NCJ Number
178106
Journal
Sexual Abuse: A Journal of Research and Treatment Volume: 11 Issue: 3 Dated: July 1999 Pages: 183-193
Date Published
July 1999
Length
11 pages
Annotation
The value of randomized controlled trials in the evaluation of sex offender treatment has been questioned, and concern has been expressed that randomization fails to produce equivalent samples.
Abstract
The author points out that the failure of randomization to produce equivalent samples is inevitable when variables are distributed by chance and that lack of equivalence can be controlled by using tests of significance. A further uncriticized and inappropriate procedure in sex offender treatment evaluation involves separation of results for subjects who do not complete treatment from those who do. Despite a recommendation from the American Psychological Association, no attention has been paid to the consistent finding that no treatment is less effective than placebo psychological therapy. The significance of Type II errors is discussed, and the recommendation that within-treatment research be encouraged as an alternative to outcome research is criticized. Demonstrating a within-treatment response when that response is associated with a better outcomes does not necessarily mean the treatment is effective. Subjects with a good prognosis may be more able to demonstrate a within-treatment response. Further, nonrandomized matched samples do not adequately control all sample differences. The post hoc statistical reversal of a reported trend for sex offenders treated with relapse prevention to show a worse outcome than untreated offenders, in order to correct lack of equivalence for the two groups, is considered inappropriate. That relapse prevention is less effective than no treatment raises the possibility of a possible negative effect. The author concludes that the continued use of relapse prevention other than in randomly controlled evaluation studies is unethical. 32 references and 1 table