Five Key Takeaways From A Comprehensive New Meta-Analysis of Youth Mentoring
By Jean Rhodes (excerpted and summarized from publication)
Along with our colleagues, we recently released a comprehensive meta-analysis of youth mentoring: Raposa, E. B., Rhodes, J., Stams, G.J., Card, N., Burton, S., Schwartz, S., Sykes, L.Y., Kanchewa, S., Kupersmidt, J., and Hussain, S. (2019). The effects of youth mentoring programs: A meta-analysis of outcome studies. Journal of Youth and Adolescence.
Background: Our review included all outcome studies of intergenerational, one-on-one youth mentoring programs written in the English language between 1975 and 2017, using rigorous inclusion criteria designed to align with developmental theories of youth mentoring. A multilevel meta-analytic approach was used to estimate the overall effect size of youth mentoring programs, as well as explore predictors of variability in effect sizes.
What we found: Analysis of 70 mentoring outcome studies, with a sample size of 25,286 youth (average age of 12 years old) revealed that the mean effect of mentoring on youth outcomes was .21. We conclude that mentoring programs remain a modestly effective intervention for youth at-risk for a range of psychosocial and academic problems across diverse outcome domains. There may be particular benefits to targeted, time-limited approaches that draw on the service of volunteers or paraprofessionals with helping experience.
5 Key takeaways
- Effects are not tracking research advances: The effect sizes observed in these analyses are remarkably consistent with past comprehensive meta-analyses of youth mentoring, which have shown overall effect sizes ranging from .18 to .21 (DuBois et al. 2002; 2011) despite the inclusion of more recent evaluations. In the past decade, there has been an increase in mentoring programs that implement evidence-based program practices, rather than relying solely on practice wisdom, and greater implementation of the benchmark practices defined in the EEPM has been linked with match longevity (Kupersmidt et al. 2017). These effect sizes may grow if the field continues to embrace more rigorous adherence to evidence-based practices that target specific mechanisms underlying particular difficulties, rather than relying on a relatively low-intensity, nonspecific approach with uneven adherence to practices that are research-informed.
- Programs serving more boys tend to produce larger effects: Larger effects were observed for programs that served a higher percentage of male youth, consistent with at least one previous meta-analysis of mentoring, which showed that programs serving more than 50% males had stronger effects (DuBois et al. 2011). Girls may enter mentoring programs with more complicated relational histories than boys, which may initially hamper mentors’ capacity to forge productive ties with them (Bogat and Liang 2005) and lead to premature closure (Kupersmidt et al. 2017a). Mentoring also had a larger impact in samples that had a higher percentage of male mentors; however, because many programs attempt to match mentor and youth gender, it is difficult to tease apart the effects of youth versus mentor gender in these findings.
- Mentor experience matters: As noted in previous columns, this is one of the loudest signals in the noise. Consistent with previous studies, programs with a greater percentage of mentors who worked in helping professions showed larger effect sizes for youth outcomes (DuBois et al. 2002). Volunteers who have had previous experience with helping youth may feel a stronger sense of efficacy, a variable that has been consistently associated with better match outcomes (e.g., Karcher et al. 2005). For example, one recent study found that volunteer mentors with greater self-efficacy and more previous involvement with youth in their communities were more successful in working with youth from high-stress backgrounds than mentors with lower self-efficacy and less previous experience (Raposa et al. 2016).
- The length of the relationship appears less consequential than what gets done We found no significant differences in effect sizes based on program length. This is inconsistent with previous research and theory indicating the importance of longer relationships (e.g., Grossman and Rhodes 2002); however, program length was also not a significant predictor of outcomes in earlier meta-analyses (DuBois et al. 2002, 2011). This finding suggests that, although relationship length may be an important determinant of youth outcomes within a specific relationship or program–where it may signify an unexpected termination, it may be less important in distinguishing overall levels of mentoring effectiveness across programs. Interestingly, programs that had expectations for longer match meeting times actually yielded smaller effect sizes. Although more research is needed to better understand this finding, it may point to the need for programs to establish realistic expectations around the time commitment to the program (Grossman et al. 2012). Programs that specify daylong activities or multiple hours of relationship-building per session might be over-taxing the commitment of the mentor, youth, or youth’s family, raising the risk for relationship dissatisfaction or premature closure. Moreover, expectations for long match meetings might be indicative of more episodic models, through which mentor and youth meetings occur for multiple hours at a time, but over the course of just a few days or weeks, or spaced over large intervals of time.
- School-and community-based mentoring programs seem to produce similar effects This is noteworthy, given that school-based mentoring has been the fastest growing program model in recent years (Wheeler et al. 2010). Initially, there was some concern that school-based mentoring relationships may be less influential than relationships forged through community-based programs. The limited time commitment may, however, protect mentors from burnout and youth from disappointment.
Taken together, the current findings provide some support for the efficacy of one-on-one, caring relationships with adults, particularly as a low-cost intervention with the potential to reach large groups of youth and reduce the need for more intensive treatments. Nevertheless, these findings emphasize the need to remain realistic about the potential of mentoring programs as currently implemented, and highlight opportunities for improving the quality and rigor of mentoring practices and evaluation strategies