[Show abstract][Hide abstract] ABSTRACT: Although the prevention of recidivism in France is currently undergoing numerous transformations, the evaluation of the risk of recidivism of individuals under the justice system is a major issue. The absence of structure and standardization in current evaluation procedures raises doubts regarding their validity and underlines the necessity to determine evaluation methods that are clinically and scientifically validated. The professionals and their administration would benefit largely from a range of strategies that would aid in executing the most just, transparent and coherent evaluation possible. This article will trace the evolution of the evaluation of the risk of recidivism, the principal strategies and methods, the most common risk factor families used, as well as illustrations of two typical strategies of evolution of the risk of recidivism; integrated evaluation with the LS/CMI and evaluation based on structured professional judgment guided by the HCR-20.
[Show abstract][Hide abstract] ABSTRACT: Clinicians have at their disposal a variety of instruments with which to evaluate the risk presented by sex offenders. These measures yield similar predictive potency, and combining them does not appear to enhance prediction. The present study examined whether classification and regression tree analysis could identify new combinations of predictors that would improve predictive validity. Items from seven actuarial instruments were used to construct classification trees. Overall, classification trees achieved slightly higher predictive accuracy than did actuarial instruments. In addition, these analyses highlight the heterogeneity among sex offenders. Despite this improvement, one should consider the incorporation of other predictors into the instruments—including dynamic factors, protective factors, and measures with strong theoretical justification.
[Show abstract][Hide abstract] ABSTRACT: Clinicians have access to several risk assessment instruments to evaluate the risk or recidivism in sexual offenders. Nevertheless, we seem to have attained a ceiling in the predictive validity of these instruments with the traditional techniques of items agglomeration. In this study, we offer a different combination of predictors with the classification and regression trees, and it, by taking into account the type of sexual offenders. The classification trees are constructed from predictors contained in seven actuarial instruments (VRAG, SORAG, RRASOR, STATIC-99, STATIC-2002, RM2000, MnSOST-R). In general, the classification trees have a higher predictive accuracy than the actuarial instruments and point out that it's not the same predictors that should be considered according to the type of offenders and the type of recidivism. Furthermore, classification trees identify correctly more recidivists than the best actuarial tool. In spite of the contribution of this approach, other types of predictors should also be considered to augment predictive accuracy: dynamic predictors, protective predictors as well as measurements based on theories like those on attachment styles (Marshall, D. R., Barbaree, H. E., 1990. An integrated theory of the etiology of sexual offending. In: Marshall, W. L., Laws, D. R. L., Barbaree, H.E. (Eds.), Handbook of sexual assault. New York: Plenum Press, pp. 257-275.) and cognitive distortions (Ward, T., Keenan, T., Hudson, S. M., 2000. Understanding cognitive, affective, and intimacy deficits in sexual offenders: a developmental perspective. Aggression and Violent Behavior, 5, 41–62.).
[Show abstract][Hide abstract] ABSTRACT: AUTHORS' NOTE: This article is a part of the master's thesis of the first author and a part of a research project directed by Raymond Knight and David Thornton. The authors would like to thank all the research assis-tants who worked in that project. This research project was supported by grants from the National Criminal Justice Reference Service (217618) awarded to the third author and David Thornton. Correspondence concern-ing this article should be addressed to Geneviève Parent, Numerous instruments are available to clinicians for evaluating sex offenders' reoffense risk. Although they have demon-strated effectiveness in predicting recidivism significantly better than unstructured clinical evaluation, little is known about their predictive accuracy in subgroups of sexual offenders or in the long term. This study was undertaken to evaluate the predictive accuracy of nine instruments (VRAG, SORAG, RRASOR, Static-99, Static-2002, RM2000, MnSOST-R, SVR-20, PCL-R) among three groups of sexual offenders across a 15-year follow-up period. The results indicate that these instruments yielded marginal to modest predictive accuracy for sexual recidivism. A more detailed study of aggressor subgroups indicated that in both the short and the long term, these instruments were more effective at predicting the sexual recidivism of child molesters and the violent and nonviolent recidivism of rapists. Finally, although mixed offenders sexually reoffend more often and more rapidly than do rapists or child molesters, firm conclusions cannot be drawn because of the small number of mixed offenders in the sample.