals who are older, successful, accessible, trustworthy, provide emotional and instrumental support, have authority, and demonstrate guidance and understanding.22, 23, 24 Finally, when so much of the focus in social services is on ameliorating family problems, mentoring can focus on the child—fostering positive youth development and nurturing his or her interests and talents.

      Although mentoring holds great promise for youth in foster care, there are also some cautions. Mentor abandonment may be more detrimental for youth in foster care than for non-foster youth, and mentors may face greater challenges, such as encountering resistance, overly rigid or blurred boundaries, mixed messages regarding youth satisfaction, or significant psychosocial needs, making their emotional connection with their mentee difficult. Youth who have experienced loss are also at particular risk for premature relationship endings, given their lack of stable living arrangements and high rates of emotional and behavioral problems. Mentors may interpret a lack of responsiveness or openness from their mentees as disinterest in the relationship without understanding the impact of their attachment history or the complicated lives that they lead. Indeed, young people in care discuss the importance of consistency and emotional closeness in mentoring relationships, which may require more flexibility, persistence, and patience on the part of the mentors.25

      While program-based mentors can certainly develop positive relationships with youth in foster care, naturally forming mentoring relationships might be particularly impactful for this population, as these relationships typically involve mutual trust and a shared understanding of the youth’s difficult background and associated emotional and behavioral problems. Because these relationships form naturally over time, there is less likelihood that they will terminate abruptly and a greater likelihood that they will last for many years, helping bridge important transitions for foster youth, especially the transition to independence.26


      Natural Mentoring.

      Young adults who emancipate from care report high rates (more than 70 percent in most studies) of nonparental natural mentorship, with relationships most likely to have begun before or during adolescence.5, 27, 28, 29 Although most of these studies involved small samples of youth that were not necessarily representative of the larger foster care population, there were many consistencies in their findings. Nonparental natural mentors most often consist of relatives, friends of the family, caseworkers, former foster parents or staff at their former placements, therapists, and teachers.5, 27, 28, 29, 30, 31, 32 Young adults describe their relationship with their natural mentor as very close, using phrases such as “like a parent,” or “trusted advisor”; in one study, over half of the respondents were in contact with their mentor nearly every day.27, 32 The types of support that natural mentors are reported to provide include instrumental, informational, and emotional support, teaching social skills, providing advice, and “keeping them on track,” which many young people described as key for preventing negative outcomes and, most importantly, supporting them in achieving positive ones.5, 27, 28, 29, 30, 31, 32

      Studies have also empirically examined whether the presence of a natural mentor is associated with better functioning. In two studies of emancipating/emancipated youth, those with a natural mentor experienced more favorable outcomes than those without a mentor: they had lower levels of stress and higher life satisfaction, were more likely to complete high school or obtain a GED, and were less likely to be arrested or experience homelessness as a young adult.29, 33 The presence of a natural mentor was, however, unrelated to employment or substance use.33 Two other studies used data from the National Longitudinal Study of Adolescent Health (Add Health) to examine the effects of a natural mentor. The Add Health study is a six-year, multiwave longitudinal study that enrolled several nationally representative cohorts of youth in grades 7 to 12. At Wave 3, when participants were between the ages of 18 and 26 (average age of 21.5), they were asked to retrospectively report whether a nonparental adult had made an “important positive difference in your life since you were 14 years old.” Among study participants who reported they had lived in foster care, having a natural mentor was associated with several positive outcomes, including participation in higher education, less suicidal ideation, lower rates of sexually transmitted infections, less physical fighting, better perceived general health, and a higher number of positive outcomes across domains.30 Natural mentorship was not, however, associated with self-esteem, depression, completion of high school, current employment, assets (including having a bank account or owning a car or home), physical activity, body mass index, substance use, arrests, or gang membership.30, 34

      Although almost all studies of natural mentorship for youth in foster care examined the effects on adolescents or young adults, one study looked at whether the presence of a natural mentor in preadolescence was associated with better psychosocial functioning. This study found that children with natural mentors reported greater attachment to friends; however, having a natural mentor was unrelated to attachment to parents (either biological or substitute), social skills, or perceived future opportunities.31

      Formal Mentoring.

      With the growth of mentoring programs nationally, more studies are beginning to examine the impact of program-based mentoring for extremely vulnerable populations. Most studies examining the impact of formal mentoring programs for youth in foster care have been published over the past decade, and have focused on a wide range of outcomes, including social skills, relationship quality, life skills, self-determination, self-confidence, academic functioning, educational outcomes, mental health functioning, delinquency, placement stability, and employment. Similar to the focus in existing studies of natural mentoring, several programs (and their evaluations) focus on transition-age youth, and these emergent adults identify the same benefits from mentoring as do youth in natural mentoring relationships.35 Interestingly, almost every evaluated formal mentoring program included in this review used paid mentors or mentors who were in college or graduate school. Several of the program evaluations used randomized controlled designs, which enables the field to make fairly strong conclusions about program efficacy.

      Big Brothers Big Sisters Studies.

      Two studies examined the impact of Big Brothers Big Sisters (BBBS) programs on youth in foster care. The first analyzed data previously collected in a randomized controlled trial (RCT) of a national study of BBBS involved youth ages 10 to 15. This study examined the 90 youth participants who identified themselves as living in foster care and compared those randomized to the BBBS intervention to those assigned to the control group. Relative to youth in the control group, foster youth in the BBBS group demonstrated larger improvements in peer support 18 months after the start of the study.36 A more recent nonexperimental pilot study examined the perceived impact of a BBBS program in North Texas involving 45 youth ages 5 to 16 in foster care. Youth were surveyed 4 and 10 months post referral to the BBBS program and reported high rates of program satisfaction and retrospective improvements in five areas: school, employment, friendships, relationships with authority figures, and self-confidence.37 It should be noted that this study encountered significant enrollment, retention, and fidelity problems, rendering the data collected difficult to interpret.

      Mentoring Programs to Improve Educational Outcomes.

      Several studies have examined the outcomes of mentoring programs specifically focused on improving education-related outcomes for youth in foster care. In the first study, a child welfare agency partnered with 18 school districts to provide individual tutoring/mentoring during the academic year to maltreated youth in foster care in grades K-12. The tutors/mentors were certified teachers, but not the child’s teacher. Most of the mentoring/tutoring activities were focused on academics and study skills. Although there was no comparison group in this study and fairly small subsets of the 206 participants completed assessments at later time points, results suggested that students had improved their academic skills.38

      A second educationally focused mentoring program, Take Charge, was tested in two RCT studies of older youth who were living in foster care and receiving special education services. The first study enrolled 69 youth ages 16 to 17 and the second study enrolled 133 youth ages 14 to 17. The youth in the intervention groups received individualized coaching by trained staff as well as group mentoring by “near-peer foster care alumni” over the academic year. Across the two studies, positive outcomes were noted in self-determination and mental health (as rated by youth and parents, but not by teachers), independent living activities, use of transition services, self-identified educational goals and accomplishments, educational planning knowledge and engagement, postsecondary preparation, and quality of life. There were no significant differences, however, between the intervention and control groups in GPA or school attitude.39, 40

      An adapted Take Charge model called Better Futures was tested in a similar population, namely young people ages 16 to 18 in foster care with serious mental health challenges. The intervention included individualized coaching for youth around key self-determination skills, an on-campus summer institute, and mentoring workshops with near peers who shared the experience of living in foster care and having mental health challenges. An RCT of the Better Futures intervention with 67 youth found that youth in the intervention group, compared to those in the control group, evidenced higher rates of enrollment in postsecondary schooling, as well as more positive reports on scales measuring self-determination, mental health empowerment, postsecondary preparation, transition planning, and hopefulness.41

      Finally, two small programs implemented through universities’ schools of social work provided mentoring to transitioning foster youth with the goal of increasing their awareness of, and access to, higher education. Although the small-scale studies of these programs did not include comparison groups, participating youth reported increases in knowledge about postsecondary options and improvements in academic self-efficacy.42, 43

      Multicomponent Programs to Improve Mental Health and Behavioral Outcomes.

      Two RCT studies evaluated multicomponent programs, each with a manualized skills group and mentoring/coaching component that focused on improving mental health and behavioral functioning of children in foster care. The first study, Fostering Healthy Futures (FHF), enrolled 156 nine- to eleven-year-old children recently placed in foster care. The children attended 30 weeks of social skills groups and received individualized mentoring over the academic year. Mentors were graduate students in social work and psychology. The study demonstrated improvements in mental health problems (including trauma symptoms) and quality of life, as well as a reduction in mental health services utilized, number of placement changes, and placement in residential treatment centers in the group assigned to receive mentoring services relative to the control group. There were no demonstrated effects on coping skills, social support, or social acceptance.44, 45 An adaptation of FHF for teens is currently being tested in another RCT. A preliminary pilot study of this adapted program, Fostering Healthy Futures for Teens, which serves older youth with a history of foster care using mentoring without skills groups, demonstrated high rates of program engagement and satisfaction.46

      Another intervention, Middle School Success, for girls in foster care was similar to FHF in that it was implemented over an academic year, although group sessions for girls and their caregivers occurred for only three weeks (twice a week) prior to the start of the school year. Parents continued weekly group meetings and girls received individualized coaching by trained staff over the academic year. Short-term results of the RCT with 100 youth included improvements in internalizing and externalizing behaviors in the girls assigned to the intervention relative to controls. At the two-year post-intervention follow-up, girls who had been enrolled in the intervention had fewer placement changes, demonstrated more prosocial behaviors, and were less like to be using substances and engaging in delinquent behaviors than girls assigned to the control group.47,48

      System-Run Mentoring/Advocacy Programs.

      Some mentoring programs for youth in foster care were administered by social services, in which trained staff provided the mentoring/coaching. One study examined the impact of “therapeutic mentoring” for 262 children ages 6 to 15 who were at risk of placement disruption from their foster homes. Caseworker reports were used to evaluate the impact of the program using a quasi-experimental design in which clinicians referred some youth to therapeutic mentoring if they determined the child was “able to benefit” from a mentoring relationship. The youth who were not referred to mentoring served as the comparison group. Mentored youth were reported by caseworkers to have fewer trauma symptoms than youth in the comparison group. Unfortunately, the small number (n = 27) of participants at follow-up limits our ability to understand the impact of the program.49, 50

      In the Massachusetts Adolescent Outreach Program, outreach workers (who were required to be licensed social workers) met individually with older, youth (ages 15 to 20) who were transitioning from foster care to help them develop skills and engage in appropriate developmental tasks including applying for jobs and college. An RCT of the program with 194 youth found positive impacts on college enrollment and retention, obtaining important documents (e.g., birth certificate, license) and receiving assistance with education, employment, housing, and financial management. There were no differences, however, between the control and intervention groups on several other target indices, including employment, economic well-being, stable housing, delinquency, pregnancy, or preparedness for independence.51, 52

      An innovative British study reviewed case files of 181 participants across 11 mentoring programs for emancipating youth ages 15 to 23. Many of the mentoring programs were situated within departments of social services as the program was mandated by the Prince’s Trust Leaving Care Initiative. The case reviews found that over three-quarters of the young people achieved program goals, but that half the relationships evidenced “negative outcomes,” including not meeting the goals and/or having unplanned endings to the mentoring relationship.53

      Finally, youth in foster care are often paired with Court-Appointed Special Advocates (CASAs). CASA volunteers are laypersons who receive specialized training to assist children through the court proceedings. They are typically assigned only a small number of children and can provide more time and attention to youth than attorneys or caseworkers.54 One key difference between CASA volunteers and typical mentors is that CASA volunteers are charged with providing information to court officials to inform decision-making. CASA volunteers are focused on the well-being of the child and may be one of the only sources of stability, as social workers, attorneys, and/or foster parents change.55 Evidence for the effectiveness of CASA volunteers is based mostly on quasi-experimental studies and is limited almost exclusively to outcomes related to permanency. A full review of CASA models is outside the scope of this review, but existing studies tentatively suggest that youth with a CASA volunteer may experience fewer placement changes, less time in care, greater likelihood of achieving permanency, greater access to services and resources, and more support through adult relationships.56,57 It is important to note, however, that not all studies have replicated these findings,58 and few differences between those with a CASA volunteer and those without have been reported on measures of well-being and development.56


      1. Although not yet confirmed through rigorous evaluation, mentees report that both natural and program mentors provide life-changing informational, instrumental, and emotional support to young people in foster care who often lack consistent adults to help them navigate the many challenges they face.

      2. Available research suggests that both natural and program-based mentoring for children in foster care (across a range of ages and mentoring formats) can have positive impacts on mental health, educational functioning and attainment, peer relationships, placement outcomes, and life satisfaction; mentoring demonstrated no impact or mixed results for other outcomes, including social skills, attachment to adults, physical health, employment and financial assets, risky behaviors, and associated negative life-course outcomes (e.g., substance use, delinquency, arrests).

      3. Most of the mentoring programs serving youth in foster care that have been evaluated to date have been multicomponent (i.e., they included components other than one-to-one mentoring, such as skills groups) and utilized mentors who were agency staff members or university students; thus, we know less about other program models serving this population.

  • What Factors Influence the Effectiveness of Mentoring for Youth in Foster Care?


      Conceptually, characteristics of the mentor/mentee and/or characteristics of the mentoring program or practice (or the combination of these factors) may influence the impact of mentoring on young people in foster care. For example, ethnic minority youth (an individual mentee characteristic) may benefit more from a mentoring program than White youth, but only if they are paired with a same-race mentor (a program practice). It is possible to theorize why different genders, racial/ethnic groups, level of adverse childhood experiences, and type of placement (kinship versus foster care) might lead to better outcomes, yet most studies that have examined the characteristics of mentors/mentees and their association with outcomes have not had strong, theoretically based, a priori hypotheses, which poses a challenge to applying these findings to mentoring youth in foster care. Previous literature provides a bit more guidance with regard to program practices. Certain program practices (e.g., mentor training, case management) are hypothesized to produce better outcomes for youth in foster care because they are associated with longer, higher-quality matches in studies focused on other youth populations.59 Mentoring uptake (e.g., number of mentoring visits, length of match) has similarly shown potential in previous studies to influence mentoring outcomes.60 Below we separately review the impact of individual and mentoring-specific factors that may condition the effectiveness of mentoring for youth in foster care.


      Youth characteristics.

      Few studies have examined whether children with certain demographic characteristics benefit more from mentoring. In terms of age, none of the reviewed studies examined whether youth at certain ages benefitted more, although one study found no differences in mentoring’s impact as a function of pubertal development in middle-school-age girls.48 Most of the studies enrolled a narrow age range of participants, which limits the ability to test for age effects. No reviewed studies tested (statistically) whether boys or girls benefitted more from mentoring. Race/ethnicity did not show a consistent pattern of influencing the impact of mentoring in two studies,38, 52 but youth in foster care are a heterogeneous group and both of these studies grouped all non-White youth into a “Minority” category.

      Some studies have examined whether maltreatment or placement history, as well as youths’ risk level prior to their receipt of mentoring, contributed to the impact of the intervention. Two studies did not find differences in program efficacy as a function of maltreatment severity.48, 61 A study of social support (not specifically mentoring), however, found that the impact of social support on depressive symptoms was greater for youth who experienced fewer types of maltreatment,52 leading the authors to concluded that “. . . as cumulative risk grows, the adequacy of compensating factors may diminish, and the chances of negative consequences may increase. Complex trauma may represent a circumstance when, even with social support, coping is taxed beyond the limit . . .” (p. 110).62 In support of this hypothesis, a recent study of Fostering Healthy Futures found that as the number of adverse childhood experiences (ACEs) increased, the impact of the intervention on trauma-related symptomatology diminished.63 At the same time, being placed in foster care is also seen as an ACE, and two studies of youth in foster care found that they achieved more benefits from mentoring than did youth not placed in foster care.36, 38 The type of foster care may also matter. One study found that the youth placed in kinship (relative) care benefitted more from mentoring than youth in non-relative foster care.36 Finally, study results are unclear in determining whether pre-program rates of behavior problems are associated with the strength of program outcomes. One study found no differential impact by behavior problems on psychosocial and behavioral outcomes,48 while another study found that the reduction in number of placement changes was greatest for those youth who had more behavior problems prior to the intervention.61

      Program practices and mentoring relationship characteristics.

