Association for Public Policy Analysis & Management

Existing research into mentoring interventions has demonstrated positive impacts on social, emotional, behavioral, academic and criminal justice outcomes for youth and young adults, with greater gains in programs with higher intensity and quality of relationships between mentors and mentees. However, there has been limited evaluation of emerging promising strategies designed to support mentor-mentee relationship development, such as “credible messenger” mentoring and group mentoring. Credible messenger mentoring utilizes mentors with background similar to those of their mentees, an approach that is theorized to expedite and enhance relationship development; in mentoring programs that serve justice system-involved young people, this approach means employing mentors who are themselves formerly justice system-involved. Group mentoring models expose mentees to multiple mentors with opportunities for more organic matching and relationship development. Furthermore, group mentoring offers expanded opportunities for the establishment of social ties with peers.

This presentation will discuss findings from new research on Arches Transformative Mentoring (Arches), a New York City mentoring program for young adult probation clients ages 16 to 24 years. Arches offers group mentoring delivered by credible messengers, combined with motivational interviewing and an evidence-based interactive journaling curriculum centered on cognitive behavioral principles. Launched in 2012 as a component of the New York City Young Men’s Initiative, and implemented in partnership with the New York City Center for Economic Opportunity and the New York City Department of Probation (DOP), Arches supports participants to improve problem solving and social skills, and to achieve the attitudinal and behavioral change necessary to avoid future criminal activity and reengage with education and the workforce.

The Urban Institute and DOP have collaborated to conduct a mixed-method implementation evaluation and quasi-experimental impact evaluation of Arches. Drawing upon multiple extant data sources, including probation client casefiles maintained by DOP, program data maintained by Arches provider organizations, and criminal record data maintained by the New York State Division of Criminal Justice Services, the research compares 279 Arches program participants with a matched comparison group of 682 young adult probation clients. This analysis finds that one- and two-year felony re-conviction rates are over 50 percent lower among Arches participants than among comparison group members. The presentation will also include findings from a broad qualitative examination of the program model, highlighting best practices in program implementation and operations.