Additional Resources:

Program Profile: Arches Transformative Mentoring Program

Evidence Rating: No Effects - One study No Effects - One study

Date: This profile was posted on November 05, 2019

Program Summary

This is a group mentoring program that seeks to reduce recidivism of youth on probation in New York City, using an interactive journaling curriculum based on cognitive-behavioral principles. The program is rated No Effects. Program participants showed a statistically significant reduction in felony reconvictions, compared with comparison group youth at 24 months; however, there were no statistically significant differences on arrests, felony arrests, or reconvictions.

Program Description

Program Goals/Target Population
Arches Transformative Mentoring Program is a community-based mentoring program for youth on probation in New York City. The program is designed to reduce youth’s recidivism and juvenile justice-system involvement and strengthen communities by improving outcomes for youth through positive attitudinal and behavioral changes. Arches serves youth, ages 16 through 24, who are on probation across the five boroughs of New York City, with an emphasis on those who are at a high risk of reoffending.

Program Activities
Youth are referred to the program by their probation officers. Participants meet in open and ongoing groups, which are led by a mentor, for 1 hour, twice a week for 6 months (48 sessions in total). New participants can join existing groups at specific time points (e.g., first week of each month). Mentors facilitate group sessions using an interactive journaling (IJ) curriculum, which was developed using cognitive-behavioral therapy and positive youth development frameworks. The curriculum focuses on helping participants make positive changes to their thoughts, feelings, and behaviors. Journaling usually occurs once a week and emphasizes topics associated with anger management and communication skills. Participants are required to complete four journals throughout the course of the program. The IJ curriculum is also supplemented by motivational interviewing techniques such as open-ended questions, reflective listening, and affirmations.

Mentors are expected to build rapport with youth, encourage them to build a support system for each other, teach them how to interact with law enforcement, fill gaps in services to reduce recidivism, and serve as a bridge among participants, probation officers, and families. Mentors are expected to be available to youth at all times for support, advice, and guidance. Mentors receive mandatory trainings on group facilitation techniques during program orientation and ongoing technical assistance with the IJ curriculum and group facilitation.

Participants receive cash stipends of up to $800 for attending mentoring sessions, distributed through debit cards that are loaded on a biweekly basis. Participants also receive hot meals prior to each session and a MetroCard following each session.

Key Personnel
In addition to referring youth to the program, probation officers provide case management and coordinate referrals to different types of services. Program sites are staffed with three to four paid mentors. One serves as the full-time lead mentor and is responsible for managing mentorship at each site, and the others are part-time mentors. One full-time program coordinator organizes program activities, supervises mentors, and acts as a liaison with the Department of Probation.

Program Theory
The program’s IJ curriculum is rooted in cognitive theory, which posits that people’s automatic reactions influence their perceptions of situations more than the actual situations do (McGuire 1996). Cognitive-behavioral therapy addresses this through brief problem-oriented conversations with a therapist. In the Arches program, these conversations are led by mentors and driven by youth’s journaling assignments. Arches mentors are trained in motivation enhancement therapy techniques, which is a person-centered approach to shifting attitudes toward personal change, especially for youth who show little desire to change, despite understanding the impact of their negative behaviors (Miller et al. 1992). The Arches model also uses a positive youth development framework by focusing on youth’s positive traits and resilience rather than their problem behavior (Lerner et al. 2018).

Evaluation Outcomes

top border
Study 1
Lynch and colleagues (2018) found Arches Transformative Mentoring Program participants showed a statistically significant reduction in felony reconvictions, compared with comparison group youth at 24 months; however, there were no statistically significant differences on arrests, felony arrests, or reconvictions. Overall, the preponderance of evidence suggested the program did not have the intended effects on participants.

At 24 months, there were no statistically significant differences in the number of arrests between youth in the Arches program and youth in the comparison group.

Felony Arrests
At 24 months, there were no statistically significant differences in the number of felony arrests between youth in the Arches program and youth in the comparison group.

At 24 months, there were no statistically significant differences in reconviction rates between youth in the Arches program and youth in the comparison group.

Felony Reconvictions
At 24 months, felony reconvictions were 57 percent lower for youth in the Arches program, compared with youth in the comparison group. This difference was statistically significant.
bottom border

Evaluation Methodology

top border
Study 1
Lynch and colleagues (2018) conducted a quasi-experimental study to compare rearrests and reconviction rates between 279 Arches Transformative Mentoring Program participants and a matched comparison group of 682 New York City youth on probation.

Demographic and criminal history data were compiled from three sources: Department of Probation Connect, Caseload Explorer (the Department of Probation’s management system), and the New York State Division of Criminal Justice Services. Risk of recidivism was measured using the Level of Service Inventory–Revised (LSI–R) screening instrument (Andrews, Bonta, and Wormith 2006).

In the final sample, the average age of Arches participants was 19 years old and the average age of comparison participants was 20.1 years old. Participants were predominantly male (86.5 percent of the treatment group and 85.3 percent of the comparison group), and most identified as non-Hispanic Black (75.6 percent treatment; 56.7 percent comparison), followed by Hispanic (17.9 percent treatment; 26.1 percent comparison), non-Hispanic white (0.7 percent treatment; 6 percent comparison), other (4 percent treatment; 9.1 percent comparison), or their information was missing (1.8 percent treatment; 2.1 percent comparison). In terms of education levels, 73.5 percent of the treatment and 61.1 percent of the comparison participants had not completed high school, 17.9 percent of the treatment and 19.8 percent of the comparison participants were high school graduates or had a GED, 2.5 percent of the treatment and 7.5 percent of the comparison participants had some postsecondary education, and 6.1 percent of treatment participants and 11.6 percent of comparison participants had some education data missing.

