NCJ Number
194873
Journal
Sexual Abuse: A Journal of Research and Treatment Volume: 14 Issue: 2 Dated: April 2002 Pages: 139-153
Date Published
2002
Length
15 pages
Annotation
This paper reports on the creation and analysis of a framework for dynamic risk assessment in the prediction of sexual offender recidivism.
Abstract
The author conducted two studies in this article. The first study compared male sexual offenders who re-offended to offenders who do not re-offend. The second study drew on the results of the first study to create an overall “deviance” category that should aid in the prediction of recidivism among sexual offenders. The first study divided 158 adult male sex offenders into 2 groups: those with a previous conviction for similar crimes and those who were serving sentences on their first sexual offense conviction. The purpose of the study was to test whether psychometric indicators for distorted attitudes, socioaffective functioning, and self-management were useful in predicting recidivism among sexual offenders. The results verified that these indicators were indeed useful: the “repeaters” group showed more distorted attitudes, worse socioaffective functioning, and poorer self-management than did the group serving sentences for their first sexual offense. The second study combined these three psychometric indicators into a “deviance” classification and tested whether this classification offered predictive data concerning sexual recidivism. Information about 117 male sexual offenders currently serving prison sentences was gathered from a database and used to rank offenders as high, medium, or low on deviance. The findings showed that the deviance classification was reliable in predicting recidivism among sexual offenders; 15 percent of those offenders who ranked as high on deviance were reconvicted of sexual offenses while only 5 percent who ranked as medium were reconvicted. None of the offenders who earned a low ranking on deviance were reconvicted of sexual crimes. 1 Table, references