      Although no reviewed studies examined the effects of program practices (e.g., amount of training or case management) on outcomes, several examined whether mentoring frequency/duration, quality of the mentoring relationships, or the characteristics or role of the mentor were associated with natural and programmatic mentoring outcomes. Because none of these factors was systematically varied in the studies, causal associations cannot be inferred, and the findings must be viewed as only suggestive.

      Two studies of mentoring frequency (i.e., how often matches met) and duration (i.e., the length of the mentoring relationship) found no differences in psychosocial or behavioral outcomes based on these factors.30, 33 Another study found that dosage (i.e., the number of hours of mentoring received) was related to better outcomes, and that those youth who received fewer hours of mentoring did worse on some measures than youth who received no mentoring.50 Similarly, in a systematic review of studies of natural mentoring among older youth in foster care, longevity and consistency were noted to be important factors.64

      Others have hypothesized that relationship quality and the role or characteristics of the mentor are important in determining outcomes. Although several qualitative studies (those based on interviews with participants) have identified key characteristics of the mentoring relationship or mentor as important to achieving good outcomes (e.g., acceptance, encouragement, reliability, patience, responsiveness, openness, understanding),28, 29 one prospective study found that quality of mentoring (as rated on an 11-point scale by the mentees) was not associated with strength of outcomes.33 While having a mentor who was “like a parent” or “serves as a role model” was associated with some favorable outcomes in one study of natural mentoring,34 another study found poorer outcomes for youth with natural mentors who engaged in health-compromising behaviors, such as substance use.65


      1. The existing literature suggests that the effects of mentoring may differ by children’s demographic characteristics, but the literature is insufficient to provide firm conclusions about their effects.

      2. Youth who are at very high levels of risk may not benefit as much as youth at lower levels of risk, and risk may also differentially affect distinct outcomes (e.g., psychosocial outcomes vs. child welfare outcomes).

      3. Characteristics of the mentoring relationship, including frequency of meetings, duration, and quality of the match, are inconsistently related to mentoring outcomes, although few of the rigorously designed studies examined these characteristics and no studies systematically varied these indices to test their importance.

      4. Studies of natural mentoring suggest that mentor characteristics and the role mentors play may be linked with youth’s receipt of benefits.

  • What Pathways Are Most Important in Linking Mentoring to Outcomes for Youth in Foster Care?


      Youth who have experienced trauma, compromised attachment, and weak social scaffolding are in need of reparative relational experiences that allow for the development of positive youth outcomes. Although no known research has empirically examined whether mentoring is a corrective attachment experience, theory suggests that the improvement of working models may be a key mechanism in mentoring’s effects. Multiple placement changes, fluctuating living and school situations, and uncertainty about one’s future all affect youth’s sense of constancy and hope. For youth in foster care, increased stability, future orientation, and self-advocacy processes may be needed to translate mentoring into intended outcomes. Programs that foster these outcomes may be most effective in yielding positive youth development among those in foster care.

      Additionally, as is true for most populations, youth in foster care are not a homogenous group, and the programs serving these youth are diverse in their goals. Therefore, the processes critical to realizing specific outcomes are dependent on the program aims, theory of change, and needs of youth. For example, a formal program designed to increase foster youth’s academic achievement may focus on the development of academic engagement as a key process, whereas a program designed to address mental health may focus on emotion regulation or coping.


      Four studies were located that examined processes through which mentoring relationships are linked to positive youth outcomes among youth in foster care. The first was a study of natural mentoring among former foster youth in which asset accumulation (i.e., owning a car, owning a residence, and having a bank account) was the main outcome of interest. It was hypothesized that having a natural mentor during adolescence would improve foster youth’s future expectations, which would then increase the likelihood of accumulating those assets during emerging adulthood. Results, however, indicated that this was not the case; having a natural mentor was not associated with the accumulation of any of the assets examined for former foster youth.34

      The second study investigated whether improvements in self-determination as a result of participation in the Take Charge program was related to greater quality of life. Self-determination was defined as having the power to make decisions, to direct one's actions, to dream and take risks, and to exercise rights and responsibilities. As hypothesized, mentoring in this educationally focused program improved youth’s belief in their ability to take action on their own behalf, which was, in turn, associated with greater quality of life, including connections with others, social inclusion, individual control, community integration, productivity, and overall satisfaction and well-being.40

      The third study assessed the Massachusetts Adolescent Outreach Program and found that in addition to the positive effects on college enrollment and persistence, youth in the outreach program were also significantly more likely to remain in foster care. The authors conducted additional analyses to examine whether the effects on college outcomes were a direct result of the outreach program or whether they operated through the program’s impact on remaining in care. Results indicated that, in fact, more than 90 percent of the program’s effect on outcomes was accounted for by program youth remaining in care longer than control youth, suggesting that there may be an educational benefit of remaining in care.51

      Finally, the fourth study examined whether decreases in substance use and delinquency were accounted for by changes in prosocial skills in the Middle School Success program. The authors’ hypothesis was partially supported; that is, for delinquency, the intervention achieved positive effects mainly through increased prosocial skills. However, the intervention also had a direct effect on substance use that was not explained by increased prosocial skills.47


      1. Mechanisms, or processes, through which mentoring may affect outcomes include improving future expectations and self-determination and increasing time in care, but research is extremely limited and thus inconclusive.

      2. One well-designed study found that improvement in prosocial skills was critical to avoiding some delinquent behaviors, but more research is needed to generalize these findings to other programs and outcomes of interest.

  • To What Extent Have Mentoring Initiatives for Youth in Foster Care Reached and Engaged These Youth, Been Implemented with High Quality, and Been Adopted and Sustained?


      Before mentoring programs can be successfully and widely disseminated to reach more youth in foster care, it is important to understand whether the findings of the studies examined in this review can be generalized to the broader population of youth in foster care. Recruiting representative samples of youth in foster care for research studies as well as program participation is complicated for a number of reasons, including:

      • There are unique human protection, consent, and confidentiality procedures for wards of the state;

      • Determining eligibility is difficult when caseworkers and foster parents have limited knowledge of the child’s background and functioning and/or there are missing health and/or educational records;

      • Coordination of involved systems (e.g., child welfare, juvenile justice, mental health) to work collaboratively on the child’s behalf can be difficult; and

      • There are scheduling and transportation challenges for participation in both the research and mentoring, as children’s living situations (and therefore caregiver availability) change over time.66, 67

      Despite these challenges, programs that have demonstrated positive outcomes should be carefully considered for dissemination and implementation among the populations for which they have demonstrated efficacy, as there are far too few evidence-based programs for this vulnerable population and we know that youth in foster care participate in far fewer organized activities than youth not in care.68

      As we think about the expansion of mentoring for youth in foster care, however, there are a number of cautions. Two thoughtful articles articulate serious concerns about widespread expansion of mentoring for youth in foster care without careful consideration of contextual and programmatic features. They caution that mentoring should not be seen by social services as a substitute for parental support, especially as young people emancipate from care, nor should mentors be expected to independently navigate key instrumental supports such as housing, education, or health care for their mentees.25,69


      Reach and Engagement.

      In the outcome studies reviewed in these pages, there was great variability in the programs’ recruitment and retention rates. A recent paper focused on barriers to implementing mentoring programs for youth in foster care provides great detail on the implementation of a BBBS program for youth in foster care that was mandated by the Texas legislature. More than 200 youth were referred to the program, but only 46 were matched with a mentor and only 3 participated for a full year. Sixty percent of the youth had either very little contact with their mentor or dropped out of the program.37 Low rates of recruitment, engagement, and/or retention were noted by other studies as well,35, 43, 48, 50, 53 and no studies examined the impact of “failed matches” on outcomes for youth.

      Low rates of recruitment and samples of convenience in the available studies of these programs make it difficult to know whether the studies’ findings can be generalized to the broader foster care population. For example, the Massachusetts Adolescent Outreach Program served only youth who were in “intensive foster care and who had a case plan goal of independent living or long-term substitute care” (p. vi)51 and the authors report that the findings should not be generalized to other populations of youth in foster care. Although the program-based mentoring studies reviewed above varied in their recruitment strategies, most of the studies of natural mentoring recruited convenience samples, which limits the generalizability of their findings.

      In terms of mentoring’s reach, it is also important to understand who engages in mentoring. No reviewed studies of formal mentoring programs have examined the characteristics of youth who engage in these programs, but two of the more rigorous studies of natural mentoring examined characteristics of foster youth who were most likely to have a natural mentor. Youth in foster care who had a natural mentor did not differ from those without a mentor in terms of gender, race/ethnicity, socioeconomic status, maltreatment type, number of placements, or mental health and behavioral problems.30, 31 In one of these studies, however, those youth with a natural mentor were older, had been in foster care for a shorter period of time, and were more likely to live in a residential treatment facility than children without a natural mentor.31

      Quality of Implementation.

      Unfortunately, no reviewed studies examined whether the quality of the program (e.g., training of mentors, monitoring of matches) was important in achieving outcomes, although in one study, mentors reported difficulty securing necessary resources and services for their mentees, even though resource acquisition was a key goal of the program. The mentors reported wanting more support and supervision around these goals.35 A few studies have delineated their training and supervision of mentors and/or surveyed the mentors for their perceptions of the mentoring experience,35, 37, 49, 70, 71 but these descriptions tend to be very program specific and no conclusions can be drawn about the impact of program practices on outcomes.

      Specificity of program goals is also an important factor to consider. For example, a study that examined whether providing mental health and educational assessments and recommendations led to an increase of services for children in foster care, found that the presence of a mentor did not impact service receipt post-program.72 This was likely because the goal of mentoring in that program was to improve social skills and reduce mental health problems, not advocate for services. Similarly, the goals of the program may not be well aligned with the mentees’ goals for themselves. In one study, coaches reported that their mentees were resistant to identifying proximal educational goals (the focus of their program) and wanted instead to focus on life after high school (e.g., getting a job).39

      Adoption and Sustainability.

      A recent review of mental health interventions for children in foster care found that there were many fewer evidence-based interventions that were adapted for youth in foster care than there were newly designed programs.73 This is likely due to the challenges of adapting existing evidence-based programs for children in foster care, which is reflected in the current review—other than the two BBBS studies, all of the programs reviewed were specifically designed for children with child welfare involvement. While mentoring programs for youth in foster care face some of the same barriers to implementation as mentoring programs for other populations (e.g., finding sustainable funding sources), there are other barriers that make implementation of programming with this population particularly difficult. No studies were found which examined the implementation of existing evidence-based mentoring practices for youth in foster care although two studies adapted their own practices for slightly different populations.39, 40, 45

      Although there are no published studies of its implementation, the Fostering Healthy Futures (FHF) program has been implemented through two community-based agencies over the past four years with high rates of fidelity to the model and better youth attendance rates than were found in the earlier studies of its impacts.44, 45 More widespread implementation of FHF, however, is limited due to: (1) the narrow population for which it has demonstrated efficacy (e.g., 9- to 11-year-old children recently placed in foster care); (2) the available funding sources, as the child welfare system typically pays for services only if cases are open and children stay in the FHF program even if they reunify or their cases close; and (3) the use of graduate student mentors, which limits the reach beyond urban areas. Barriers to the implementation of other programs reviewed here would include the high cost of paid mentors.


      1. Both natural and program-based mentoring appear to appeal to and engage youth who are diverse in sociodemographic and behavioral/emotional functioning, although mentoring programs (especially those with less structure) often have difficulty retaining foster youth.

      2. Studies have not examined whether adherence to a given program model predicts better outcomes for youth, although alignment of program goals and outcomes is reported to be important by program developers and participants.

      3. Although there are many conceptual reasons why mentoring is an excellent fit for youth in foster care, there are pragmatic challenges, both logistical and financial, that make widespread implementation difficult and no studies have been conducted that examine program expansion or adaptation.

      4. Because of the high potential for adverse outcomes among this vulnerable population, great care and coordination is required for implementing mentoring programs and supporting natural mentoring relationships for youth in foster care; if done well, however, the benefits of mentoring appear to outweigh the risks and foster youth may experience positive outcomes across a range of domains.

  • Implications for Practice

    (Mike Garringer, MENTOR: The National Mentoring Partnership)

    As noted in the preceding review of evidence for serving youth in the foster care system with mentoring relationships and targeted mentoring services, there is plenty of research that suggests just how impactful these relationships can be for youth who have experienced the trauma, pain, and uncertainty of being abused and neglected and becoming a ward of the state. But, this review also notes several cautions against applying these research findings too broadly and warns policymakers and other stakeholders against delivering mentoring relationships to these youth without fully considering the nuances and pitfalls. As emphasized in the review, “mentoring should not be seen by social services as a substitute for parental support, especially as young people emancipate from care, nor should mentors be expected to independently navigate key instrumental supports, such as housing, education, or health care for their mentees.”

    Thus, the first bit of advice for funders or practitioners who want to alleviate some of the challenges faced by these youth with mentoring is to proceed with caution. Serving these youth involves additional planning and safeguards not often considered in traditional mentoring programs. As noted many times in the review, these youth have often been through complex, adult-instigated trauma and may be dealing with cumulative risks that are beyond what many children who participate in traditional youth mentoring programs face. The last thing any mentoring program wants to do is somehow make the circumstances foster youth face worse by underestimating the challenges and risks.

    That said, the basic focus on building a healthy relationship that typically defines mentoring is shown here to be of central importance when mentoring youth in foster care, so foundational principles of mentoring remain useful and applicable. This review further suggests a number of key insights that practitioners can use to build and implement responsive and meaningful mentoring interventions for youth who are in, or exiting out of, the foster care system. Relative to other evidence reviews conducted by the National Mentoring Resource Center, there is an emerging yet solid amount of research on which programs can build effective services and avoid doing harm.


      One of the main themes noted in this review was how often programs designed for this population used highly qualified staff or highly trained graduate students or undergrads in the mentoring role, as well as how common it is to offer mentoring alongside other more clinical forms of support. In fact, only two of the studies included in this review offered mentoring in what might be considered “traditional mentoring program” settings (in these cases, Big Brothers Big Sisters agencies).

      Given the struggles several programs faced when working with these youth, it seems critical that any program interested in serving youth in foster care should clearly define its target population as well as its recruitment, engagement, and retention strategies before enrolling children. Coordinating efforts with foster care and other agencies to identify eligible youth and share resources necessary to recruit, launch, and provide high-quality programming is also needed.39, 37, 66

      Because these youth are engaged with complicated and traumatizing systems and because they have already experienced serious problems with adults they have trusted in the past, this seems to be a population of youth that may benefit most from mentoring services that are specifically tailored to their needs and integrated with the other supports they are receiving to the degree possible. Programs may want to work with clinicians, social workers, or others that have deep experience in meeting the needs of foster youth to design and develop match activities and teachings that augment relational mentoring experiences with skills training, therapeutic practices, or other clinically-informed interventions that can more holistically support the needs of foster youth.

      Most studies of mentoring programs to date have used college or paid mentors. Programs may have targeted these individuals to help ensure that they have the knowledge or skills needed to work with this population or because programs can more easily ensure that these mentor groups have strong incentives to meet program requirements. However, this focus should not imply that other mentor groups are less effective; we simply know less about other types of mentors and the supports they might need to be effective.


      The review notes that one of the main ways that mentors, both in and out of programs, can benefit foster youth is by serving as a corrective experience in the wake of dysfunctional and harmful relationships. Mentors can help rebuild trust and transfer interpersonal skills to youth who have unfortunately often been taught to keep adults and others at bay as a result of their painful experiences. All mentors need training, but these mentors, in particular, must be trained to:

      • Recognize the importance of consistency, not only in terms of meeting frequency and stability, but also in how they carry themselves and the way they demonstrate dependability and build trust with the mentee. It is especially critical that they remain consistent across placements, when possible, as those are the most disruptive times for these mentees.

      • Offer genuine caring, warmth, and opportunities for fun⎯foster youth have many other adults invasively in their lives due to their status, but a mentor can be an oasis of fun, unconditional support, and relaxation.

      • Understand that foster youth may struggle to form relationships because of their past experiences and that demonstrating “disinterest” is often just a coping mechanism. Too many mentoring relationships end early because the mentor feels like the youth is not “into” the relationship, when in fact that feigned disinterest is a put-on born of self-preservation from prior harmful experiences. Patience and perseverance are musts when mentoring youth in foster care.

