A Long-Term Follow-Up Evaluation of an Employment Assistance Reentry Program

Abstract

This study uses a quasi-experimental design to evaluate the effectiveness of an employment assistance program implemented in Minnesota prisons by examining recidivism and post-release employment outcomes of 2,814 individuals released between 2009 and 2018. Observable selection bias was reduced by using propensity score matching to create similar treatment and comparison groups. Results of Cox regression analyses showed that completing EMPLOY reduced the hazard ratio for recidivism by 27–66%. The findings further showed that those who completed EMPLOY were more likely to gain post-release employment within two years after release from prison and worked more hours, had higher hourly wages, and earned more total wages during the two-year follow-up period.