Proximal peer-level effects of a small-group selected prevention on aggression in elementary school children: an investigation of the peer contagion hypothesis.

Examined peer contagion in small group, selected prevention programming over one school year. Participants were boys and girls in grades 3 (46 groups, 285 students) and 6 (36 groups, 219 students) attending school in low-resource, inner city communities or moderate resource urban communities. Three-level hierarchical linear modeling (observations within individuals within groups) indicated that individual change in aggression over time related to the average aggression of others in the intervention group. The individual child was "pulled" toward peers' mean level of aggression; so the intervention appeared to reduce aggression for those high on aggression, and to make those low on aggression more aggressive. Effects appeared to be magnified in either direction when the child was more discrepant from his or her peers. From these results we derive a principle of "discrepancy-proportional peer-influence" for small group intervention, and discuss the implications of this for aggregating aggressive children in small group programs.