A quasi-experimental retrospective evaluation of a prison policy change.

Interrupted time-series design and quasi-experimental logic were employed to assess the behavioral effects of a prison disciplinary intervention on a population of approximately 2000 adult male inmates. The uncontrolled nature of the prison environment prevented experimental analysis of intervention effects. The quasi-experimental methodology applied in this study exemplifies how data, otherwise lost, can be retrieved. In addition, the process by which qualified conclusions can be drawn from the retrieved data is demonstrated.