Description of original award (Fiscal Year 2014, $299,808)
This program furthers the Department's mission by enhancing what is understood about mentoring as a prevention and intervention strategy for youth who are at high risk of involvement or involved in the juvenile justice system.
Research demonstrates that youth at risk for poor social outcomes may benefit from interventions that include mentor guidance, but there is lack of systematic, long-term research on mentoring programs with juvenile offenders. The mechanisms by which mentorship promotes positive long-term gains and the specific role of peer versus adult mentors are understudied.
The proposed project will extend an ongoing prospective, longitudinal study on influence of personal and environmental factors on youth recidivism. In addition, it will propose expanded analyses to address knowledge gaps concerning mentor/mentee relationships within the juvenile justice system. The proposed study will use a quasi-experimental design as in the ongoing study. Participants will be 75 male juvenile offenders and their mentors currently enrolled in the ongoing study. Youth participants are representative of the population in the underserved areas where they live, and comprise approximately 65% Hispanic, and 35% Black.
Although the overall study is longitudinal, a number of the new hypotheses proposed are best addressed as cross-sectional questions. In these, the statistical methods used will be logistic regression and general linear modeling. For the longitudinal analysis, a joint modeling approach, including survival-type analyses when appropriate, will be used since the longitudinal prosocial outcomes of the juvenile offenders, the recurrent technical violations of probation, and the subsequent re-offenses are expected to be correlated. Such an approach will maximize statistical power for their hypothesis and statistical efficiency in the estimation of effect sizes.
The specific questions they ask are: (1) Do juvenile offender participants in a peer-based community mentoring program show decreased recidivism and prosocial gains, and are these outcomes related to participants readiness-for-change at 24-months post-detention? (2) What is the degree of mentor/mentee engagement following probation and does mentor/mentee engagement predict prosocial outcomes and recidivism? (3) Do specific temperament and personality characteristics of mentors predict mentor/mentee engagement? (4) Do specific temperament and personality characteristics of mentors predict recidivism and prosocial gains of their mentees? (5) Is youth personality/temperament, mentor personality/temperament, or youth/mentor alliance the strongest predictor of mentor/mentee engagement? (6) What youth cognitive predictor variables are most related to mentor/mentee engagement and mentor-youth alliance?
The knowledge gained will be disseminated by professional seminars, including those with Juvenile Justice, conferences, and peer-reviewed publications. It is their belief that such knowledge has the potential to improve the potency of mentorship programs and allow for a better use of sparse resources.