      This review notes that natural mentoring relationships⎯those formed outside of an intentional program context⎯have plenty of evidence that they provide valuable support to foster youth. Perhaps one of the greatest gifts a programmatic mentor can give a youth in foster care is to broaden that child’s network of support⎯the individuals that scaffold that young person and nurture their development in different ways. Approaches to youth-initiated mentoring, in which the youth is guided and empowered to identify and gather support from a range of other adults, is a model that shows considerable promise. Of course, mentors who work with a mentee in this way will need to be trained in how to help the youth identify appropriate adults to take on additional mentoring roles given the findings noted in this review on the potential negative effects of mentors who engaged in health-compromising behaviors. Identifying additional positive role models may be a challenge for some foster youth, but the payoffs of helping a young person build more and deeper connections may be worth the risks.

      It is also worth noting that these youth have often had healthy relationships with adults in their “circle” taken away from them as a result of their entry into the child welfare system. So, in addition to helping youth find new adults to rebuild that web of support, programs should also see if there are adults who were part of the child’s life for whom they can work with the foster care system to help reconnect and rebuild supportive relationships for the youth they serve. Programs should not assume that youth in foster care need an entirely new group of caring adults and should train their mentors accordingly. Sometimes restoring the relationships lost is a good starting point.


      One of the stronger programs in this review, Take Charge, utilized a set of mentoring activities designed to promote youth-empowerment, self-determination, and goal setting around educational and post-emancipation dreams. This may be an especially important approach as young people reach the end of their time in the foster care system by strengthening their goal-attainment skills and empowering them to exercise agency over the next steps in their lives. This program’s approach resulted in reports of greater quality of life for mentored youth, including connections with others, social inclusion, individual control, community integration, productivity, and overall satisfaction and well-being.

      As noted toward the end of the review, programs serving foster youth face challenges when the goals of the program are not aligned with the goals and needs of the youth being served. A self-determination approach alleviates this challenge by putting these youth in charge in a way that is unfortunately uncommon given their circumstances.

  • References

    1. Child Welfare Information Gateway. (2017). Foster care statistics 2015. Washington, DC: U.S. Department of Health and Human Services, Children’s Bureau. Retrieved from

    2. Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., … Marks, J. S. (1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) study. Am Journal of Preventive Medicine, 14, 245–58.

    3. Weiler, L. M., Garrido, E. F., & Taussig, H. N. (2016). Well-being of children in the foster care system. In M. R. Korin (Ed.), Health promotion for children and adolescents (pp. 371–388). New York, NY: Springer.

    4. Flynn, R. J., Ghazal, H., Legault, L., Vandermeulen, G., & Petrick, S. (2004). Use of population methods and norms to identify resilient outcomes in young people in care: An exploratory study. Child & Family Social Work, 9, 65–79.

    5. Hass, M., & Graydon, M. (2008). Sources of resiliency among foster youth. Children and Youth Services Review, 31, 457–463.

    6. Simmel, C. (2007). Risk and protective factors contributing to the longitudinal psychosocial well-being of adopted foster children. Journal of Emotional and Behavioral Disorders, 15, 237–249.

    7. Luthar, S. S., Cicchetti, D., & Becker, B. (2000). The construct of resilience: a critical evaluation and guidelines for future work. Child Development, 71, 543–562.

    8. Egeland, B., Jacobvitz, D., & Sroufe, L. A. (1988). Breaking the cycle of abuse. Child Development, 59, 1080–1088.

    9. Masten, A. S., Cutuli, J. J., Herbers, J. E., & Reed, M.-G. J. (2009). Resilience in development. In C. R. Snyder & S. J. Lopez (Eds.), Oxford handbook of positive psychology (pp. 117–131). Oxford, England: Oxford University Press.

    10. Lynch, M., & Cicchetti, D. (1992). Maltreated children’s reports of relatedness to their teachers. New Directions for Child Development, 57, 81–107.

    11. Bowlby, J., (1988). A secure base: Parent-child attachment and healthy human development. New York, NY: Basic Books.

    12. Britner, P. A., Randall, K. G., & Ahrens, K. R. (2014). Youth in foster care. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp. 341–354). Thousand Oaks, CA: SAGE.

    13. Cassidy, J., & Shaver, P. R. (Eds.). (2016). Handbook of attachment: Theory, research, and clinical applications (3rd ed.). New York, NY: Guilford Press.

    14. Aquilino, W. S. (2006). Family relationships and support systems in emerging adulthood. In J. J. Arnett & J. L. Tanner (Eds.), Emerging adults in America: Coming of age in the 21st century (pp. 193–217). Washington, DC: American Psychological Association.

    15. Bronfenbrenner, U., & Morris, P. A. (2006). The bioecological model of human development. In W. Damon & R. M. Lerner (Eds.), The handbook of child psychology. Vol. 1: Theoretical models of human development (6th ed., pp. 793–828). Hoboken, NJ: Wiley.

    16. Scharf, M., Mayseless, O., & Kivenson-Baron, I. (2004). Adolescents’ attachment representations and developmental tasks in emerging adulthood. Developmental Psychology, 40, 430−444.

    17. Overton, W. F. (2010). Life-span development: Concepts and issues. In R. M. Lerner & W. F. Overton (Eds.), The handbook of life-span development. Vol 1: Cognition, biology, and methods (pp.1–29). Hoboken, NJ: Wiley.

    18. Samuels, G., & Pryce, J. (2008). “What doesn’t kill you makes you stronger”: Survivalist self-reliance as resilience and risk among young adults aging out of foster care. Children and Youth Services Review, 30, 1198–1210.

    19. Samuels, G. M. (2008). A reason, a season, or a lifetime: Relational permanence among young adults with foster care backgrounds. Chicago: Chapin Hall Center for Children at the University of Chicago. Retrieved from

    20. Collins, M. E. (2004). Enhancing services to youths leaving foster care: Analysis of recent legislation and its potential impact. Children and Youth Services Review, 26, 1051–1065.

    21. Avery, R. J. (2010). An examination of theory and promising practice for achieving permanency for teens before they age out of foster care. Children and Youth Services Review, 32, 399–408.

    22. Greeson, J. K.P., & Bowen, N. K. (2008). “She holds my hand”: The experiences of foster youth with their natural mentors. Children and Youth Services Review, 30, 1178–1188.

    23. Greeson, J. K.P., Thompson, A. E., Evans-Chase, M., & Ali, S. (2015). Child welfare professionals’ attitudes and beliefs about welfare-based natural mentoring for older youth in foster care. Journal of Social Service Research, 41, 93–112.

    24. Hudson, A. (2013). Career mentoring needs of youths in foster care: Voices for change. Journal of Child and Adolescent Psychiatric Nursing, 26, 131–137.

    25. Spencer, R., Collins, M. E., Ward, R., & Smashnaya, S. (2010). Mentoring for young people leaving foster care: Promise and potential pitfalls. Social Work, 55, 225–235.

    26. Greeson, J. K. P. (2013). Foster youth and the transition to adulthood: The theoretical and conceptual basis for natural mentoring. Emerging Adulthood, 1, 40–51.

    27. Duke, T., Farruggia, S. P., & Germo, G. R. (2017). “I don't know where I would be right now if it wasn't for them”: Emancipated foster care youth and their important nonparental adults. Children and Youth Services Review, 76, 65–73.

    28. Ahrens, K. R., DuBois, D. L., Garrison, M., Spencer, R., Richardson, L. P., & Lozano, P. (2011). Qualitative exploration of relationships with important nonparental adults in the lives of youth in foster care. Children and Youth Services Review 33, 1012–1023.

    29. Collins, M. E., Spencer, R., & Ward, R. (2010). Supporting youth in the transition from foster care: Formal and informal connections. Child Welfare, 89, 125–143.

    30. Ahrens, K. R., DuBois, D. L., Richardson, L. P., Fan, M., & Lozano, P. (2008). Youth in foster care with adult mentors during adolescence have improved adult outcomes. Pediatrics, 121, e246–e252. Retrieved from

    31. Greeson, J. K. P., Weiler, L. M., Thompson, A. E., & Taussig, H. N. (2016). A first look at natural mentoring among preadolescent foster children. Journal of Community Psychology, 44, 586–601.

    32. Munson, M. R., Smalling, S. E., Spencer, R., Scott Jr., L. D. S., & Tracy, E. M. (2010). A steady presence in the midst of change: Non-kin natural mentors in the lives of older youth exiting foster care. Children and Youth Services Review, 32, 527–535.

    33. Munson, M. R., & McMillen, J. C. (2009). Natural mentoring and psychosocial outcomes among older youth transitioning from foster care. Children and Youth Services Review, 31, 104–111.

    34. Greeson, J. K. P., Usher, L., & Grinstein-Weiss, M. (2010). One adult who is crazy about you: Can natural mentoring relationships increase assets among young adults with and without foster care experience? Children and Youth Services Review, 32, 565–577.

    35. Osterling, K. L., & Hines, A. M. (2006). Mentoring adolescent foster youth: Promoting resilience during developmental transitions. Child and Family Social Work, 11, 242–253.

    36. Rhodes, J. E., Haight, W. L., & Briggs, E. C. (1999). The influence of mentoring on the peer relationships of foster youth in relative and nonrelative care. Journal of Research on Adolescence, 9, 185–201.

    37. Scannapieco, M., & Painter, K. R. (2013). Barriers to Implementing a Mentoring Program for Youth in Foster Care: Implications for Practice and Policy Innovation. Child and Adolescent Social Work Journal, 31, 163–180.

    38. Mallett, C. A. (2012). The School Success Program: Improving maltreated children’s academic and school-related outcomes. Children & Schools, 34, 13–26.

    39. Geenen, S., Powers, L. E., Powers, J., Cunningham, M., McMahon, L., Nelson, M., … & Consortium to Increase the Success of Youth in Foster Care. (2013). Experimental study of a self-determination intervention for youth in foster care. Career Development and Transition for Exceptional Individuals, 36, 84–95.

    40. Powers, L. E., Geenen, S., Powers, J., Pommier-Satya, S., Turner, A., Dalton, L. D., … Swank, P. (2012). My Life: Effects of a longitudinal, randomized study of self-determination enhancement on the transition outcomes of youth in foster care and special education. Children and Youth Services Review, 34, 2179–2187.

    41. Geenen, S., Powers, L. E., Phillips, L. A., Nelson, M., McKenna, J., Winges-Yanez, N., ... & Swank, P. (2015). Better Futures: A randomized field test of a model for supporting young people in foster care with mental health challenges to participate in higher education. The Journal of Behavioral Health Services & Research, 42, 150–171.

    42. Bruster, B. E., & Coccoma, P. (2013). Mentoring for educational success: advancing foster care youth incorporating the core competencies. Journal of Human Behavior in the Social Environment, 23, 388–399.

    43. Kirk, R., & Day, A. (2011). Increasing college access for youth aging out of foster care: Evaluation of a summer camp program for foster youth transitioning from high school to college. Children and Youth Services Review, 33, 1173–1180.

    44. Taussig, H. N., & Culhane, S. E. (2010). Impact of a mentoring and skills group program on mental health outcomes for maltreated children in foster care. Archives of Pediatric and Adolescent Medicine, 164, 139–146.

    45. Taussig, H. N., Culhance, S. E., Garrido, E., & Knudtson, M. D. (2012). RCT of a mentoring and skills group program: Placement and permanency outcomes for foster youth. Pediatrics, 130, e33–e39.

    46. Taussig, H. N., Weiler, L. M., Rhodes, T., Hambrick, E., Wertheimer, R., Fireman, O., & Combs, M. (2015). Fostering healthy futures for teens: Adaptation of an evidence-based program. Journal of the Society for Social Work and Research, 6, 617–642.

    47. Kim, H. K., & Leve D. L. (2011). Substance use and delinquency among middle school girls in foster care: A three-year follow-up of a randomized controlled trial. Journal of Consulting and Clinical Psychology, 79, 740–50.

    48. Smith, D. K., Leve, L. D., & Chamberlain, P. (2011). Preventing internalizing and externalizing problems in girls in foster care as they enter middle school: Impact of an intervention. Prevention Science 12, 269–77.

    49. Johnson, B., & Pryce, J. M. (2013). Therapeutic mentoring: Reducing the impact of trauma for foster youth. Child Welfare, 92, 9–25.

    50. Johnson, B., Pryce, J. M., & Martinovich, Z. (2011). The role of therapeutic mentoring in enhancing outcomes for youth in foster care. Child Welfare, 90, 51–71.

    51. Courtney, M., Zinn, A., Johnson, H., & Malm, K. (2011). Evaluation of the Massachusetts Adolescent Outreach Program for Youths in Intensive Foster Care: Final Report. OPRE Report # 2011–14. Washington, DC: Office of Planning, Research, and Evaluation, Administration for Children and Families, U.S. Department of Health and Human Services. Retrieved from

    52. Greeson, J. K. P., Garcia, A. R., Kim, M., & Courtney, M. E. (2015). Foster youth and social support: The first RCT of Independent Living Services. Research on Social Work Practice, 25, 349–357.

    53. Clayden, J., & Stein, M. (2005). Mentoring young people leaving care: ‘Someone for me.’ York, United Kingdom: Joseph Rowntree Foundation. Retrieved from

    54. Litzenfelner, P. (2000). The effectiveness of CASAs in achieving positive outcomes for children. Child Welfare, 9, 179-193.

    55. Youngclarke, D., Dyer Ramos, K., & Granger-Merkle, L. (2004). A systematic review of the impact of court-appointed special advocates. Journal of the Center for Families, Children and the Courts, 5, 109–126. Retrieved from

    56. Caliber Associates (2004). Evaluation of CASA representation – Final report. Fairfax, VA: Author. Retrieved from

    57. Calkins, C. A, & Millar, M. (1999). The effectiveness of court-appointed special advocates to assist in permanency planning. Child and Adolescent Social Work Journal, 16, 37–45.

    58. Tuff, L. J. (2014). Court-Appointed Special Advocates: Is their impact effectively evaluated by current research methodology? (Master’s thesis.) Seattle, WA: University of Washington. Retrieved from

    59. Herrera, C., DuBois, D. L., & Grossman, J. (2013). The role of risk: Mentoring experiences and outcomes for youth with varying risk profiles. New York, NY: A Public/Private Ventures project distributed by MDRC. Retrieved from

    60. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219.

    61. Taussig, H. N., Culhane, S. E., Garrido, E., Knudtson, M. D., & Petrenko, C. L. M. (2012). Does severity of physical neglect moderate the impact of an efficacious preventive intervention for maltreated children in foster care? Child Maltreatment, 18, 56–64.

    62. Salazar, A. M., Keller, T. E., & Courtney, M. E. (2011). Understanding social support’s role in the relationship between maltreatment and depression in youth with foster care experience. Child Maltreatment, 16, 102–113.

    63. Weiler, L. M., & Taussig, H. N. (2017). The moderating effect of risk exposure on an efficacious intervention for maltreated children. Journal of Clinical Child & Adolescent Psychology [Epub ahead of print]. Retrieved from

    64. Thompson, A., Greeson, J. K. P., & Brunsink, A. M. (2016). Natural mentoring among older youth in and aging out of foster care: A systematic review. Children and Youth Services Review, 61, 40–50.

    65. Farruggia, S. P., & Sorkin, D. H. (2009). Health risks for older US adolescents in foster care: The significance of important others’ health behaviors on youths’ health and health behaviors. Child: Care, Health and Development, 35, 340–348.

    66. Blakeslee, J., Quest, A. D., Powers, J., Powers, L. E., Geenen, S., Nelson, M., … & Research Consortium to Increase the Success of Youth in Foster Care. (2013). Reaching everyone: Promoting the inclusion of youth with disabilities evaluating foster care outcomes. Children and Youth Services Review, 35, 1801–1808.

    67. Taussig, H. N., Harpin, S., Betts, W., Melnicoe, L., & Russo, G. J. (2016). Youth in foster care. In L. S. Neinstein, D. K. Katzman, S. T. Callahan, C. M. Gordon, A. Joffe, & V. I. Rickert (Eds.), Neinstein’s adolescent and young adult health care: A practical guide (pp. 642–645). Philadelphia, PA: Lippincott Williams & Wilkins.

    68. Kwak, Y., Lu, T., & Christ, S. L. (2017). Organized and unstructured activity participation among adolescents involved with child protective services in the United States. Child and Youth Care Forum, 46, 495–517.

    69. Avery, R. J. (2011). The potential contribution of mentor programs to relational permanency for youth aging out of foster care. Child Welfare, 90, 9–26.

