Absence of conflict of interest.
Citation
Nally, J., Lockwood, S., Knutson, K., & Ho, T. (2012). An evaluation of the effect of correctional education programs on post-release recidivism and employment: An empirical study in Indiana. The Journal of Correctional Education, 63(1), 69-89.
Highlights
- The study examined the effect of Indiana’s correctional education program on individuals’ employment, earnings, and recidivism outcomes after release from prison.
- The authors used administrative data from the Indiana Department of Corrections (IDOC) and Indiana Department of Workforce Development to compare individuals who participated in correctional education programs while in custody with those who did not.
- The study found that participation in correctional education programs was associated with a lower likelihood of recidivism.
- The quality of causal evidence presented in this study is low because the authors did not ensure that the groups being compared were similar before the intervention.
Intervention Examined
Indiana’s Correctional Education Programs
Features of the Intervention
Intervention group members received U.S. Department of Education funding to attend the correctional education program while incarcerated. Indiana’s correctional education programs included postsecondary, job-oriented certificate training programs to increase inmates’ employability and reduce recidivism post-release.
Features of the Study
The authors used data obtained from IDOC and Indiana Department of Workforce Development to examine employment, earnings, and recidivism outcomes of individuals who were released from IDOC during 2002 to 2009. The intervention group consisted of 1,077 inmates who received federal funding to attend correctional education programs before release and were released from IDOC custody during the period 2002–2009. The comparison group included 1,078 individuals who did not receive federal funding for education programs and were randomly selected from a larger pool of sample members released from IDOC in 2005. Most sample members were men (about 85 percent) and Caucasian (65 percent), between the ages of 20 and 39 (77 percent), and had a high school diploma or GED (80 percent).
The study examined the effect of correctional education on participants’ employment, earnings, and recidivism post-release. Employment and earnings were measured in the first quarter of 2008 through the second quarter of 2009. To compare outcomes, the authors compared the raw, unadjusted percentages in different categories (for example, employed or not employed, percentage employed in cumulative quarters, categorical earnings in quarters). The impacts on recidivism were estimated using a statistical model that controlled for race, gender, age, education level, and employment status.
Findings
Employment
- Individuals who participated in correctional education programs were less likely to be employed since release than individuals who did not attend correctional education programs while incarcerated (the comparison group). This outcome was not tested statistically.
Earnings
- Among individuals who were employed, those in the comparison group were less likely to earn more than $1,000 in each quarter. This outcome was not tested statistically.
Recidivism
- Individuals who participated in correctional education programs were significantly less likely to recidivate than individuals who did not attend correctional education programs while incarcerated (the comparison group).
Considerations for Interpreting the Findings
Time of release from prison differed between the intervention and comparison groups and this could have affected the observed outcomes. Intervention group members were released from prison between 2002 and 2009, while comparison group members were released in 2005. Thus, an intervention group member released before 2005 had a greater period of time to reoffend. Alternately, an intervention group member released in 2007–2009 would have re-entered society during a recession, which may have negatively impacted his or her labor market success.
The analysis also did not account for differences between the groups before the intervention, for example, their baseline employment status (for employment outcomes) or criminal history (for recidivism outcomes). Preexisting differences between the groups—and not the intervention— could explain the observed differences in outcomes.
Causal Evidence Rating
The quality of causal evidence presented in this study is low because the authors did not ensure that the groups being compared were similar before the intervention. In addition, the time in which the sample members were released from prison differed between the intervention and comparison groups. Thus, it is a confounding factor. This means we are not confident that the estimated effects are attributable to correctional education programs; other factors are likely to have contributed.