Although the association between unemployment and offending is well established, relatively little is known about the impact of vocational education and training programs on re-offending, with much of the previous work in this area failing to control for, or correct, selection bias. This article reports the findings of a systematic review, which considers the findings of only those studies that have used experimental or quasi-experimental designs to evaluate vocational training and employment program outcomes for adult offenders. The analysis identifies key features, based on these studies, of those programs associated with the best outcomes and recommends selection criteria for those who are most likely to benefit from prison vocational education and training.