An empirical study of the efficacy of social skills training (SST) with young male offenders in custody was conducted. Two forms of SST, a traditional approach and one with some elements of cognitive behavior modification, were compared with attention-placebo controls (unstructured discussion), a no-training group drawn from those judged suitable for SST, and an unreferred control group drawn from the remainder of the borstal population. A variety of pre- and post-training measures were taken, including a wide range of personality measures, staff ratings of institutional behavior, frequency of institutional discipline reports, and ratings of role-play performance. The two behavioral training groups improved significantly on the behavioral role-play assessment, while the controls did not. Institutional discipline reports also improved for the two behavioral training groups, although a similar improvement also was found for the attention-placebo controls. No other measures differentiated between the experimental groups; nor could any distinctions be drawn between the two forms of behavioral training on any measures. It is suggested that the improvement in behavioral ratings may have been a function of familiarity with role-play, and, therefore, the only positive training effect — improvement in institutional discipline — can be attributed to an attention component. A number of explanations are offered for the lack of generalization of the training effects: It is further suggested that the cognitive training may have been particularly sensitive to a number of inhibiting factors. Finally, several suggestions for future experimentation and training strategies are made.
Read full abstract