How Effective Are Mentoring Programs for Youth? A Systematic Assessment of the Evidence

During the past decade, mentoring has proliferated as an intervention strategy for addressing the needs that young people have for adult support and guidance throughout their development. Currently, more than 5,000 mentoring programs serve an estimated three million youths in the United States. Funding and growth imperatives continue to fuel the expansion of programs as well as the diversification of mentoring approaches and applications. Important questions remain, however, about the effectiveness of these types of interventions and the conditions required to optimize benefits for young people who participate in them. In this article, we use meta-analysis to take stock of the current evidence on the effectiveness of mentoring programs for youth. As a guiding conceptual framework for our analysis, we draw on a developmental model of youth mentoring relationships (Rhodes, 2002, 2005). This model posits an interconnected set of processes (social-emotional, cognitive, identity) through which caring and meaningful relationships with nonparental adults (or older peers) can promote positive developmental trajectories. These processes are presumed to be conditioned by a range of individual, dyadic, programmatic, and contextual variables. Based on this model and related prior research, we anticipated that we would find evidence for the effectiveness of mentoring as an approach for fostering healthy development among youth. We also expected that effectiveness would vary as a function of differences in both program practices and the characteristics of participating young people and their mentors. The meta-analysis encompassed 73 independent evaluations of mentoring programs directed toward children and adolescents published over the past decade (1999-2010). Overall, findings support the effectiveness of mentoring for improving outcomes across behavioral, social, emotional, and academic domains of young people's development. The most common pattern of benefits is for mentored youth to exhibit positive gains on outcome measures while nonmentored youth exhibit declines. It appears then that mentoring as an intervention strategy has the capacity to serve both promotion and prevention aims. Programs also show evidence of being able to affect multiple domains of youth functioning simultaneously and to improve selected outcomes of policy interest (e. g., academic achievement test scores). From a developmental standpoint, benefits of participation in mentoring programs are apparent from early childhood to adolescence and thus not confined to a particular stage of development. Similarly, although programs typically have utilized adult volunteers and focused on cultivating one-to-one relationships, those that have engaged older peers as mentors or used group formats show comparable levels of effectiveness. Collectively, these findings point toward the flexibility and broad applicability of mentoring as an approach for supporting positive youth development. Several other aspects of our findings, however, underscore a need for caution. These include a failure of evaluations to assess several key outcomes of policy interest (e. g., juvenile offending, obesity prevention) or to determine whether benefits for youth are sustained at later points in their development. More generally, we find that gains on outcome measures for the typical young person in a mentoring program have been modest (equivalent to a difference of 9 percentile points from scores of nonmentored youth on the same measures). This level of impact is within the range of effects observed for other types of interventions for children and adolescents but fails to reflect discernible improvement over the previous generation of mentoring programs (DuBois, Holloway, Valentine, & Cooper, 2002). Variability in program effectiveness, although less pronounced, also continues to be evident even after accounting for methodological differences in studies. In analyzing this variability, we find that programs have been more effective when (a) participating youth have either had preexisting difficulties (including problem behavior specifically) or been exposed to significant levels of environmental risk, (b) evaluation samples have included greater proportions of male youth, (c) there has been a good fit between the educational or occupational backgrounds of mentors and the goals of the program, (d) mentors and youth have been paired based on similarity of interests, and (e) programs have been structured to support mentors in assuming teaching or advocacy roles with youth. These findings suggest that effects may hinge to a noteworthy extent on decisions that are made regarding which youth and mentors to involve in a program and on the care with which mentoring relationships are established and then guided toward specific types of activities. Taking stock of the available evidence leads us to see value in continued support for youth mentoring programs. The argument for using mentoring as an intervention strategy is particularly strong when there is interest in promoting outcomes across multiple areas of a young person's development. For investments to yield optimal returns, however, there is a need for policy to be directed toward several critical areas of concern: (a) ensuring adherence to core practices (e. g., screening and training of mentors) that both research and common sense dictate to be essential elements of program quality, (b) facilitating ongoing refinement and strengthening of programs using the available evidence as a guide, and (c) fostering stronger collaborations between practitioners and researchers as a framework for evidence-driven dissemination and growth within the field. From a research standpoint, to support and inform these efforts there is a pressing need to (a) gauge the impact of mentoring interventions on key outcomes of policy interest and on the outcomes of participating youth at later points in their development; (b) utilize study designs and analyses that are capable of addressing the relative effectiveness of competing models and practices, the unique contributions of mentoring within more complex, multi-component interventions, and differences in youth responsiveness (including potential harmful effects for some youth); (c) investigate increasingly well-specified models of how different types of program practices and processes may be instrumental in shaping consequential features of mentoring relationships and ultimately, the realization of particular desired outcomes for youth; and (d) establish a research registry to improve the quality and synthesis of available evidence regarding the effectiveness of youth mentoring as an intervention strategy.