    70. Petrila, A., Fireman, O., Schnoll-Fitzpatrick, L., Wertheimer, R., & Taussig, H. N. (2015). Student satisfaction with an innovative internship. Social Work Education, 51, 121–135.

    71. Taussig, H. N., Culhane, S. E., Raviv, T., Schnoll Fitzpatrick, L. E., & Wertheimer, R. W. (2010). Mentoring children in foster care: Impact on graduate student mentors. Educational Horizons, 89, 17–32.

    72. Petrenko, C. L. M., Culhane, S. E., Garrido, E. F., & Taussig, H. N. (2011). Do youth in out-of-home care receive recommended mental health and educational services following screening evaluations? Children and Youth Services Review, 33, 1911–1918.

    73. Hambrick, E. P., Oppenheim-Weller, S., N’zi, A. M., & Taussig, H. N. (2016). Mental Health Interventions for Children in Foster Care: A Systematic Review. Children and Youth Services Review, 70, 65–77.

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.

Each Mentoring Model/Population Review is conducted by the National Mentoring Resource Center Research Board with the intention of examining the full body of rigorous evidence as it pertains to either mentoring for a specific population of youth (e.g., youth with disabilities, immigrant youth) or a specific model of mentoring (e.g., group mentoring, e-mentoring). Each review is built around a thorough literature review for the topic in an attempt to answer key questions about mentoring’s effectiveness, participant characteristics and program processes that influence that effectiveness, and successful implementation of relevant programs to date.

Each Review also contains an “Implication for Practitioners” section that highlights steps programs can take to use or build on this evidence base. A draft version of each review and accompanying implications for practice is anonymously reviewed by at least one practitioner and one researcher who have expertise in the topic. A Research Board member serves as the coordinating editor for each review and makes final decisions regarding the acceptability of its content, prior to submission for final review and approval by the Office of Juvenile Justice and Delinquency Prevention.

 Mentees and Mentor
Title Summary
American Indian and Alaska Native Youth This review examines research on mentoring American Indian and Alaska Native (AI/AN) youth.
Black Male Youth This review examines research on mentoring for black male youth.
Children of Incarcerated Parents This review examines research on mentoring for children of incarcerated parents (COIP).
E-Mentoring This review examines research on digital and electronic forms of mentorship, or e-mentoring.
Group Mentoring This review examines the research evidence for mentoring programs that use a group format, in which one or more mentors is matched with a group of youth for a shared mentoring experience.
Immigrant and Refugee Youth This review examines research on mentoring for first-generation immigrant and refugee youth (FG-IRY).
LGBTGI-GNC This review examines research on mentoring for youth who are lesbian, gay, bisexual, transgender, questioning, intersex, and gender nonconforming (LGBTQI-GNC).
Mentoring and Domestic Radicalization This review examines research as it relates to mentoring and domestic radicalization.
Mentoring for Enhancing Educational Attitudes, Beliefs, and Behaviors This review examines research that addresses the potential influence of mentoring for youth on their educational attitudes, beliefs, and behaviors (EABBs).
Mentoring for Preventing and Reducing Delinquent Behavior Among Youth This review examines research as it relates to mentoring as a prevention strategy for delinquent behavior.
Mentoring for Preventing and Reducing Substance Use and Associated Risks Among Youth This review examines research on youth mentoring as a strategy for preventing and reducing adolescent substance use, including opioids.
One-to-One Cross-Age Peer Mentoring This review addresses four topics related to one-on-one cross-age peer mentoring for children and adolescents.
Youth and Young Adults During Reentry from Confinement This review examines research findings related to mentoring youth and young adults who are reentering their communities after confinement by the justice system.
Youth in Foster Care This review examines research on mentoring youth in foster care.
Youth Involved in Commercial Sex Activity This review examines research on mentoring for youth with backgrounds of involvement (or high-risk for involvement) in commercial sex activity (YCSA).
Youth with Disabilities This review examined research on mentoring for youth (ages 25 and younger) who have a disability, including physical, cognitive, learning, and developmental disabilities, and excluding psychiatric disabilities which have been discussed elsewhere.
Youth with Mental Health Challenges This review examined research on mentoring for youth (ages 18 and younger) who are experiencing mental health challenges.


Mentoring and Domestic Radicalization
PDF button Facebook button Twitter button

May 2017

This review examines research as it relates to mentoring and domestic radicalization. The review is organized around four questions:

  1. What is the effectiveness of mentoring for preventing or reducing domestic radicalization among youth?

  2. What factors influence the effectiveness of mentoring for preventing or reducing domestic radicalization among youth?

  3. What pathways are important in linking mentoring to prevention or reduction of domestic radicalization among youth?

  4. To what extent have mentoring initiatives with potential to prevent or reduce radicalization reached youth most likely to benefit, been implemented with high quality, and been adopted and sustained by host organizations?

Research directly addressing mentoring as it relates to domestic radicalization among youth is extremely limited in amount (nine studies) and scope (e.g., carried out primarily in non-U.S. contexts) and as a whole is lacking in methodological rigor (e.g., no well-controlled studies of program effectiveness). As such, available evidence is largely insufficient for answering any of the above questions. Currently, a significant proportion of research and interventions addressing radicalization and violent extremism have focused on acts committed by those affiliated with Islam. However, extant findings do suggest a number of noteworthy possibilities. These include:

  • The potential for program-supported mentoring to enhance core indicators of positive development among youth who, collectively, may be relatively more vulnerable to radicalization (e.g., those from marginalized communities or stigmatized cultural groups); such indicators include social connections with diverse peers, and confidence in being able to successfully pursue postsecondary education and obtain employment;
  • The potential for mentoring to help forestall or interrupt the emergence of attitudes that may reflect tendencies toward radicalization among youth (e.g., a belief that violence toward others in society is justified based on religious or political tenets);
  • The potential for processes significant in linking mentoring to prevention or reduction of radicalization and violent extremism among youth to include both a) those identified as being of general importance when mentoring youth—such as forging of a close and trusting bond and engaging in activities to promote core aspects of positive youth development—and b) other processes that have more specific relevance to susceptibility to radicalization—such as direct discussion of ideological beliefs and engineering of positive contacts with members of other cultural groups;
  • The value of partnerships comprised of diverse local community government and nongovernment entities and stakeholders for facilitating the development, implementation, and reach of initiatives involving mentoring that have aims of contributing to prevention or reduction of radicalization among youth.
  • Both practical (i.e., identifying young persons expected to be most appropriate for participation) and sociopolitical concerns (e.g., perceptions of stigmatization and stereotyping) as barriers to the engagement of youth in mentoring initiatives associated with efforts to prevent radicalization and violent extremism as well as enhanced youth engagement when young persons have meaningful roles in the development or implementation of programs (e.g., peer mentoring).

Insights for practice based on currently available knowledge are appended to this review. This commentary notes that, even when not specifically targeting the prevention or reduction of radicalization and violent extremism, mentoring programs may be in a position to influence factors that have been theoretically linked to radicalization. These include a sense of community and connectedness for youth who may otherwise feel isolated from and marginalized by the dominant culture. Mentoring programs are also encouraged to examine how they may enhance their existing programming to provide training to mentors and program staff on warning signs of possible radicalization among the youth they serve. The commentary also takes note of the encouraging examples of programs that have involved law enforcement in their work to their advantage (e.g., helping to break through stereotypes and foster constructive dialogue) while at the same time highlighting a range of potentially formidable dynamics (e.g., feelings of distrust) and safeguards (e.g., protection of rights to privacy) that merit careful attention in any such efforts.

  • Introduction

    Acts of terror have an enormous economic and human cost. In 2015, the global economic impact of terrorism was estimated at $89.6 billion, its second highest level since 2000, with the human cost reflected (in part) in the 29,376 associated deaths.1 The last 15 years have seen the economic and opportunity costs arising from terrorism grow approximately elevenfold and deaths of private citizens due to terrorism increase approximately sixfold.1 Geographically, terrorism tends to be concentrated in a relatively small number of countries, with four (Afghanistan, Iraq, Nigeria, and Pakistan) being the location for more than half (57 percent) of terrorism-related deaths since 2000.1 The United States and other Western countries account for a relatively small proportion of terrorism globally. Notably, though, 2015 was the worst year on record (since tracking began in 2000) for terrorism in Organisation for Economic Co-operation and Development (OECD) countriesi, with the number of attacks rising for the sixth consecutive year to a total of 731.1 Among OECD countries, the United States had the third highest number of deaths from terrorism in 2015 and in the first half of 2016.1

    Although there is no one agreed upon definition of terrorism, it is often thought of as involving the use of violence against multiple targets/victims to effect societal, political, religious, or ideological change.2 Domestic terrorism in the United States includes acts committed by right-wing, antiabortion, environmental, and religious extremists. For example, the United States had 4,420 right-wing terrorism incidents between 1990 and 2012.3 However, in more recent years, the threat of domestic terrorism by individuals affiliated with ISIL and Al-Qaeda has dominated the conversation. According to the Center on National Security at Fordham Law School4, there were 368 cases of terrorist attacks associated with Islamic groups in the United States between 2001 and 2013. However, it is important to note that there is no evidence that violent extremism is particularly likely to emanate from any specific established religious tradition, ideology, or belief system.5 In recent times, most terror attacks have been committed by lone actors rather than domestic or international terrorist organizations, accounting for 98 percent of all deaths from terrorism in the United States since 2006.1 These include the San Bernardino attack in which 14 were killed, the attack on attendees of the Emanuel African Methodist Episcopal Church in South Carolina that killed 9, and the Orlando nightclub shooting that killed 50 and is suspected to be inspired by ISIL.ii Furthermore, individuals involved in terrorist attacks in recent times have been relatively young, with mean ages between 25 and 293, 4 and the Internet and social media are increasingly being used as tools for recruiting and planning for terrorist attacks.6

    Closely intertwined with the problem of terrorism is the concept of radicalization. Definitions of radicalization also vary considerably.7 For the most part, though, there appears to be agreement that radicalization is the process of developing extremist ideologies and beliefs,7 with extremism (in the context of liberal democracies) understood to refer to “an ideology that advocates racial or religious supremacy and/or opposes the core principles of democracy and universal human rights”.7 USAID’s definition of violent extremism as “advocating, engaging in, preparing, or otherwise supporting ideologically motivated or justified violence to further social, economic, and political objectives”8 (page 2, emphasis added) illustrates the manner in which radicalization and terrorism are understood to be interconnected. Yet it should be kept in mind that many (possibly most) individuals who might be described as “radicalized” do not engage in or actively support terrorism-related violence;iii likewise, not all terrorist acts need be committed by those with deeply held extremist beliefs or ideologies.7

    Research on factors leading to radicalization, and among this group factors leading some to engage in or otherwise support violent extremism, is quite limited to date. Available findings indicate that these influences are multifaceted, distributed across individual/psychological, social/group, and societal levels, and variable across individuals, time, and context.9 Illustratively, based on extensive analyses of over 5,000 data sets, the Institute for Economics and Peace (IEP)1 has identified numerous country-level factors that are associated statistically with terrorism, but also found that these factors tend to be distinctively different for OECD and non-OECD countries. For OECD countries, the IEP1 noted that “socioeconomic factors, such as youth unemployment, militarization, levels of criminality, access to weapons and distrust in the electoral process, are the most statistically significant factors correlating with terrorism” (page 3). Similarly, there is accumulating empirical support for a range of psychological and social/interpersonal processes as potentially contributing to radicalization and violent extremism. These factors include personal motivations to redress grievances and receive anticipated rewards (e.g., money), socially-facilitated entry through family/kinship and other close network ties, and needs for belonging, sense of identity, or personal meaning that may be met through group affiliation.9,10

    Some observers of the foregoing types of potential influences on susceptibility to radicalization or violent extremism have proposed the usefulness of distinguishing what have been referred to as “push” and “pull” factors. Hassan11 describes the two types of factors as follows:

    Push factors are the negative social, cultural, and political features of one’s societal environment that aid in “pushing” vulnerable individuals onto the path of violent extremism. Push factors are what are commonly known as “underlying/root causes” such as poverty, unemployment, illiteracy, discrimination, and political/economical marginalization. Pull factors, on the other hand, are the positive characteristics and benefits of an extremist organization that “pull” vulnerable individuals to join. These include the group’s ideology (e.g., emphasis on changing one’s condition through violence rather than “apathetic” and “passive” democratic means), strong bonds of brotherhood and sense of belonging, reputation building, prospect of fame or glory, and other socialization benefits.

    To some extent, mirroring this diversity of suspected drivers of violent extremism, approaches to combating terrorism, and radicalization as a precursor to it have been quite varied. These also continue to evolve, seemingly at least in part in response to emerging understandings from research. Of particular note, in 2015 the Institute for Economics and Peace12 noted that whereas “traditional counterterrorism approaches have targeted terrorist activity directly through increased security measures . . . as the understanding about the drivers of terrorism improves, discussion has shifted to prevention strategies so as to reduce the pool of individuals that may choose to participate in terrorist activities” (page 74). In line with this trend, the National Institute of Justice has focused its funding for research and evaluation in the area of domestic radicalization on “Community-level demonstration programs to prevent radicalization to violent extremism [that] may involve a variety of strategies and activities” (page 5). These programs include primary prevention strategies (i.e., those focused on reducing the likelihood of radicalization by working with broad groups, communities, or populations through such activities as antiviolence messaging and education), secondary prevention strategies (i.e., programs directed at individuals who have been identified as being at high-risk for becoming radicalized), and intervention strategies (i.e., approaches that have the aim of aiding the disengagement of radicalized individuals and/or de-radicalizing those who have already adopted extremist ideologies, but are not engaged in planning or carrying out acts of violence). Also of note are more theoretically derived taxonomies of strategies. Davies,13 for example, recently drew on insights from complexity science and theory to outline four types of strategies within education that could interrupt the spread of violent extremism: introducing turbulence through value pluralism, working within the enabling constraints of human rights, building confidence and resilience, and developing networking for social change.

    Given the social and interpersonal factors that have been implicated in radicalization (e.g., need for belonging) and the growing interest in preventive approaches and those that are predicated on adult guidance and influence, it is not surprising that mentoring of youth has been widely proposed as a potentially useful approach for combating violent extremism both in the United States and abroad. The present review takes stock of research that pertains to this topic, with a focus on the following four questions:

    1. What is the effectiveness of mentoring for preventing or reducing domestic radicalization among youth?

    2. What factors influence the effectiveness of mentoring for preventing or reducing domestic radicalization among youth?

    3. What pathways are most important in linking mentoring to prevention or reduction of domestic radicalization among youth?

    4. To what extent have mentoring initiatives with potential to prevent or reduce domestic radicalization reached youth most likely to benefit, been implemented with high quality, and been adopted and sustained by host organizations?

    A systematic literature search was conducted to identify journal articles, book chapters, and other types of reports that have reported findings pertinent to one or more of the preceding questions. This included searches of PubMed, Proquest Dissertations and Theses, PsycINFO, and Google Scholar using an established set of keywords as well as outreach to a listserve on youth mentoring research and practice. Additionally, research referenced in relevant chapters (e.g. those in either edition of the Handbook of Youth Mentoring14, 15) and prior literature reviews in the areas of both mentoring and radicalization/violent extremism was reviewed for potential relevance. Several notes are in order as to what type of research was considered to be within the scope of the review.

    First, the primary interest was in research focused on mentoring as defined by the National Mentoring Resource Center (i.e., relationships and activities that take place between youth [i.e., mentees] and older or more experienced persons [i.e., mentors] who are acting in a nonprofessional helping capacity—whether through a program or more informally—to provide support that has its aim or realistic potential benefitting one or more areas of the young person’s development; for further detail, see What is Mentoring?). This definition excludes services and supports that are offered in formal professional roles by those with advanced education or training (e.g., social work, counseling) as well as those that are exclusively or predominantly didactic in orientation (e.g., structured curriculum). However, for purposes of the present review, these requirements were relaxed to some degree in view of the limited amount of available research. Notable, too, is that, although “criminal mentoring” and related processes have been implicated in the radicalization of some young persons (see, e.g., Simi et al.14), the present review focused on mentoring with more salutary and prosocial intentions.

