The planned study will randomly assign aggressive children to one of two versions of the Coping Power child component. The two versions of Coping Power will either deliver the child component of the program in the usual small group format (Group Coping Power: GCP) or in a newly-developed individual format (Individual Coping Power: ICP). By providing a direct comparison of two different formats of the same intervention, the planned study's design will fill a critical gap in our current understanding of the relative effectiveness of group vs. individual programs. Further, this study will allow for examination of the specific factors that influence relative effectiveness of these two formats, important information with broad implications for program development, training of clinicians, and intervention implementation.

Specific Aim 1: The study will test the hypothesis that the Coping Power intervention will produce larger effect sizes when delivered in an individual format in comparison to a group format. Although there are advantages of both formats, pilot data suggests that the group format may be diminishing the strength of outcome effects of intervention in comparison to the same intervention delivered in individual format. This pilot data is consistent with some prior findings, but a direct randomized comparison of children assigned to group versus individual formats has not been conducted, despite the critically important conceptual, clinical, and policy implications.

Hypothesis 1-1: It is hypothesized that ICP will produce greater reductions in behavior outcomes including substance use, externalizing behavior problems, and delinquency at a 1-year follow-up, in comparison to GCP.

Hypothesis 1-2: it is hypothesized that the ICP condition will produce greater improvements in children's social competence, which is directly targeted by the intervention, in comparison to GCP.

Specific Aim 2: Individual and group variation in effect sizes will be an outcome of youth behavior in the group (i.e., deviancy training) and group leader behavior management skill. We see youth behavior to be highly influenced by group leader management practices. We understand that some groups and/or individual children present challenges to even the most competent group leaders, and therefore, variation will be observable and meaningful. The design of the study allows for the testing of both group level and individual effects, and linkage of these effects to specific behaviors. Such information will provide an empirical basis for clinical training for group interventions with youth in general and Coping Power in particular.

Hypothesis 2-1: It is hypothesized that peer escalation in the GCP condition will predict worse outcomes, and that the level of group deviance in the GCP condition will moderate the effectiveness of the GCP condition, with better outcome effects for the groups with the highest initial screening scores.

Hypothesis 2-2: It is hypothesized that group interventions will be compromised by individual children's reactions to the interpersonal dynamics of the groups, such as inadvertent attention to deviant behavior and talk provided by group members and/or the group leader.

Hypothesis 2-3: It is hypothesized that level of positive group leader behaviors (directing attention to rules, correcting behavior, providing praise for compliance, introduction and review of activities, clear directions) will moderate the effectiveness of the GCP condition.

Specific Aim 3: Variability in outcome scores will differ between conditions. Hypothesis 3-1: It is hypothesized that there will be greater variability in the outcome scores of children in the GCP condition than in those of children in the ICP condition.

Specific Aim 4: Child characteristics will be examined as potential moderators of intervention effects.

Hypothesis 4-1: It is hypothesized that youth with low effortful control will be most vulnerable to deviancy effects in group interventions and therefore will show lower effect sizes than youth higher in effortful control at baseline. Thus we expect effortful control to function as a moderator of group intervention effectiveness, but not individual intervention effectiveness.

Research Question 1: In addition, we will investigate the possibility that characteristics of the youth's decision-making (impulsive decision-making; outcome expectations), affective arousal (callous-unemotional traits; low physiological arousal in response to negative consequences), temperament and behavior characteristics (baseline severity of aggressive behavior); perceived and actual peer reactions (perceived peer competence; peer rejection; peer victimization; deviant peers) and demographic characteristics (sex; age; race) will moderate the effectiveness of both interventions.

Inclusion Criteria:

• top 25% in teacher-rated aggression and above low-aggressive range in parent-rated aggression

Exclusion Criteria:

-