NCJ Number
89870
Date Published
1982
Length
86 pages
Annotation
Papers consider the Federal approach to presumptive parole dates, the effects of presumptive parole dates on inmate behavior, a comparison of recidivism between males and females, the clinical prediction of violent behavior, and employment and recidivism for Federal releases in community treatment centers.
Abstract
The first paper presents an overview of the goals and structure of the U.S. Parole Commission's approach to parole release decisionmaking, and the relationship of this approach to the concerns of equity and determinacy is highlighted. What has emerged in the Federal parole system is a conceptually simple system which provides for the early setting of a tentative date of release based on factors known at the time of commitment. The primary finding of the second paper is that the inmates given presumptive parole dates did not commit disciplinary infractions any more frequently than those in a control group, nor were differences observed in the seriousness of the infractions recorded. The finding of the third paper is that, although limited by the small size of the female sample, no substantial difference in recidivism rate exists between male and female Federal releasees once control has been exercised for salient factor score. New arrests within a 2-year followup period were used for the criterion. The fourth presentation examines common clinical errors in predicting violent behavior, research on clinical prediction, and statistical approaches to improving clinical prediction. It is concluded that the improvement of prediction accuracy by incorporating statistical information can best be done by making the base rates of violent behavior a prime consideration. Extensive references accompany this paper. The concluding paper found that release through a community treatment center significantly improves postrelease employment success of both white and minority parolees, with treatment center release found to be significantly associated with reduced recidivism for minority offenders but not for white offenders. The papers include relevant tabular data and references.