    Second, research needed to either focus on domestic radicalization (e.g., evaluation of a mentoring program with this goal) or on selected outcomes that may represent particularly notable risk or protective factors for radicalization. These latter outcomes include selected indicators of positive youth development (e.g., morality-based components of character, feelings of belonging to a prosocial group or institution, confidence or optimism about future possibilities such as postsecondary education or employment) as well as beliefs and attitudes concerning violent extremism or closely related ideologies. Consideration of a relatively broad array of outcomes, rather than only those with obvious direct relevance to radicalization or violent extremism, is in line with the argument made later in this review that the Positive Youth Development framework17 provides a useful lens through which to consider factors that may be more broadly and fundamentally protective in relation to susceptibility to radicalization.

    Third, youth of all backgrounds (e.g., religious, racial, socioeconomic) were included in keeping with the understanding that the potential for violent extremism is not restricted to any one particular group or population of young persons.

    Fourth, in keeping with research suggesting that radicalization is a process that may occur gradually over an extended period of time and thus not necessarily confined to a single phase of development,9 the age range of youth in eligible studies was extended to include young adulthood (i.e., samples of young persons as old as age 25).

    Finally, a large proportion of initiatives to counter or prevent violent extremism (including those making use of mentoring specifically) have taken place in other countries; for this reason, research was not limited to the U.S. context. Clearly, though, it is extremely important to keep in mind the potential, and indeed likely, limitations of attempting to generalize findings from research conducted in other countries to the United States.

    A total of nine studies met criteria for inclusion in the review; four studies were primarily or exclusively quantitative, two were qualitative, and the remaining three were mixed methods (i.e., combination of quantitative and qualitative). Table 1 includes a description of several different programs that have been evaluated in these studies as well as additional programs or initiatives that were identified in the process of conducting this review, but that have not been evaluated to the best knowledge of the review authors. It should be kept in mind throughout this review that the range of potential types of radicalization and extremism is considerably greater than what is reflected in the identified research and program examples. Concern with radicalization linked to either intentional distortion or unintentional misunderstanding of Islamic religious teachings is well-represented (and, in fact, arguably overrepresented). In contrast, work focused on exploring or leveraging mentoring’s potential to reduce the likelihood a young person will gravitate toward violence in conjunction with other religious viewpoints (e.g., Christianity) or any of a variety of nonreligious beliefs or grievances (e.g., racial superiority, animal rights) is notably absent.

    i The OECD currently consists of 35 member countries and includes many of the world’s most advanced countries, such as the United States, Germany, and Australia, but also emerging countries like Mexico, Chile, and Turkey. For a complete list go to

    ii Although clearly significant, the economic and human costs of terrorism are relatively small in comparison to other forms of violence. During 2015, for example, terrorism accounted for only 1 percent of the total global economic impact of violence, which reached $13.6 trillion;12 likewise, globally, the homicide rate is 13 times that of the rate of deaths attributable to terrorism.1

    iii It is worth noting that this perspective, in which radicalization does not always serve as a precursor to terrorism, does not appear to be universally held. Illustratively, the National Institute of Justice, in its most recently issued guidelines for applications for research and evaluation of prevention and intervention demonstration programs focused on domestic radicalization, defines radicalization as “the process by which individuals enter into violent extremism” (page 5).

  • What is the Effectiveness of Mentoring for Preventing or Reducing Domestic Radicalization among Youth?


      Mentoring is frequently included in initiatives to prevent violent extremism, suggesting that it is viewed as having the capacity to make a measurable contribution to preventing or reducing radicalization among young persons. As with other suggested strategies to curb radicalization or extremist activity, these efforts as a whole have not been guided by well-developed theories of change (i.e., delineation of specific processes through which mentoring could affect radicalization and related outcomes and the conditions under which these are expected to be most likely to occur). Among a myriad of theoretical perspectives that could be brought to bear on this question, one that appears particularly promising is positive youth development. Below, this framework is considered with respect to the question of why mentoring could be expected to be effective in preventing or reducing radicalization. Attention is also given to aspects of the framework that suggest ways that mentoring can be ineffective or even counterproductive for this purpose. Examples of mentoring programs that appear to be aligned with important components and assumptions of the framework are provided.

      Overview of Positive Youth Development framework.

      Positive Youth Development (PYD) is a strength-based perspective that has provided an influential counterpoint to more deficit-oriented approaches to research and practice directed toward “at risk youth” over the past two decades.34 As summarized recently by Erdem and colleagues,35 “One of the major premises of the PYD framework is that youth development takes place in a system of bidirectional processes between youth and their ecological context through which youth build strong relationships with parents, peers, teachers, and other adults and connect to communities, schools, and other institutions.36, 37, 38 Youth, as constructive agents of their development, use such ecological assets and resources to achieve healthy development in academic, psychological, social, and moral domains and continue to both grow and ultimately contribute to civil society as young citizens”37 (page 466). Along those same lines, the Social Development Model39 postulates that when youth have opportunities for bonding to persons and institutions whom they regard as holding prosocial values, this will reduce the youths’ susceptibility to involvement in antisocial behavior such as substance use or violence. Conversely, if such assets are lacking (for example, youth have opportunities primarily to form positive bonds only with those not perceived to hold prosocial values or those who overtly espouse antisocial viewpoints), involvement in problem behavior is expected to become more likely.

      Lerner and colleagues17, 37, 40 have proposed that PYD consists of assets in five key domains, referred to frequently as the “5 Cs”: competence (social, academic, and/or cognitive skills); confidence (positive self-worth, self-efficacy); connection (positive bonds with people and/or institutions); character (sense of morality and integrity); and care and compassion (sense of sympathy and empathy for others). When the 5 Cs develop, they are expected to contribute to the emergence of a “sixth C” of contribution that includes (in part) actions taken to benefit one’s community and institutions of civil society (e.g., neighborhoods, schools, religious groups).17 Following from the basic premise that as positive behaviors increase negative behaviors will decrease, PYD, as reflected by the 6 Cs, is expected to be associated with a decline in problem behaviors, including aggression and delinquency.17 This idea, which also has received robust empirical support17, suggests one potential pathway of PYD to prevent or reduce violent extremism. More compelling, perhaps, are the linkages that are readily apparent between the Cs themselves and factors implicated in radicalization and violent extremism. These include, for example, unmet needs for a sense of belonging to a group or community (connection) or personal meaning and purpose (confidence), lack of opportunities for success in important arenas of development such as school and employment (competence), moral disengagement (character), and undeveloped empathy or sympathy for others whose actions one could affect (caring). Positive contributions to one’s community through activities such as volunteering and participation in more sociopolitically oriented efforts to promote social justice, furthermore, can arguably be conceptualized as the antithesis of violent extremism that abrogates the rights and welfare of other groups.

      In line with the foregoing possibilities, a systematic review of the effectiveness of interventions to prevent violent extremism among young people found that those emphasizing capacity building or empowerment were among the most effective.41 For programs focused on violent extremism generally, effectiveness was associated with debate and discussion as well as education and training, whereas in the case of those emphasizing prevention of violent extremism in the name of religion “outreach/peripatetic mechanisms and multiagency working” were prominent among factors linked to apparent effectiveness. The authors of this review (which also considered programs directed toward other “target population groups,” such as women and entire communities) cautioned that none of the studies involved robust quantitative analyses, making their findings and conclusions highly tentative.

      Application of PYD to mentoring.

      Within the framework of PYD, mentoring can be conceptualized as having the potential to serve as an important ecological asset that may promote positive development for youth.35 Consistent with this understanding, Lerner and colleagues17 highlighted a range of specific ways in which mentors of youth could potentially promote each of the 6 Cs. These include, for example, helping mentees to identify and explore their special interests or hobbies (competence); feel loved and valued (confidence); have their voices heard in the community in ways that support a sense of mattering (connection); understand the values that their mentors hold, the behaviors or activities they do and do not find acceptable, and why (character); have a positive role model for caring in their mentors (caring); and participate in causes that align with their interests and concerns (contribution).

      In line with these possibilities, activities or opportunities aimed at fostering positive adult-youth relationships have been identified as an important feature of programs that contribute to improvements on indicators of positive development among youth38 and thus are widely recognized as a core feature of positive developmental settings.42 In support of these possibilities, a recent study of youth mentored in the Big Brothers Big Sisters community-based program in Canada found that reports of greater support from mentors (including scales assessing “Developmental Support”—an example item for which was “Tries to find out what I like to do”—and “Practical-Oriented Support”—an example item for which was “Teaches me a skill or how to do things”) was predictive of higher assessed levels of PYD (the indicators for which were measures of the 5 core Cs).35 In this way, mentoring may be able to serve, in effect, as scaffolding that enables youth to satisfy, in a prosocial and adaptive manner, the same types of basic needs (e.g., for belonging and a positive identity) that appear to attract some young persons to extremist ideas or groups when otherwise unmet. Viewed from the perspective of the distinction between the previously described “push” and “pull” factors in the process of radicalization,9this possibility suggests mentoring may have the potential to help mitigate the influence of “pull” factors by engaging young persons in functionally equivalent, but prosocial pathways of development. This might take the form of either preventing the emergence of extremist attitudes or behaviors or counteracting such tendencies if already apparent. Reflecting this viewpoint, a recent report of the Federal Bureau of Investigation43 on preventing violent extremism noted the following: “Normal developmental vulnerabilities common to adolescents make some amenable to the influence of violent extremism, a trajectory that, through inhibitors such as community engagement, mentoring, therapy, and education, can be altered or suppressed.” (emphasis added, p. 16).

      A set of mentoring programs developed by Mosaic, a government-supported organization in the United Kingdom, with the aim of supporting youth in marginalized, predominantly Muslim communities, are illustrative of a PYD-aligned approach to supporting young persons whose life circumstances have the potential to make them vulnerable to radicalization (see Table 1). The organization’s Primary School Programme, which serves 9- to 11-year-old girls and their mothers, is school-based and uses a group format in which each mentor (a female volunteer from a professional background) is paired with a small number of girls and their mothers. Sessions are facilitated by a teacher and structured so that the children’s mothers variously participate together with or separate from their daughters, with a resource booklet used to help mentors address themes such as confidence, communication, and educational aspirations for girls and citizenship and understanding the British education system for their parents. The Secondary School Programme, which also uses a group format and serves youth ages 11 to 18, is geared toward enhancing life skills and employability (competence), improving self-efficacy and educational aspirations (confidence), and reducing feelings of isolation (connectedness). Mosaic’s Ex-Offender Programme is directed toward young persons ages 18 to 30 and uses a one-to-one mentoring model. The PYD-oriented aims of these programs include establishing a long-lasting relationship that continues into the community-integration phase and in which mentors provide practical assistance in areas such as securing housing and employment.

      These examples illustrate ways that mentoring efforts, which extend across youth of varying developmental levels and life circumstances, can benefit efforts to counter violent extremism. They also are consistent with the potential suggested above for mentoring to be oriented toward either interrupting or redirecting putative risk processes for radicalization, such as criminality, or preventing such processes altogether.

      By the same token, mentoring relationships may not necessarily encompass the attributes expected to promote core indicators of PYD. Illustratively, the formation of a close bond in which the mentee feels deeply cared for and validated by a mentor may facilitate growth in areas such as connection and confidence. Yet the realization of this type of bond is by no means a routine feature of either naturally occurring or program-supported mentoring relationships among youth (e.g., see Bayer et al.44).45 Thus, it is not difficult to imagine that there might be limited or even no benefits of mentoring relationships with respect to reducing radicalization or associated behaviors among young persons. Of further note is the possibility for mentoring to undermine aspects of PYD in ways that could increase potential for radicalization and associated behavior. These include, for example, diminished feelings of self-confidence and sense of connectedness in the wake of perceived abandonment when mentors fail to follow through on commitments46, 47 as well as inadvertent or even intentional encouragement of problematic values or behaviors through the example that is set by the mentor’s own actions.48 The recruitment of individuals into violent extremism, in fact, has been described as involving processes of social bonding and influence that bear noteworthy similarities to those understood to be central to prosocial forms of mentoring.7


      Six of the studies included in this review reported findings that address the question of how effective mentoring may be for prevention or reduction of radicalization. Three of the studies reported on outcomes relating predominantly to PYD, whereas the remaining three reported on outcomes with more direct conceptual ties to the radicalization process and extremism. Three of the studies included a comparison group of youth not participating in the mentoring program; however, in all instances the design used was quasi-experimental (i.e., the comparison group was constructed by researchers from a pool of youth not participating in the program, rather than on the basis of random assignment to the mentoring program or comparison group). Other notable limitations of the extant evaluations include small sample sizes, lack of details regarding procedures for constructing the comparison group, substantial percentages of study participants without post-test data on outcomes, and unclear or missing information on statistical tests for significance. Further details on the methodology and findings of each study are provided in Table 1.

      Positive Youth Development.

      In a quasi-experimental evaluation of the Nightingale Project mentoring program in Spain for students ages 10 to 16 of foreign origin, findings were consistent with benefits of the program for participating youth, relative to comparison group youth, in several areas.23 These include greater learning of the Catalan language, establishing broader and more diverse networks of friends in school, developing higher educational aspirations and expectations, becoming better acquainted with the new municipality in which they were living, and improved self-confidence and self-esteem. Findings of another quasi-experimental evaluation of outcomes for youth participating in the Secondary School Programme of Mosaic21 (referenced above) similarly suggested improved aspirations and expectations for postsecondary education as well as greater confidence in being able to find a job after schooling for mentored youth relative to those in the comparison group. In contrast, in the US context, a quasi-experimental evaluation of a school-based group mentoring program within a highly ethnically diverse high school (see Lapidus29 in Table 1) found no evidence of an effect of mentoring program participation on students’ reports of ethnic group belonging and an unexpected negative effect on their reports of school belonging.


      A pre-post design without a comparison group was used to evaluate a program within an Islamic association in the United States with the stated goals: “to implant correct Islamic teachings and the seeds of peace” (see Harun20, in Table 1). The program involved the Imam teaching youth about the Islamic faith in the context of a support group comprised of 10 male and female teens from Yemen. The developer of the program (who also served as group facilitator and evaluator) identified the youth involved as those in the congregation who appeared to be at risk for radicalization due to factors such as poverty, unemployment, and “frustration over and discontent at the injustices that Muslims face” (page 12).

      The evaluation reported pre-post improvements (but no clearly described procedures for testing statistical significance) on survey questions assessing participating youths’ knowledge and beliefs related to Islamic teachings on extremism and violence. Examples of the questions asked include: “I comprehend that Islam prohibits extremism and evil activities”, “Indulging in extremism and evil activities do not serve Islam”, and “Advocates for religious violence do not adhere to Islamic teachings.”

      A mixed-methods evaluation was conducted of the Being Kenyan Being Muslim (BKBM) intervention,32 which uses an interactive/experiential group learning format to promote value complexity as a strategy for countering violent extremism and other forms of intergroup conflict. The evaluation found that scores on a measure of the construct of Integrative Complexity (IC), intended to assess the complexity with which participants think about conflicted social issues relevant to extremism, increased significantly by the end of the 16-hour intervention. In line with this finding, qualitative analyses indicated that participants demonstrated an ability to perceive some validity in different viewpoints in their oral presentations at the end of the intervention. A noteworthy feature of this intervention and evaluation is that participants included not only individuals deemed vulnerable to involvement in violent extremism (including six former members of the extremist group al-Shabaab), but also staff of the USAID-funded Kenya Transition Initiative and its grantees/beneficiaries. Although gains in IC were apparent for both types of participants, they were larger for the latter group. Of note, the authors of this evaluation recommended that one-to-one IC mentoring be provided for the most vulnerable participants, in addition to the BKBM course, to help them consolidate the gains that the group sessions produce.

      Another mixed-methods evaluation examined the Australian sports-based More than a Game program (see Table 1).25 This program, which is designed to address issues of identity, sense of belonging, and cultural isolation among young Muslim men, includes a component in which participants receive mentoring from members of law enforcement. Both qualitative and quantitative findings indicated improved attitudes to and understanding of other cultural groups, particularly Jewish people, among participants. Qualitative findings were assessed as also indicating benefits for participants in a number of areas that suggested enhanced resilience due to improvements in core components of PYD. These included character and confidence building, skills for conflict resolution, and sense of belonging. A further noteworthy finding, similar to those for the BKBM intervention, is that there was qualitative evidence of positive transformation among program stakeholders (e.g., mentors) in their perceptions of cultural differences.


      1. Preliminary evidence suggests the potential for program-supported mentoring provided to youth from marginalized communities and those with recent immigrant backgrounds to enhance indicators of PYD that may reduce their susceptibility to radicalization or violent extremism (e.g., rewarding social connections with diverse peers, confidence in being able to successfully pursue postsecondary education and obtain employment); however, the amount and quality of this research is notably limited and restricted to non-US contexts.

