The purpose of this journal article is twofold. First, it provides a discussion of the use of recidivism as a measure of effectiveness of criminal justice interventions, and, secondly, there is discussion of an evaluation of a juvenile post-release support program. The article argues that there has been a significant growth in recidivist studies, particularly as a measure of effectiveness. However there has been less and less attention placed on the limitations of measures of recidivism, or the nature of extraneous factors that influence re-offending. We use an evaluation study we conducted of the Post Release Support Program (PRSP) for juvenile offenders in New South Wales to explore these issues further. One of the interesting points to the study was that, while the statistical results on re-offending were not conclusive, the qualitative interviews among staff and offenders were overwhelmingly positive about the program. For us this raises the question: what value do we place on recidivism in evaluating a program when qualitative outcome information appears more conclusive?