the older children. This finding suggests the need to target programs to children's different developmental needs.
More targeted social competence interventions have been designed for populations of children with risk factors for later disorders, such as peer rejection or early aggressive behavior. Bierman's (1986) program, *Social Skills Training, examined the impact of a social skills training program that involved role playing naturalistic interactions between a target child and peer partners. This selective intervention was targeted at 11-year-olds who had low peer acceptance and were deficient in conversation skills. Children were randomly paired with two same-gender peers and assigned to one of two experimental conditions: social skills training or peer experience. On short-term follow-up, the social skills training showed an overall positive effect on conversational skill acquisition and peer response.
Lochman and co-workers tested an expanded social competence curriculum, the *Social Relations Intervention Program, with a sample of 86 African-American fourth-grade children identified through sociometric ratings as the most socially rejected and physically aggressive in their classrooms and then randomly assigned to intervention and control groups (Lochman, Coie, Underwood, and Terry, in press). The indicated intervention, based on earlier quasi-experimental work (Lochman and Curry, 1986), consisted of positive social skills training to promote prosocial behaviors and cognitive-behavioral training to decrease anti-social reflexive responses and foster adaptive social problem solving. The curriculum included 26 30-minute individual sessions and 8 small-group sessions. It had four components: social problem-solving training, positive play training, group entry skills training, and training on coping with anger (including how to identify and reduce impulsive behavior and how to use self-talk to regulate behavior). Sessions were held twice weekly at school from early October to late April, for a total of 12 to 18 sessions, by a team of psychology graduate students and a university psychologist.
Small samples, nonequivalent comparison groups, and substantial attrition posed threats to this study, but teacher-rated aggression and rejection were significantly lower among a subgroup of rejected-aggressive children who had received the intervention than among rejected-aggressive children in schools not offered the intervention. One year later, for those remaining in the study, significantly lower teacher ratings of aggression and higher teacher ratings of prosocial behavior were observed in the aggressive-rejected subsample. The intervention appeared to have little effect on the behaviors or ratings of rejected children not also identified as aggressive.