This study examined high-aggressive and low-aggressive boys' ratings of the effectiveness of aggressive and assertive strategies for solving social problems involving hypothetical peers and actual peers. Subjects were 66 third-grade boys (11 groups of 6 boys each for a total of 22 high-aggressive, 22 low-aggressive, and 22 average aggressive boys) who participated in a series of 5 play sessions and interviews. After the fifth play session, each boy rated each of his five play partners' aggressiveness on a five-point scale. In the interviews, boys were read stories in which a social conflict with a hypothetical or actual peer was described. For each story, each boy evaluated whether he would achieve a nonsocial goal (instrumental effectiveness) and whether the hypothetical or actual peer would want to be his friend if he used an aggressive strategy or an assertive strategy to solve the conflict. Results indicated that outcome expectations did not differ when boys rated the instrumental effectiveness of aggression for solving social problems with hypothetical peers. However, when the boys rated the instrumental effectiveness of aggression for solving social problems with actual peers, high-aggressive boys more frequently believed that aggression would help them to achieve their own goals than did low-aggressive boys. (MM)