![[ Visit Client Website ]](images/banner.gif)
Formal youth mentoring programs can enhance social, emotional, behavioral, and academic outcomes for children and adolescents (DuBois et al., 2011). However, in some programs, positive effects are associated with the length of the relationship, and mentoring relationships that terminate prematurely can have negative consequences for youth participants (Grossman & Rhodes, 2002). Important early research investigating mentoring relationship endings employed retrospective qualitative methods and identified the significance of participant expectations (Spencer, 2007). The current study used a prospective design tracking a large sample of new community-based matches to examine potential predictors of relationship development and duration. It was hypothesized that expectations of the mentor and parent/guardian assessed prior to making the match would predict the likelihood and timing of mentoring relationship closure.
Methods
Adults volunteering to become mentors in four agencies affiliated with a large national mentoring organization were recruited into the study and assessed for a variety of factors expected to be associated with relational success. When these mentors were proposed for specific matches, the youth and parent/guardian were recruited and completed baseline surveys for the study prior to matching. Matches with all parties consenting to the study (n=356) were followed prospectively for at least 15 months. Several brief measures of expectations regarding the anticipated mentoring relationship were developed for this research. Match length and predictors of match closure were analyzed with event history analysis, using Kaplan-Meier and Cox regression procedures.
Results
A total of 109 matches (30.6%) ended before reaching the initial 12-month commitment expected in the mentoring program. At the conclusion of the 40-month window of study observation, 119 matches (33.3%) were still intact and consequently censored in the analyses. The estimated mean survival time for matches was 21.8 months (95%CI: 20.4—23.3), and the estimated median survival time was 18.4 months (95%CI: 15.6—21.2). Mentors who expected to enjoy fun activities together had longer lasting matches than those who wanted to focus on achieving goals, developing a meaningful bond, or connecting around intellectual interests (p=.048). Mentors who expected to have challenges spending time with their mentees were more likely to have early-ending relationships (p=.047). When mentors were asked how long they expected their relationships to last, those who were unsure or indicated “forever” were more likely to have early-ending relationships than those giving more reasonable timeframes (p=.029). In contrast, initial parent/guardian desires for relationship length corresponded directly with actual match length (p=.026).
Conclusions and Implications
Almost a third of new mentoring relationships in this study ended prematurely, before completing the yearlong commitment made at the outset. Matches ending early are less likely to achieve positive outcomes for youth and are more likely to result in disappointment and distress. The findings suggest that some straightforward questions asked of participants about their expectations for an impending match can predict the actual duration of the relationship. Such questions could be incorporated into the volunteer screening and assessment process to enable mentoring program staff to address training issues and make more informed enrollment and matching decisions.