According to the risk principle, correctional programs are more likely to have an effect when they are delivered to high-risk groups of offenders. The current study probed this risk principle by comparing data on 7,306 offenders who successfully completed programming at 1 of 53 community-based residential post-release programs in Ohio with data on 5,801 offenders under post-release supervision who did not complete residential programming. Data under examination included demographic information, substance use history, criminal history, current offense information, and mental health. The analysis of program effectiveness controlled for risk and a risk-by-group interaction effect. Results of statistical analyses indicated significant differences in the effectiveness of residential programming by risk level of offender. Residential programs serving low- and moderate-risk offenders were associated with an increase in recidivism rates when compared to the control group. However, residential programs were effective with higher risk offenders, reducing recidivism rates among participants. The findings seriously question criminal justice policies that reserve residential programming for low-risk offenders. Future research should focus on investigating the relationships between program characteristics and program effectiveness. Tables, figures, footnotes, references