      2. Very limited research has examined the potential for mentoring to help forestall or interrupt the emergence of attitudes or behaviors that may reflect tendencies toward radicalization among youth; there is, however, limited “proof of concept” evidence for this possibility with respect to attitudes for mentoring carried out with Muslim youth and young adults in varying contexts (i.e., faith- or community-based).

  • What Factors Influence the Effectiveness of Mentoring for Preventing or Reducing Domestic Radicalization among Youth?


      The extent to which mentoring proves to be useful for preventing or reducing radicalization and violent extremism among young persons has the potential to be conditioned (i.e., amplified or diminished) by a wide range of factors. Theory49 and prior research50, 51 on mentoring for youth suggest the potential importance of characteristics of both the young persons who are intended to receive mentoring and those who provide mentoring to them, as well as programmatic and other contextual considerations.

      Youth characteristics.

      Research on mentoring more generally suggests that it may be particularly beneficial for youth who face conditions of environmental risk or disadvantage.50, 51 As noted previously, literature on radicalization and violent extremism frequently references factors such as socioeconomic deprivation within one’s neighborhood or community and marginalization or stigmatization of one’s cultural group by other segments of society to contribute to the emergence of radicalization and its progression toward violent extremism. Youth with higher levels of exposure to these types of contextual adversities could be among those most likely to benefit from mentoring, in part because of its potential to be supportive of young persons in ways that help to avoid or at least mitigate their harmful effects (e.g., through supporting youth with educational attainment and accessing opportunities for employment or offering experiences of positive contact with members of other cultural groups). A similar line of reasoning suggests the potential for young persons with higher levels of individual-level risk factors or experiences tied to radicalization and extremist behavior to be especially likely to show gains in outcomes of interest. A cross-sectional study of Dutch Muslim youth (N = 131), for example, found that personal uncertainty, perceived injustice, and perceived group threat experiences of being discriminated against or devalued by other groups were linked to proposed indicators of a radical belief system (e.g., perceived illegitimacy of Dutch authorities and superiority of Muslims) that, in turn, predicted more accepting attitudes of violence by other Muslims.52 Mentoring could potentially be helpful in lessening such vulnerabilities (e.g., through improvements in self-confidence, exposure or viewpoints or perspectives that offer nonviolent ways of acting on perceptions of threat or injustice, or guidance that affirms the salutatory components of a youth’s religious or other deeply held beliefs while fostering reconsideration of those that could promote intensified radicalization).

      Another possible conditioning factor in this domain suggested by a reading of the available literature is how far along (if at all) the young person being mentored is in the process to radicalization. For example, compared to youth exhibiting initial signs of radicalization (e.g., demonstrating positive interest in some of the ideas associated with an extremist ideology), youth or young adults who have already committed themselves to extremist attitudes or activity may be more difficult to reach and influence through mentoring for a variety of reasons, both psychological (e.g., confirmation bias—the tendency to attend selectively to information that supports one’s already held beliefs) and contextual (e.g., fears of loss of support and potentially even retribution from individuals or groups with whom the young person has become affiliated in conjunction with radicalization). On the other hand, considering that violent extremist behavior is rare, there may be less opportunity to demonstrate impact when mentoring is directed toward youth who do not exhibit signs of radicalization, even if they are more susceptible to radicalization based on the types of environmental and individual risk factors discussed above. Possibly, too, the degree of existing radicalization or extremist behavior could operate differently depending on the outcomes involved. Marginalized youth without such tendencies already evident, for example, may be primed to show relatively immediate improvement on indicators of PYD in response to mentoring but less so on measures of radicalization (although in the long run such benefits could well become more apparent). Further complicating matters, the manner in which youth characteristics condition the effectiveness of mentoring with regard to outcomes of interest for prevention or reduction of domestic radicalization may be contingent on the types of mentor characteristics and program practices and design features that are discussed in the following sections.

      Mentor characteristics.

      One characteristic of mentors that could theoretically condition the effectiveness of mentoring for prevention or reduction of radicalization or extremism is whether they share the same religious or cultural background as their youth mentees. It may be that mentors whose backgrounds align with youth in this regard could be particularly effective. They may, for example, have knowledge and experiences that make them especially adept at helping youth to cultivate positive ethnic or racial identities or non-extremist understanding of tenets of religious or other beliefs and may be viewed as more credible sources of guidance in these respects by youth. In alignment with these possibilities, several of the mentoring programs summarized in Table 1 have utilized mentors who share the cultural or religious backgrounds of the youth involved. In a recent National Mentoring Resource Center review of research on mentoring for first-generation immigrant and refugee youth, for example, Oberoi54 noted some support for the idea that mentors associated with the country of resettlement may be able to serve as cultural and system translators for immigrant and refugee youth, such as by facilitating language learning, providing exposure to the behavioral and social norms of the new country’s culture, and serving as a source of “bridging” social capital in ways that help connect youth to important resources and institutions. Programs that are clearly consistent with this perspective can be found in Table 1. The same review also noted a theoretical potential for second-generation mentors who are from the same cultural background as immigrant or refugee youth to prove especially effective to the extent that they are able to capitalize on their familiarity with important aspects of both the youth’s culture of origin and that of the new country. Similar benefits may be realized (at least in part) even when mentors do not share the youth’s background, if they are culturally competent, potentially with the aid of training or other programmatic supports (as is discussed further below). On the other hand, it should be kept in mind that the broader research literature has failed to reveal consistent differential benefits for same-race/ethnicity versus cross-race/ethnicity mentoring relationships.53 Furthermore, keeping in mind the potential for experiences of marginalization and exclusion to increase risk for radicalization, positive ties with mentors who do not necessarily share the youth’s cultural background have potential to offer important benefits.

      Another noteworthy feature of the mentors utilized by some programs and initiatives with aims of curbing domestic radicalization is that the mentors themselves have previously been involved in activities or groups that support extremism. This approach is illustrated by the EXIT program, a Swedish organization supporting neo-Nazis’ disengagement from the extremist right (see Table 1). By drawing on their own experiences, such mentors (sometimes referred to in the literature as “formers”) may be particularly well positioned to help youth who are open to the prospect of exiting extremist groups or organizations to safely and effectively navigate this process.33 A risk involved with utilization of formerly radicalized mentors is that in some instances these individuals may not have fully extricated themselves from extremist influences or have satisfactorily resolved their own issues or questions, thus suggesting a potential for selective instances of serious harm in which mentee radicalization is accentuated rather than abated. Along these lines, Christensen55 argued that the use of formers as mentors in EXIT may fail to contribute to mentees’ development and reintegration into democratic society if they (the mentors) have not contextualized and reinterpreted their own narrative of (dis)engagement and combined it with deliberate practices aligned with EXIT’s approach when interacting with mentees. Several program practices, ranging from appropriate provisions for mentor screening and ongoing support and monitoring of mentoring relationships once established, could serve to avoid these potential pitfalls. Others practices, such as utilization of developmentally advanced peers as mentors, could potentially amplify them. These possibilities further illustrate the ways in which the design features and other characteristics of mentoring initiatives with aims relating to prevention or reduction of radicalization among young persons could be influential in conditioning their effectiveness.

      Program design and practices.

      Programmatic considerations that research and/or practice-based experience have pointed to as having the potential to enhance the effectiveness of mentoring for youth more generally may well have similar implications for the impact of initiatives in which radicalization of youth is an area of concern (see, for example, recent meta-analyses34, 51 and the Elements of Effective Practice for Mentoring). Benefits may be most likely to be realized when the design features and practices involved are tailored or adapted specifically for this purpose. Possibilities already mentioned in this regard or that can be inferred from program examples include mentor screening and training, relationship monitoring and supervision, and structuring mentoring so that it incorporates opportunities for religious or ideological guidance or positive contact with youth from other identity groups (e.g., cultural, racial). Another could be the expansion of the traditional role of mentors to include advocacy,34, 51 with a focus on facilitating access to supports for PYD (e.g., skill development, educational attainment) and giving “voice” to mentees in ways to help ensure that they receive fair and just treatment within different settings, such as school or the juvenile justice system.

      Other decisions with potential consequences for effectiveness when designing mentoring programs or initiatives include the age of the mentors involved and whether mentoring is provided in a one-to-one or group format. Peer mentoring has been utilized as a strategy in several instances (see Table 1, for examples). In discussing needs for positive relational experiences among American Muslim youth, Ahmed and colleagues56 note that mentors who are closer in age may often be better able to relate to and engage these youth, and in so doing may serve as role models and as a valuable resource for navigating intersecting developmental contexts and concerns (e.g., family, friends, religion). Favorable findings reported in an evaluation of the US-based WORDE program,57 which is a multicomponent program that includes a peer gatekeeper training program, also suggest that peer mentors, particularly those within an individual’s network, may be well positioned to play an important role in recognizing signs of radicalization and intervening. However, the transient life stage that many younger mentors may be in could result in a high turnover rate, making longer-term (and potentially more impactful) relationships relatively difficult to achieve.56 Ahmed and colleagues56 also elaborate on a range of possible distinctive benefits of older adult mentors, including advocacy through their social and professional networks, serving as aspirational role models, and helping to build positive lines of communication between youth and their parents. These considerations largely echo those that have been discussed in the broader literature with respect to mentor age. As has been found in that research, both younger and older mentors may have the potential to be effective for preventing or reducing radicalization when programs are thoughtfully designed with both the opportunities and challenges of the selected age group in mind.58, 59 Similar arguments could be developed for the potential effectiveness of both one-to-one and group mentoring formats and programs in which mentoring takes place in either a particular setting (e.g., school, faith-based), the community-at-large, or online. Examples of most of these possible program design variations can be found in Table 1.

      In addition to these types of considerations are two less commonplace programmatic considerations that may have particular importance in the context of efforts to use mentoring for prevention or reduction of radicalization among young persons. One of these is whether the aim of the program is to establish new relationships or draw upon those that already exist in the youth or young adult’s social network. Although seemingly most programs have adopted the former approach, arguments also have been made in support of capitalizing on existing social connections. Illustratively, the Danish VINK program aims to support frontline workers to assume a mentoring role with young persons deemed potentially susceptible to radicalization: “Rather than assigning an external mentor that the radicalizing youth might not know and trust, VINK believes that those that are best positioned to influence radicalizing individuals are those frontline workers who already know them.”27 In line with this approach, the systematic review of programs to curb violent extremism referred to previously concluded that “outreach/peripatetic work” (which includes use of community-embedded persons) was one of the most salient factors of programs showing signs of possible effectiveness.

      A second potentially key area for strategic decision-making that merits attention has to do with whether—and if so how—to involve law enforcement in mentoring efforts directed toward preventing or reducing radicalization among young persons. A discussion of police-community engagement in counterterrorism efforts in the United Kingdom drew a distinction between strategies focused on “community cohesion” and “liberal freedoms associated with liberal democracy,” respectively.60 In the community cohesion strategy, Islamic ideology is portrayed as being in conflict with Western values and as being associated with a perceived increased risk of committing acts of violence, such that political, religious, and ethnic identities associated with being Muslim are “securitised and responded to by the state above and beyond established rules and frameworks that exist within what might be termed ‘normal politics’” (page 14).60 In comparison, the liberal freedoms approach seeks to “enable individuals to draw upon the liberal freedoms associated with liberal democracy” (page 14) so that a range of actions other than violence are considered legitimate for individuals to pursue their aims (e.g., social and political activism).

      The context and strategies for involving law enforcement in mentoring would clearly differ between these approaches. When taking an approach emphasizing liberal freedoms and democracy, police officers and policing units have become involved in partnership approaches with those community members who are “formerly and/or currently practising ‘securitised identities’” (page 14)60 to support their efforts to work with individuals deemed “at risk” of committing acts of violence, but without such mentors themselves being problematized or securitized by law enforcement. Mentoring schemes mounted from this perspective have also involved multiagency partnerships between police and other statutory agencies such as probation or housing. Such efforts broadly align approaches suggested as potentially effective earlier in this review (e.g., facilitating positive development, use of “formers” as mentors, and advocacy). Approaches associated with the other framework may alternatively position law enforcement in roles (e.g., surveillance, questioning of religious beliefs) that make it difficult to establish the conditions of trust that are widely understood as fundamentally important for mentoring of young persons to be effective.49 Spalek and Davies61 offer the following caution: “It is important to stress that there is a danger that mentoring schemes can be part of broader net-widening strategies to bring . . . particular groups of individuals to the attention of law enforcement authorities [and] can, if appropriately governed, comprise of overzealous and ill-informed flagging of individuals for ‘vulnerability’ ensuring unnecessary collection of personal data” (page 365).


      None of the research identified in the literature search for this review reported findings addressing the potential role of different factors in conditioning the effectiveness of mentoring for prevention or reduction of radicalization among young persons. It is notable, though, that the potential effectiveness of approaches reflecting different variations on several of the mentor characteristics and programmatic considerations discussed above is suggested by the findings of the evaluations discussed previously (see the initial section of this review addressing overall effectiveness of mentoring for prevention or reduction of youth radicalization). These include mentors who share the religion of the youth and those who come from more diverse cultural backgrounds, as well as those who have affiliations with law enforcement; programmatic use of schools, faith-based organizations, and the broader community as settings for mentoring; and structure mentoring to include religious guidance or positive contact with youth from other cultural groups.


      1. A wide range of youth and mentor characteristics and programmatic considerations have the potential to condition the effectiveness of mentoring for prevention or reduction of radicalization among young people, potentially in interaction with one another; however, research to address such possibilities is lacking.

      2. Existing evidence, although preliminary, suggests that the potential for mentoring to advance aims of reducing or preventing violent extremism may extend across mentors with varying backgrounds as well as programs utilizing a range of settings and strategies directed toward this aim.

  • What Pathways Are Important in Linking Mentoring to Prevention or Reduction of Domestic Radicalization among Youth?


      As noted in an earlier section of this review, developmental processes, particularly those involving the promotion of core components of PYD (i.e., the 6 Cs), have the potential to be important in linking mentoring to reduction or prevention of radicalization. The following discussion focuses on two additional types of intervening processes that may be significant in this regard: attitudes, beliefs, and experiences that may serve as precursors to radicalization or violent extremism and features of the mentoring process itself.

      Attitudes, beliefs, and experiences.

      Perceived credibility of a range of different ideologies or other sets of beliefs—particularly those that propose that the rights or welfare of one group are being fundamentally and unfairly threatened by the actions or even the mere existence of others—may serve as important contributors to radicalization among young persons. To some extent, in fact, sympathy toward such viewpoints and other related attitudes (e.g., negative views of members of other racial, cultural, or religious groups) can be regarded as indicative of an early stage of radicalization itself.7, 9 These types of attitudes may be fueled by a range of more personalized factors, including grievances stemming from perceived experiences of discrimination or unjust treatment, social isolation, and identity concerns. It will be recalled, for example, that one recent study found that personal uncertainty, perceived injustice, and perceived group threat experiences of being discriminated against or devalued by other groups were linked to indicators of a radical belief system.52 Another recent study of a large sample of Arab-Palestinian adolescents (N = 3,178) similarly found that the association of the adolescents’ perceived ethnic discrimination with reports of engaging in serious physical violence against others was partially mediated by normative (i.e., accepting) beliefs about violence.62 Notably, the associations between adolescents’ reports of direct and indirect exposures to violence in their neighborhoods and engaging in violent behavior were similarly mediated. Given that mentoring is primarily an individual-level intervention with demonstrated effects that are limited to the attitudes and behaviors of the youth who receive mentoring,51 there is a theoretical potential to contribute to amelioration of some, but clearly not all, processes that may be involved in setting the stage for radicalization and extremism.

      Mentoring process.

      The strength of the affective bond that develops between youth and their mentors is one of most robustly established processes through which such relationships can be beneficial for a range of different outcomes (e.g., psychological well-being, risk behavior, academic success).44, 63 Previously referenced theory and research also point toward social disconnectedness and unfulfilled needs for belonging as contributors to radicalization among young persons. It thus would seem that features of mentoring relationships that cultivate feelings of closeness, such as mutual sharing, engaging in activities of shared interest, and duration over time,64, 65 could all be important in processes linking mentoring to outcomes of interest for prevention or reduction of radicalization. In a group mentoring format, feelings of closeness toward not only the mentor(s) but also other participating youth could be similarly important. Other processes that could be important, as suggested earlier in this review, include teaching or guidance, intentional efforts to promote core components of positive development (e.g., self-confidence, character, leadership skills), and advocacy on behalf of youth.