Participants’ prior offenses were characterized as follows: most serious (35.1 percent treatment; 24.2 percent comparison), more serious (8.6 percent treatment; 12.9 percent comparison), less serious (32.3 percent treatment; 34 percent comparison), least serious (18.3 percent treatment; 21.6 percent comparison), and missing (5.7 percent treatment, 7.3 percent comparison). Participants’ risk levels were categorized as follows: low (6.8 percent treatment; 5.4 percent comparison), medium (52.3 percent treatment; 54.4 percent comparison, high/highest (22.9 percent treatment; 15.8 percent comparison), or risk score not completed (17.9 percent treatment; 24.3 percent comparison).

At intake, the sample of Arches participants was more likely to be black, less educated, less likely to be employed, and more likely to have committed serious crimes and be deemed high risk. compared with comparison group participants. These differences were accounted for using nearest-neighbor propensity score matching.

Data on arrests, felony arrests, reconvictions, and felony reconvictions were compared between groups using t-tests. Subgroup analyses were conducted for age, gender, and risk level.
bottom border


top border
Each site received $170,800 to support program activities (Lynch et al. 2018).
bottom border

Implementation Information

top border
A nonprofit organization, Community Connections for Youth, provides both training and technical assistance, including onsite observations with feedback, to all Arches sites. Training includes mandatory orientation and follow-up, in-service trainings. Four journals and writing supplies are required for each youth to complete the program. Debit cards are distributed to youth to be reloaded every 2 weeks in exchange for their program attendance, as recorded by staff on the Department of Probation website, for an amount up to $800. Sites also provide hot meals and MetroCards at each of the 48 meetings over 6 months.
bottom border

Other Information (Including Subgroup Findings)

top border
Lynch and colleagues (2018) found no differences in estimated program effects by gender or risk status on any of the four outcome measures (arrest, felony arrest, reconvictions, and felony reconvictions) at 24 months. In regard to age, estimated program effects were more favorable for younger participants. Compared with their matched counterparts, program participants 17 years of age and younger had a statistically significant lower likelihood of being arrested for a felony, having a reconviction, or having a felony reconviction at 24 months. There were few statistically significant differences in estimated program effects among older age groups, and those that were significant were not in the expected direction (i.e., Arches participants fared worse than the comparison group). Finally, a subgroup analysis of Arches participants found that youth who completed the program had better outcomes across all four outcomes, compared with non-completers at 24 months; however, this difference was statistically significant only for felony arrests.
bottom border

Evidence-Base (Studies Reviewed)

top border
These sources were used in the development of the program profile:

Study 1
Lynch, Mathew, Nan Marie Astone, Juan Collazos, Micaela Lipman, and Sino Esthappan. 2018. Arches Transformative Mentoring Program: An Implementation and Impact Evaluation in New York City. Washington, D.C.: Urban Institute.
bottom border

Additional References

top border
These sources were used in the development of the program profile:

Andrews, D.A., James Bonta, and J. Stephen Wormith. 2006. “The Recent Past and Near Future of Risk and/or Need Assessment.” Crime & Delinquency 52(1):7–27.

Lerner, Richard M., Jacqueline V. Lerner, Jason B. Almerigi, Christina Theokas, Erin Phelps, Strinunn Gestsdottir, Sophie Naudeau, et al. 2005. “Positive Youth Development, Participation in Community Youth Development Programs, and Community Contributions of Fifth-Grade Adolescents: Findings from the First Wave of the 4-H Study of Positive Youth Development.” Journal of Early Adolescence 25(1):17–71.

McGuire, James 1996. Cognitive Behavioral Approaches: An Introductory Course on Theory and Research. Liverpool, England: University of Liverpool.

Miller, William R., Allen Zweben, Carlo C. DiClemente, and Robert G. Rychtarik. 1992. Motivational Enhancement Therapy Manual: A Clinical Research Guide for Therapists Treating Individuals with Alcohol Abuse and Dependence. Rockville, Md.: National Institute on Alcohol Abuse and Alcoholism.
bottom border

Related Practices

top border
Following are practices that are related to this program:

This practice provides at-risk youth with positive and consistent adult or older peer contact to promote healthy development and functioning by reducing risk factors. The practice is rated Effective in reducing delinquency outcomes; and Promising in reducing the use of alcohol and drugs; improving school attendance, grades, academic achievement test scores, social skills and peer relationships.

Evidence Ratings for Outcomes:
Effective - One Meta-Analysis Crime & Delinquency - Multiple crime/offense types
Promising - More than one Meta-Analysis Drugs & Substance Abuse - Multiple substances
Promising - One Meta-Analysis Education - Multiple education outcomes
Promising - One Meta-Analysis Mental Health & Behavioral Health - Psychological functioning
bottom border

Program Snapshot

Age: 16 - 24

Gender: Both

Race/Ethnicity: Black, Hispanic, White, Other

Geography: Urban

Setting (Delivery): Other Community Setting

Program Type: Cognitive Behavioral Treatment, Diversion, Mentoring, Probation/Parole Services

Targeted Population: Young Offenders

Current Program Status: Active