Changing Attitudes and Motivation in Parolees (CHAMPS) Evaluation


With 750,000 people released from prisons each year, there is a pressing need for rigorous evidence on the effectiveness of reentry strategies. Returning citizens face a range of challenges to successful reentry into the community, including low levels of employment and substance use problems, all of which impact recidivism rates. Although the issue of reentry among individuals formerly incarcerated has attracted substantial attention and funding in recent years, very little is known about the components of effective reentry programs. It is unknown what in-prison activities are best able to prepare individuals for the return to the community, what works best to stabilize people after they are released, and what long-term efforts are needed to help people become productive citizens. One avenue for affecting outcomes may be through parole, but little is known about what parole practices are most effective for whom and how additional services aimed at improving returning citizens’ cognitive and behavioral functioning can complement the work of parole officers.

The Bureau of Justice Assistance, in a collaborative effort with the National Institute of Corrections and the National Institute of Justice, is implementing an innovative parole-based intervention with a well-known cognitive behavioral therapy program as part of a Demonstration Field Experiment on reentry of the formerly incarcerated, known as Changing Attitudes and Motivation in Parolees (CHAMPS).

The National Institute of Justice selected MDRC and its partner, George Mason University, to conduct a multisite random assignment study to test this reentry model intended to: (1) improve individuals’ motivation to change; (2) address cognitive and behavioral functioning regarding crime-prone thoughts and behaviors; and (3) address core factors that affect individuals’ performance while under community supervision following release from prison.

Agenda, Scope, and Goals

The overarching goal of this study was to test the effectiveness of parole supervision strategies and a targeted cognitive behavioral intervention to improve outcomes and reduce recidivism for individuals on parole. The evaluation used a random assignment research design to measure the impact of the following interventions: 

  • The National Institute of Corrections’ Next Generation relationship and desistence model, which is designed to improve the techniques used by parole officers in supervising and interacting with clients. The model stems from a relationship theory as well as risk-need-responsivity framework where services are recommended to address risk factors that may create problem behaviors for individuals on parole. Selected parole officers delivered this model after receiving training.
  • Cognitive behavioral therapy consisting of Motivational Enhancement Therapy sessions followed by Thinking for a Change sessions. Treatment providers, not parole officers, delivered this intervention.

Design, Sites, and Data Sources

The CHAMPS evaluation targeted individuals who were recently released from prison to parole supervision. The evaluation used a random assignment design to compare the efficacy of three approaches with parolees: (1) parole officers use standard supervision practices (the “business as usual” control group); (2) parole officers use core correctional practices with a focus on desistance (the Next Generation model); and (3) parole officers use the Next Generation model and parolees participate in Motivational Enhancement Therapy-Thinking for a Change (MET-T4C), a cognitive behavioral program for individuals formerly incarcerated offered by service providers.  

The three sites participating in the evaluation were: Dallas, Texas; Des Moines, Iowa; and Denver, Colorado.

Outcome data for all study groups was collected from sample member interviews and administrative records. Administrative records data were collected from each of the state agencies to measure criminal history at baseline and key outcomes over time, including arrest, conviction, reincarceration, and parole violations. We collected data from each of the sites’ management information systems in order to measure treatment session attendance, employment, discharge status, parole contacts, drug testing, and other events and processes as available. The survey, administered up to one year after study enrollment, focused on key outcomes that cannot be measured with the administrative records, including receipt and dosage of reentry services, educational and employment services, and substance use and mental health services; housing stability and services; and other outcomes.