      Any of these processes could prove to be more or less salient in facilitating desirable outcomes with respect to radicalization or extremism depending on the extent to which they take the form of addressing issues or concerns specific to this goal. Possibilities include teaching or guidance that incorporates discussion and perhaps even debate about ideology,41 approaches to promoting development that bring youth into positive contact with members of other groups that are most likely to be devalued or mistrusted in conjunction with radicalization, and advocacy efforts that place special emphasis on ensuring that young persons’ rights and privileges are not infringed upon.


      A few studies have reported quantitative or qualitative findings that are relevant to possible processes through which mentoring may be linked to outcomes of interest in efforts to prevent or reduce radicalization. These investigations are limited methodologically by a number of considerations, including lack of robust assessment of chains of influence from mentoring to possible intervening processes to outcomes of interest.

      An evaluation of a school-based group mentoring program referenced previously (see Lapidus29 in Table 1) found that higher scores on a combined measure of mentor/mentoring group belonging predicted increases in reported levels of both school and ethnic group belonging. In combination with the finding that participation in the mentoring program was not related to improvement on these measures, these results were interpreted as evidence of the importance of mentoring group cohesion and feelings of connection to the group’s mentor as conditions required for mentoring to provide benefits in these areas. Interestingly, if the student’s mentoring group’s focus on cultural issues was high, there was a heightened risk for the student to report a decrease in ethnic belonging. Qualitative findings from the same evaluation suggested factors that might have contributed to this finding. These include the possible downsides of heightening mentees’ awareness of critical problems they faced relating to their ethnic minority backgrounds (e.g., racism) without corresponding attention to helping them to deepen their understanding of these issues (i.e., critical consciousness) and identifying active ways of responding to them that are affirming of their ethnic identities.

      Qualitative interviews with 16 stakeholders (mentors and project staff) of the West Midlands 1-2-1 Mentoring Scheme in the United Kingdom61 revealed several points of general agreement about desirable aspects of the mentoring process for the young persons served through this program, all of whom were deemed to be at risk for violent extremism. These include utilization of both “befriending and interventionist” strategies. Befriending strategies emphasize communication of empathy and efforts to foster mutual feelings of trust, in part through resolution of issues of confidentiality. Interventionist strategies are oriented toward fostering youth empowerment without a focus on attempting to convert a youth to a specific school of theological thought. There also were several areas in which stakeholders had varying views, raising several questions about the goals and intended outcomes of mentoring in this context. Is the overall aim to support vulnerable individuals or to change them in some way? Should mentoring always involve challenging beliefs or should emphasis be placed on the strategies to achieve personal or political goals while leaving beliefs alone? Should mentoring styles be hard and confrontational or soft and empathetic; and when and for whom are varying styles appropriate? Should mentors disclose personal information about themselves?

      Finally, analysis of qualitative data collected in the evaluation of the More than a Game program25 identified several processes that appeared to be important to the program’s assessed success. One of these was the creation of a safe and supportive environment in which the participating youth felt free to explore issues that may otherwise not have been addressed. This included their relationship with police, which was addressed through “enthusiastic and wide-ranging debates” and role-plays as part of the mentoring component of the program that involved police. Also judged to be significant was the emergence and strengthening of trust and respect among young people from different communities and between young people and police across the different program activities, enhanced sensitivity to the harmful effects of dehumanizing stereotypes, and skills development in areas such as communication, leadership, and personal accountability.


      1. Available evidence suggests that several of the processes understood to be important in connecting mentoring to positive youth outcomes more generally—such as forging of a close and trusting bond and engaging in activities to promote core aspects of PYD—can also be significant in linking mentoring to prevention or reduction of radicalization among youth; however, these findings are highly preliminary due, in part, to a lack of examination of the viability of potential pathways in their entirety (i.e., from mentoring to lower levels of radicalization).

      2. Some of the processes that tentatively appear to represent viable routes for connecting mentoring to prevention or reduction of radicalization—such as direct discussion of ideological beliefs and engineering of positive contacts with members of other cultural groups—extend beyond those that have been most widely addressed in the general literature on youth mentoring; however, there is also preliminary evidence to suggest such processes (e.g., discussions focusing on culture and ethnicity) may prove ineffective or problematic when initiated with limited preparation or response planning.

  • To What Extent Have Mentoring Initiatives with Potential to Prevent or Reduce Radicalization Reached Youth Most Likely to Benefit, been Implemented with High Quality, and been Adopted and Sustained?


      Potential barriers.

      Efforts to prevent violent extremism—also often grouped together under the umbrella terms of “preventing violent extremism”67 (PVE) or “countering violent extremism” (CVE)iv—have come under considerable critical scrutiny both abroad (e.g., Lindekilde et al.68) and within the United States (e.g., LoCicero & Boyd69) for a host of reasons. These include their potential to violate civil liberties (e.g., freedom of expression, right to privacy) and to unfairly stigmatize entire religious, cultural, or political groups as well as their perceived lack of demonstrated effectiveness and potential to do harm. Illustratively, a 2012 process evaluation of programs for preventing violent extremism among young people in the United Kingdom as part of its Prevent Strategy67 found that negative reactions from the local community, partner agencies, and other practitioners regarding the initiative and its predominant focus on Muslims presented a significant challenge to implementation. Approximately half of the projects in this effort included mentoring in some form. This fact speaks to the reality that mentoring strategies operating in the same general space of concern with prevention or reduction of radicalization among youth are unlikely to be immune to the foregoing types of concerns or their fallout with respect to issues such as program adoption, reach, implementation, and sustainability.v

      There is also reason to be concerned about the capacity of initiatives to sustain youth involvement for significant durations of time (e.g., one year or more). In the above referenced process evaluation, approximately two-thirds of the young people participating in those interventions that took a targeted or preventative approach were involved with them during only one quarter (i.e., a three-month period) of their time of operation; just over 3 percent were engaged with the projects for a year or more. The absence of collaborative relationships with social service and law enforcement agencies also appear to be barriers to help-seeking behaviors among individuals who may notice signs of radicalization in youth.57

      Potential facilitators.

      As noted previously, one identified trend in the broader literature on programs to prevent violent extremism among youth is that programs appear to have had greater success when incorporating multiagency partnerships and outreach/peripatetic mechanisms for engaging participants (for supporting data, see also Hirschfield et al.67). In line with the potential relevance of such factors to mentoring efforts specifically, a recent study of 21 programs categorized as “community/mentoring” (e.g., Big Brothers Big Sisters) found that predictors of program sustainability included connection to a well-functioning coalition.70 Both the serious concerns voiced regarding initiatives to prevent violent extremism and the identified value of localized outreach efforts suggest that partnerships may be most successful when they include groups and organizations that legitimately represent the interests of youths and their communities and have ongoing relationships with them. Illustrating this approach, Building Resilience Against Violent Extremism (BRAVE), currently being implemented in Montgomery County, Maryland, seeks to foster community engagement and representation by incorporating a wide range of stakeholders, including faith community leaders, public officials, law enforcement officers, educators, social service providers, and civic activists. Youth participants in this program indicated that service learning opportunities, inclusion of arts and music, as well as encouragement from significant adults (e.g., parents and counselors) were compelling factors in their participation.57 Even in the context of such efforts, there is ample reason to expect that direct involvement of youth in areas such as program planning and implementation (e.g., peer mentoring) may still be critically important both for ensuring their engagement and facilitating implementation. This may be especially true of mentoring initiatives considering their likely dependence on relative levels of trust and investment on the part of the young persons who are intended to be served.

      iv In February 2015, a White House Summit on Countering Violent Extremism (CVE) was held; in January 2016, the US Department of Justice launched a Countering Violent Extremism Task Force.

      v To some observers in the United States (Brennan Center for Justice, n.d,), concerns about government-sponsored CVE initiatives have led them to conclude that it is inappropriate to even attempt the implementation of social programs and services such as mentoring within such initiatives.


      With relevance to partnerships and community involvement, in the evaluation of the Midlands 1-2-1 Mentoring Project in the United Kingdom,66 “caution was expressed with regard to outsourcing, as it was felt that local expertise and ownership was important” (page 5). Additional findings identified potential risk to the host organization associated with loss of reputation if a client committed an extremist act or the project was not generally successful, as well as general agreement that an appropriate “business model” for the program would include both accreditation and modest remuneration for mentors, in part so as to ensure professionalism and accountability.

      The More than a Game evaluation25 similarly identified promoting linkages between local communities, government and nongovernment entities (e.g., local media), and community consultation with a local Islamic organization as key contributors to the assessed success of the program. This evaluation also highlighted the value of being able to make appropriate changes to the program as it was being implemented. In particular, although not an initial aim of the program, the spontaneous creation of a Jewish-Muslim football team provided an avenue for cross-cultural engagement and breaking down of stereotypes that contributed considerably to the overall achievement of the program’s objectives.

      Findings pertinent to reaching and engaging youth in mentoring supports or services with connections to efforts to prevent radicalization/violent extremism were also reported in a study that included interviews with 39 young Muslims in Denmark.68 It was found that the “vast majority” of these young persons “protested against the ‘governance through individual support and response’ initiatives, in particular the role model/mentoring schemes” (page 122).68 Their concerns included the perception that the implicit message of such efforts is “discriminatory against the target groups, as it suggests [incorrectly] that these groups (i.e., young Muslims) are in particular need of role models” (page 122).68 Challenges to engagement of young persons, specifically how to identify those who were most appropriate for services, were also noted in the previously described process evaluation of projects in the United Kingdom. This evaluation also reported difficulties with recruitment and retention of both mentors (including Muslim men in particular) and staff within projects. These appeared to stem from concerns about the aims and approach of the overall initiative (in part as portrayed by press coverage), as well as, in one instance, objections to the delivery of a project by a criminal justice agency. In contrast, projects utilizing young persons both in peer mentoring and leadership roles reported relative success, including greater sustainability for projects.


      1. Partnerships comprised of diverse local community government and nongovernment entities and stakeholders (e.g., community activists) may be important for facilitating the development, implementation, and reach of initiatives involving mentoring that have aims of contributing to prevention or reduction of radicalization among youth.

      2. Barriers to the engagement of youth in mentoring initiatives associated with efforts to prevent radicalization and violent extremism have included practical challenges associated with identifying young persons expected to be most appropriate for participation as well as overt resistance stemming from sociopolitical concerns, including perceptions of stigmatization and stereotyping. Preliminary evidence suggests that the effects of such barriers can be at least partially offset through meaningful involvement of young persons in programs both as peer mentors and in leadership roles.

  • Implications for Practice

    (Mike Garringer, MENTOR: The National Mentoring Partnership & Dr. David DuBois, NMRC Research Board Chair)

    Most youth mentoring programs likely do not see their work as being part of a global response to terrorism or violent radical extremism. While many mentoring programs in the United States do focus on supporting groups of youth who are marginalized, disenfranchised, or otherwise neglected, often to the point of anger, by mainstream American society, it would be quite a stretch to describe most mentoring programs as being part of “the war on terror.” As noted in the review, radicalization and violent extremism are extraordinarily rare pathways for young people to take. In spite of the seriousness of the crimes committed by extremists, this is simply not something that most mentoring programs will seek to address directly. However, looking beyond just mentoring programs, it’s easy to see how mentoring is a strategy that may be considered by many organizations to combat radicalization of young people. Thankfully, this review offers several good starting points for thinking about the role mentors might play in this work.

    As suggested by the findings of this review that point to thwarted opportunities for PYD (as reflected in the 6 Cs: caring/compassion, connection, character, confidence, competence, and contribution), we can see how just about any mentoring program or mentor can play a role in preventing youth from becoming radicalized. Mentoring programs, and the work of mentors, in fact, are almost inherently focused on fostering a sense of belonging and meaning or purpose in young people, which are central facets of pathways to healthy development. The role of a mentor, as traditionally conceived, is to help a person find out who they are, who they can become, and how they can find acceptance, validation, and some measure of success in the world. Jean Rhodes’ conceptual model of youth mentoring speculates that identity development is one of the cornerstones of personal growth in a mentoring experience and that the development of a strong and positive identity helps youth find their way in the world. In this regard, to the extent they reach and serve marginalized youth, all mentoring programs are part of the fight to help such youth build a positive sense of self and experience love and acceptance, even if their world or community and the messages they receive from them about themselves or groups of which they are a part are far from ideal. Thus, simply by offering the love and support of a mentor to all youth who need it, America’s mentoring programs can be argued to be contributing to the fight against extremism and violence. And with that in mind, there are some things they should consider when thinking about how their services might contribute to the greater struggle to prevent violent extremism and radicalization.

    For “Typical” Mentoring Programs

    There are a few strategies suggested by the research and theory covered in the review that practitioners in just about any mentoring programs may want to consider, regardless of the scope and purpose of their services, which might help prevent mentees from becoming radicalized over time:


      As noted in the review, reversing the process of radicalization appears to become more challenging the deeper a person falls into those ways of thinking. Preventing extremism from taking root in the first place is certainly preferable to trying to turn it around after radicalization has occurred. But stopping this process early on requires knowing what to look for. Thus, both program staff and mentors may want to familiarize themselves with warning signs that a youth in their care is perhaps beginning to think in an extreme or radicalized way. This might be especially important for programs serving youth who may be experiencing bullying and harassment, those serving youth of a specific ethnic group or religion, those serving recent immigrants and refugees from war-torn parts of the globe, and programs serving older youth who may be trying to find employment and economic stability in their young adulthood.

      Although in the United States we often think of violent extremism as being something limited to certain ethnic groups or religions, it’s important to remember that violent extremism can also be part of the dominant culture or based on political views. The tenor and tone of the 2016 presidential election certainly appears to have sparked a renewed wave of white nationalistic views, armed antigovernment groups, and violent racially motivated attacks on minorities and women. So be sure to look for signs of radicalization or extreme views in any youth served by your program.

      It is worth noting here that “extreme views” does not refer to strongly held opinions or beliefs that the mentor does not personally agree with⎯every mentoring relationship is likely to have those differences of opinion. It also does not encompass the natural anger and strong emotions that can result from a young person’s newfound deeper understanding of systemic injustices. Rather, we are referring here to beliefs that categorically demonize another group of people (the “us” vs “them” mindset referenced elsewhere in this review), views that condone or accept violence as a primary solution to social or political issues, and other viewpoints or desires for action that seem driven by newfound dogma or are otherwise out of character given the mentor’s understanding of the youth’s personality and values.

      While there is no single magical list of signs that definitively predicts radicalization, there are things mentors and staff can look for, and the following resources may be helpful to programs that want to educate staff or work these concepts and warning signs into mentor training:


      Perhaps the clearest guidance in the review for mentoring programs in preventing radicalization is found in the aforementioned “6 Cs” of PYD. In these words, mentoring programs can find the strategies to combat radicalization, help youth feel better about themselves and their world, and help the disenfranchised find meaning and purpose in positive, not violent ways. Mentoring programs and mentors interested in how they might incorporate these PYD principles into their work can learn more in this series of fact sheets housed on the National Mentoring Resource Center website. Think carefully about how your program brings each of those elements to life for young people and if you find that you are missing some of those Cs, work to fill the gaps in what you are offering young people. (As an aside, these resources focus on the first of the 5 Cs of PYD as listed earlier; research suggests that bolstering these lays an important and perhaps essential foundation for enhancing young persons’ efforts to make a difference in their communities through contribution, the sixth C.)

      It’s also worth noting that those same principles that define healthy and positive development are also used by those who would radicalize our youth. Those who mold young persons into terrorists certainly spend their time teaching their versions of character and values, building up feelings of confidence and competence in executing a violent plan, and cultivating a sense of compassion for whatever disenfranchised group they claim to represent. Arguably, the fundamental difference between PYD and this misappropriated version of personal growth is the definition of “community.” Proponents of violent extremism simply don’t see the people they oppose and the victims of their violence as part of a “community” worth belonging to or even one that should be free to exist unharmed. So, if we are going to counter this approach, which might even be thought of as “negative mentoring,” we will need to provide strong and meaningful opportunities for youth to connect to the larger community and see themselves as part of a greater whole than the dim worldview offered by those who bend toward extremism. If we don’t meet those key Cs in our mentoring work, someone with much less positive ideas about the world just might.


