The present investigation extended the methodology to investigate the effects of viewed aggression on groups of two and four children. The present study employed a 2 x 2 x 3 factorial design, using as independent variables sex of the child, group size (two or four children), and cartoon condition (aggressive cartoon, nonaggressive cartoon, and no cartoon). Forty-eight boys and 48 girls of kindergarten age were randomly assigned to the experimental groups. A factor analysis of the dependent measures identified four factors, accounting for 85 percent of the total variance, each of which responded differently to the independent variables. The finding that (a) transgressive-aggressive was not affected by viewing aggressive cartoons, and (b) that normative-aggression loads on the same factor as normative-play, rejects the assumption that aggression is a single class of behavior and rejects the hypothesis that the effect of viewing aggressive models reduce inhibitions against aggression. The results are interpreted in terms of factors affecting levels of activity. (Author)