Principal Investigator: John E. Lochman and John D. Coie

Level of Intervention: Selected

Target Population: Aggressive, rejected, and non-aggressive rejected fourth grade, African-American students.

References: Lochman, Coie, Underwood, & Terry (1993)

Theory (Risk and Protective Factors Targeted):
This program is based on the significant amount of research that relates children’s aggressive behavior to adjustment difficulties in the peer context and specifically to peer rejection. Rejected children exhibit a variety of social skill deficits that contribute to active disliked by peers. Research suggests that aggressive-rejected children have more significant social-cognitive deficits and are at greater risk for poor adolescent outcomes. Social skills training programs have improved the social acceptance of rejected children and cognitive-behavioral programs have shown success in improving the behavioral outcomes of conduct problem children.

Description of Intervention:
The program consists of 4 components: 1) Social problem solving (7 sessions), 2) positive play training (9 sessions), 3) group-entry skill training (14 sessions), and 4) anger-control (4 sessions). In the first component the children learned problem solving steps including identification of a problem and goals of the situation, how to inhibit impulsive behavior, and how to generate possible solutions. The second component focused on helping the children gain the skills necessary to play effectively with peers (e.g. communication, negotiation, cooperation). In the third component the children were taught how to join a group of peers including how to identify group leaders and match their behavior to the group. In the final component, the children learned how to reduce their impulsive behavior through identification, the use of self-statements, and discussion of competition in interpersonal situations. The training took place over 26, 30-minute individual sessions and 8 small group sessions. The sessions took place twice weekly between October and April of the school year. The order of the sessions was altered slightly for the aggressive-rejected students. They received the anger-control unit immediately after the problem solving unit.

Research Subjects:
Twenty-eight third-grade classrooms (602 students) in an inner-city school system were screened with peer-nomination techniques. Positive ("liked most"), negative ("liked least"), and aggressive ("starts fights") nominations were gathered on each child and a social preference score was constructed for each child. Rejected children had a social preference score less than –1.0SD, standardized liked-most scores less than zero, and standardized liked-least scores greater than zero. Aggressive children had standardized nomination totals for "starts fights" greater than 1.0SD. A sample of 86 socially rejected, African-American boys was identified. From this group, 52 (27 boys, 25 girls) students who remained in school for the entire fourth grade year and who had parental consent were used in the final sample.

Prior to the intervention, 13 students were placed in the Aggressive-Rejected intervention (ARI) group, 33 in the Rejected-Only (RI), 11 in the Aggressive-Rejected control (ARC) group, and 29 in the Rejected-Only control (RC) group. Of the 52 subjects who completed the post-test assessment, 9 were in the ARI group, 17 were in the RI group, 9 were in the ARC group, and 17 were in the RC group. At the one-year follow-up assessment the sample consisted of 44 subjects: 7 in the ARI group, 17 in the RI group, 6 in the ARC group, and 14 in the RC group.

Research Design:
Children were randomly assigned to intervention (ARI or RI) or control (ARC or RC) groups on the basis of their screening scores.

Outcomes:
Used ANCOVAs with pre-intervention scores as covariates.

Post-Test
Students in the ARI group were rated by teachers as significantly less aggressive (p<.04) on the Teacher Behavior Checklist (Coie & Dodge, 1988) and less rejected (p<.01) compared to the ARC group. The ARI group had significantly more positive social acceptance according to peers than the ARC group (P<.04). There was a significant main effect for Intervention status on children’s ratings of their self-worth (Perceived Competence Scale for Children, Harter, 1982). Control children reported higher levels of self-worth compared to intervention children.

Follow-up (1 year)
According to teacher ratings, the ARI group exhibited significantly less aggression (p<.03) and more prosocial behavior (p<.05) than the ARC group. A significant Intervention x Academic problem (academically adequate vs. academically inadequate) interaction effect for the Aggression score (p<.01) and the Prosocial score (p<.01) which was attributed to the fact that the academically adequate intervention group had lower aggression and higher prosocial scores than the control group. There was no longer a main effect on child self-reported self-worth.

Strengths and Limitations:
The Social Relations Intervention Program targeted the social cognitive risk factors associated with children who exhibit disruptive behavior problems by providing intervention on the individual child level. Based on teacher ratings, the findings indicated that the program was most effectiveness in reducing aggression and improving prosocial behavior in aggressive-rejected children at post-test. These results were maintained at one-year follow-up. The program was also successful in altering peer ratings of the children’s behavior at post-test though these results were not maintained over time. The design of this study was strong (i.e. randomized clinical trial) and it included both boys and girls which is unusual. Unfortunately, due to the small sample size and the reliance on an entirely African-American sample, the authors were unable to examine gender differences and the findings are somewhat limited in their generalizability. The authors noted that the staff who administered the program were trained for 2 weeks prior to starting the intervention and participated in weekly supervision meetings. However, there were no measures used to ensure program fidelity and dosage was not discussed. These types of measures should have been included given that the authors made some modifications to a previous program.