The meta-analysis included studies that evaluated an educational, vocational, or work program for convicted adults or persons identified by the criminal justice system and placed in a prison or a jail or diverted to another corrections-based program such as probation. Measures of post-program recidivism included arrest, conviction, self-report, technical violation, or incarceration. Of 53 program-comparison contrasts identified in the literature, 33 were coded for meta-analysis. The evidence from the 33 evaluations showed participants were employed at a higher rate and recidivated at a lower rate than non-participants. Assuming a 50 percent recidivism rate for non-participants, participants recidivated on average at a rate of 39 percent. The reduction in re-offending appeared to be greater for educational programs than for work programs. The generally weak methodological nature of the studies, however, prevented attributing observed effects to program activities. Future evaluative research in this area could be strengthened by incorporating theoretical links between program activities and future criminal involvement and through designs that control for self-selection bias beyond basic demographic differences. 70 references, 1 note, and 6 tables