Social skills training with young adolescents: group and individual approaches in a school setting

There has been relatively little research on social skills training with young adolescents. Yet adolescence is a key stage in the life cycle, and may well be important for the development of social skills. There is evidence that difficulty with the peer group early in life may be a risk factor for a range of personal problems at a later stage in the life cycle. The research programme sought to evaluate the use of group social skills training with young adolescents in a school setting, and to identify the most useful component. In the first experiment nine young adolescents aged 12-13 were identified as having social skills difficulty. The intervention took the form of a group programme of ten social skills training sessions with a strong individualised component. On measures of social competence and self-report significant gains were made which were maintained at six month follow-up. On a measure of sociometric status there was significant gain by the time of follow-up. A non-equivalent problem-free no-treatment control group (n = 10) of like age made no significant gains on any of the variables measured. There were significant differences between the control group and the treatment group in social competence and sociometric status prior to treatment, but none post treatment nor at follow-up. Using MANOVA a main effect for time was found for the combined measures of social competence and self-report. In the second experiment 41 young adolescents aged 12-13 were allocated to an individualised training (IT) Group (n = 14), a standardised training (ST) Group (n = 13) and a no treatment control group (ii = 14). Therapist effects were controlled by crossing over therapists, so that equal numbers of each group were trained by each pair of therapists. The programme ran over 2 years. A group comparison design taken together with single case designs provided evidence for the effectiveness of the interventions and for the relative superiority of individua].ised training over standardised training. Dependent measures included a self-report questionnaire, a sociometric questionnaire, a behavioural interview schedule, a role-taking test, a social assertiveness test, and a clinical change rating. There was no main effect for type of problem (rejected vs neglected) nor for subject gender. Data relating to therapist performance suggested that therapist effects might make a substantial difference to outcome. It was concluded from the two experiments that the treatments were generally effective and able to produce generalisation, that the individualised approach worked rather better than the standardised approach, and that it was important to control for therapist effects.