      As noted in the review, one of the key ingredients in becoming radicalized often appears to be a sense of isolation from others and a mindset that views groups of people as an “other” or as persons not worthy of respect and civility. One of the best ways of combating this aspect of radicalization is to expose young people to individuals, institutions, and viewpoints that they may feel very negatively about or with whom they find little common ground. Part of the magic of mentoring is widely understood to be the process of exposing youth to ideas and parts of their community they might never have seen otherwise. So, think about how your mentoring program exposes mentees to the broader community, to different cultural groups, and to new experiences that might close that distance between “us” and “them.” As noted in the review, programs that build bridges between different ethnic groups, or between oppressed groups and law enforcement, appear to have the potential to go a long way toward changing negative opinions and attitudes. These programs must bring groups together cautiously and with clear intentions, purpose, and activities, lest they inadvertently reinforce negative viewpoints and stereotypes. But it makes sense conceptually that it’s probably a lot harder to think about doing harm to an “other” after you have spent some time in their company and experienced them as people, not abstract concepts. All mentoring programs have a role to play in helping mentors and adults alike in crossing boundaries and exploring differences (and similarities!). That’s work you are likely already doing (or could do more of) that can be in service of something as seemingly distant as preventing violent extremism.

    For Programs Explicitly Focusing on Preventing Radicalization

    If your mentoring program is explicitly focused on the prevention of radicalization, there are some additional strategies or considerations noted in the review that you may want to consider:


      While the research is a bit unclear in this area, mentors from the same ethnic group or religion that are encouraging the radicalization might be particularly capable of helping youth interpret problems and understand how the world works in a more positive light. One of the precursors to extremism is a narrowing or calcification of one’s worldview and thinking; those in the process of radicalization are often victim to extreme confirmation bias that bends the world into their preconceived notions. Breaking through that way of thinking might require a mentor with extreme credibility and some knowledge of the grievances that are driving the youth’s radicalization. These mentors may be especially well-positioned to help mentees know that they understand their frustrations and grievances, while also steering them toward more positive solutions. As suggested by the encouraging findings (albeit preliminary and non-definitive) of one program that used an Imam for this purpose in mentoring Muslim youth thought to be vulnerable to radicalization, such persons may also be instrumental in helping youth interpret religious texts or other indoctrinating literature more accurately or compassionately. As noted above, exposing youth to the “other” is a key part of stopping radicalization. Yet, as we learn more about the potential contributions of mentoring in this area, it will not be surprising if it turns out that facilitating a youth’s meaningful connection with a messenger or guide who has a similar background, ethnicity, religion, or life experience may also achieve status as a “best practice.” Such persons may be more capable of role modeling clear thinking and positive responses to negative circumstances than individuals from other groups or the dominant culture. Clearly, though, as suggested in the review, this type of strategy need not exclude others that are focused, perhaps in a complementary manner, on combating other potential contributors to radicalization (e.g., absence of meaningful exposure to groups or persons that represent potential victims of a young person’s radicalization toward violent extremism).


      Given the seriousness of the work at hand, programs attempting to prevent radicalization should consider providing each mentee with a dedicated mentor who can help them grow and develop in what may be very individualized and specific ways. But, as noted in the review, there also seems to be power in going through a program with a cohort of peers and having opportunities for norming new ideas and behaviors. Peers can be powerful motivators to change ways of thinking and to build that positive sense of belonging that acts as a shield against isolation and internalized anger. Groups of youth may also have an easier time contributing in meaningful ways to the community by their collective action, reinforcing the notion that there are positive and effective ways they can begin to address the issues that may have been nudging them toward radicalization in the first place. Of course, programs looking to use peers in this role should be sure to offer plenty of monitoring and support to ensure that these influential peers are not actually reinforcing negative views of others or engaging in other behaviors that might spur further radicalization.


      The review noted many examples of programs that involve law enforcement professionals in their work, helping to break through stereotypes and foster dialogue between members of a particular group and what might have seemed to some youth like “the enemy” or a representation of the very system of oppression that was driving them toward extreme views. And while there are many good reasons to involve law enforcement in this kind of mentoring work, there are also risks. Unfortunately, these types of programs may be perceived by some communities as a strategy for “keeping tabs” on them, if not outright violating their privacy and fundamental rights as citizens. Programs that involve law enforcement must have clear lines of how they will respect their clients’ privacy, right to assembly, and civil rights, and the limits of their engagement with law enforcement professionals and institutions. A program that is perceived as being a vehicle for spying or further oppression is unlikely to get buy-in from the very individuals they need to reach most.

  • References

    1. Institute for Economics and Peace. (2016). Global Terrorism Index 2016: Measuring and understanding the impact of terrorism. Sydney, Australia: Author. Retrieved from

    2. Rausch, C. C. (2015). Fundamentalism and terrorism. Journal of Terrorism Research, 6, 28–35.

    3. Perliger, A. (2012). Challengers from the sidelines: Understanding America’s violent far-right. West Point, NY: The Combating Terrorism Center at West Point. Retrieved from

    4. Center on National Security at Fordham Law. (2013). By the numbers: US prosecutions of jihadist terror crimes, 2001–2013. Retrieved from

    5. Brooks, R. A. (2011). Muslim “homegrown” terrorism in the United States: How serious is the threat? International Security, 36, 7–47.

    6. Koehler, D. (2014). The radical online: Individual radicalization processes and the role of the internet. Journal of Deradicalization, 1, 116–134.

    7. Borum, R. (2011a). Radicalization into violent extremism I: A review of social science theories. Journal of Strategic Security, 4 (4), 7–36.

    8. USAID. (2011). The development response to violent extremism and insurgency: Putting principles into practice. Washington, DC: Author. Retrieved from

    9. Borum, R. (2011b). Radicalization into violent extremism II: A review of conceptual models and empirical research. Journal of Strategic Security, 4 (4), 37–62.

    10. McCauley, C., & Moskalenko, S. (2008). Mechanisms of political radicalization: Pathways toward terrorism. Terrorism and Political Violence, 20, 415–433.

    11. Hassan, M. (2012). Understanding drivers of violent extremism: The case of al-Shabaab and Somali youth. CTC Sentinel, 5 (8), 18–20. Retrieved from

    12. Institute for Economics and Peace. (2015). Global terrorism index 2015: Measuring and understanding the impact of terrorism. Sydney, Australia: Author. Retrieved from

    13. Davies, L. (2016). Wicked problems: How complexity science helps direct education responses to preventing violent extremism. Journal of Strategic Security, 9 (4), 32–52.

    14. DuBois, D. L., & Karcher, M. J. (Eds.). (2005). Handbook of youth mentoring. Thousand Oaks, CA: SAGE.

    15. DuBois, D. L., & Karcher, M. J. (Eds.). (2014). Handbook of youth mentoring (2nd ed.). Thousand Oaks, CA: SAGE.

    16. Simi, P., Windisch, S., & Sporer, K. (2016). Recruitment and radicalization among US far-right terrorists. College Park, MD: National Consortium for the Study of Terrorism and Responses to Terrorism. Retrieved from

    17. Lerner, R. M., Napolitano, C. M., Boyd, M. J., Mueller, M. K., & Callina, K. S. (2014). Mentoring and positive youth development. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp. 17–27). Thousand Oaks, CA: SAGE.

    18. Ferguson, B., & Walker, P. (2015, Feb 4). Talking and walking – UK mentors steer young people away from radicalization. The Guardian Newspaper. Retrieved from

    19. Home Office. (2015). Channel Duty Guidance: Protecting vulnerable people from being drawn into terrorism: Statutory guidance for Channel panel members and partners of local panels. London, United Kingdom: Crown. Retrieved from

    20. Harun, S. S. (2014). Risks for religiously infused violence in Muslim youth and successful antidotes through correct Islamic teaching (Doctoral dissertation). Retrieved from ProQuest Dissertations Publishing (#3580350).

    21. Bartlett, J. (2012). A model role evaluation of mosaic mentoring programmes. London, United Kingdom: Demos. Retrieved from

    22. MOSAIC (n.d.). Secondary school programme. Retrieved from

    23. Gelis, J. F. (2015). How an intervention project contributes to social inclusion of adolescents and young people of foreign origin. Children and Youth Services Review, 52, 144–149.

    24. Johns, A., Grossman, M., & McDonald, K. (2014). “More Than a Game”: The impact of sport-based youth mentoring schemes on developing resilience toward violent extremism. Social Inclusion, 2, 57–70.

    25. McDonald, K., Grossman, M., & Johns, A. (2012). More Than a Game evaluation report. Melbourne, Australia: Centre for Cultural Diversity and Wellbeing, Victoria University. Retrieved from

    26. Australian Multicultural Foundation. (n.d.). Australian Muslim Youth Leadership & Mentorship Program: A toolkit for young leaders. Victoria, Australia: Author. Retrieved from

    27. Vidino, L., & Brandon, J. (2012). Countering radicalization in Europe. London: The International Centre for the Study of Radicalisation and Political Violence (ICSR), King’s College London. Retrieved from

    28. Temple-Raston, D. (2016, Sept 5). Mentoring program works to prevent radicalization in Copenhagen. NPR All Things Considered. Retrieved from

    29. Lapidus, R. B. (2004). An evaluation of a high school-based group mentoring program on protégés’ sense of belonging (Doctoral dissertation). Retrieved from ProQuest Dissertations Publishing (#3170736).

    30. Ramalingam, V. (2012). Policy Briefing: Far-right extremism: Trends and methods for response and prevention. London, United Kingdom: Institute for Strategic Dialogue. Retrieved from

    31. Cultures Interactive. (n.d.). NGO for intercultural education and violence prevention. Retrieved from

    32. Savage, S., Khan, A., & Liht, J. (2014). Preventing violent extremism in Kenya through value complexity: Assessment of Being Kenyan Being Muslim. Journal of Strategic Security 7 (3), 1–26.

    33. Butt, R., & Tuck, H. (n.d.). European counter-radicalisation and de-radicalisation: A comparative evaluation of approaches in the Netherlands, Sweden, Denmark and Germany. London: Institute for Strategic Thinking. Retrieved from

    34. 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.

    35. Erdem, G., DuBois, D. L., Larose, S., DeWit, D. J., & Lipman, E. L. (2016). Mentoring relationships, 3 positive development, and emotional and behavioral problems among youth: Investigation of a mediational model. Journal of Community Psychology, 44, 464–483.

    36. Larson, R. W. (2000). Toward a psychology of positive youth development. American Psychologist, 55, 170–183.

    37. Lerner, R. M., Lerner, J. V., Almerigi, J. B., Theokas, C., Phelps, E., Gestsdottir, S., . . . von Eye, A. (2005). Positive youth development, participation in community youth development programs, and community contributions of fifth-grade adolescents: Findings from the first wave of the 4-H study of Positive Youth Development. Journal of Early Adolescence, 25, 17–71.  

    38. Roth, J., Brooks-Gunn, J., Murray, L., & Foster, W. (1998). Promoting healthy adolescents: Synthesis of youth development program evaluations. Journal of Research on Adolescence, 8, 423–459.

    39. Catalano, R. F., & Hawkins, J. D. (1996). The social development model: A theory of antisocial behavior. In J. D. Hawkins (Ed.), Delinquency and crime: Current theories (pp. 149–197). Cambridge, United Kingdom: Cambridge University Press.

    40. Bowers, E.P., Li, Y., Kiely, M. K., Brittian, A., Lerner, J. V., & Lerner, R. M. (2010). The Five Cs Model of Positive Youth Development: A longitudinal analysis of confirmatory factor structure and measurement invariance. Journal of Youth and Adolescence, 39, 720–735.

    41. Pratchett, L., Thorp, L., Wingfield, M., Lowndes, V., & Jabbar, R. (2010). Preventing support for violent extremism through community interventions: A review of the evidence. London: Department for Communities and Local Government. Retrieved from

    42. Roth, J. L., & Brooks-Gunn, J. (2003). What exactly is a youth development program? Answers from research and practice. Applied Developmental Science, 7, 94–111.

    43. Federal Bureau of Investigations. (2016). Preventing violent extremism in schools. Washington, DC: Author. Retrieved from

    44. Bayer, A., Grossman, J. B., & DuBois, D. L. (2015). Using volunteer mentors to improve the academic outcomes of underserved students: The role of relationships. Journal of Community Psychology, 43, 408–429.

    45. Spencer, R. (2007). “It's not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331–354.

    46. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219.

    47. Karcher, M. J. (2005). The effects of developmental mentoring and high school mentors' attendance on their younger mentees' self-esteem, social skills, and connectedness. Psychology in the Schools, 42, 65–77.

    48. Beam, M. R., Gil-Rivas, V., Greenberger, E., & Chen, C. (2002). Adolescent problem behavior and depressed mood: Risk and protection within and across social contexts. Journal of Youth and Adolescence, 31, 343–357.

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

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

    51. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, 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.

    52. Doosje, B., Loseman, A., & Van den Bos, K. (2013). Determinants of radicalization of Islamic youth in The Netherlands: Personal uncertainty, perceived injustice, and perceived group threat. Journal of Social Issues, 69, 568–904.

    53. Sánchez, B., Colón-Torres, Y., Feuer, R., Roundfield, K. E., & Berardi, L. (2014). Race, ethnicity, and culture in mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp. 145–158). Thousand Oaks, CA: SAGE.

    54. Oberoi, A. K. (2016). Mentoring for first-generation immigrant and refugee youth. National Mentoring Resource Center Research Review. Retrieved from

    55. Christensen, T. W. (2015). When good intentions are not enough: A successful mentor-mentee relation requires a deliberated practice. Psyke & Logos, 36, 242–265.

    56. Ahmed, S., Patel, S., & Hashem, H. (2015). State of American Muslim youth: Research & recommendations. Dearborn, MI: The Family & Youth Institute, and Institute for Social Policy and Understanding. Retrieved from

    57. Williams, M. J., Horgan, J. G., & Evans, W. P. (2016). Evaluation of a multi-faceted, US community-based, Muslim-led CVE program. National Criminal Justice Reference System. Retrieved from

    58. Karcher, M. J. (2014). Cross-age peer mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp. 233–257). Thousand Oaks, CA: SAGE.

    59. Taylor, A. S., LoSciuto, L., & Porcellini, L. (2005). Intergenerational mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 286–299). Thousand Oaks, CA: SAGE.

    60. Spalek, B., & McDonald, L. Z. (2011). Preventing religio-political extremism amongst Muslim youth: A study exploring police-community partnership. Unpublished report, Institute of Applied Social Studies, University of Birmingham. Retrieved from

    61. Spalek, B., & Davies, L. (2012). Mentoring in relation to violent extremism: A study of role, purpose, and outcomes. Studies in Conflict & Terrorism, 35, 354–368.

    62. Massarwi, A. A., & Khoury-Kassabri, M. (2017). Serious physical violence among Arab-Palestinian adolescents: The role of exposure to community violence, perceived ethnic discrimination, normative beliefs, and parental communication. Child Abuse & Neglect, 63, 233–244.

    63. DuBois, D. L., & Silverthorn, N. (2005). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26, 69–92.

    64. Rhodes, J. E., & DuBois, D. L. (2008). Mentoring relationships and programs for youth. Current Directions in Psychological Science, 17, 254–258.

    65. Karcher, M. J., & Hansen, K. (2014). Mentoring activities and interactions. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (2nd ed., pp. 63–82). Thousand Oaks, CA: SAGE.

    66. Spalek, B., & Davies, L. (2010). Key evaluation findings of the West Midlands (WM) 1-2-1 mentoring scheme. Unpublished report, University of Birmingham. Retrieved from

    67. Hirschfield, A., Christmann, K., Wilcox, A., Rogerson, M., & Sharratt, K. (2012). Process evaluation of preventing violent extremism programmes for young people. London: Youth Justice Board for England and Wales. Retrieved from

    68. Lindekilde, L. (2012). Neo-liberal governing of “radicals”: Danish radicalization prevention policies and potential iatrogenic effects. International Journal of Conflict and Violence, 6, 109–125. Retrieved from

    69. LoCicero, A., & Boyd, J. W. (2016, July 16). The dangers of countering violent extremism (CVE) programs [Blog post]. Retrieved from

    70. Cooper, B. R., Bumbarger, B. K., & Moore, J. E. (2015). Sustaining evidence-based prevention programs: Correlates in a large-scale dissemination initiative. Prevention Science, 16, 145–157.

  • Table

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.