A Comparison of Recidivism Patterns of Pennsylvania's Community Correction Program: A Two-Year Survival Analysis

• In Event: Reducing Offender Recidivism

Abstract

Reducing recidivism is a crucial goal for public and for policy makers. Many studies have demonstrated that recidivism rates varied between offenders with different criminogenic characteristics, different follow-up periods, and different outcome measures. A few studies, however, investigated the difference in timing of recidivism between offenders under different type of intermediate sanction programs. This study explored the relative intervention effect of the Pennsylvania Community Correction Programs using survival analysis. The cases of 7,846 offenders were examined in a two-year follow-up period between 2006-2007. Re-offending is defined as days survived in the community without any involvement in illegal activities, including parole violations, arrests, and re-incarceration. Predictors includes participation in correctional treatment programs, socio-demographic characteristics, Level of Service Inventory-Revised (LSI-R) score, sex offender status, and committing county. Cox's proportional hazard analysis was employed to estimate the hazard rate of re-offending in a statistical model. The findings from this study suggested that the re-offending rates and survival curves were essentially different depending on treatment program participation, LSI-R score, sex, and age. The implication of the findings was discussed in the study.

Authors

• Hyejin Kim, University of Cincinnati

• Edward Latessa, University of Cincinnati