| Abstract: | The present study, utilizing an analogue procedure and a short-term longitudinal design, examined the relative strength of observed behavior in predicting mid- to late-adolescents' ratings of that behavior. Direct observations were made of 89 adolescents during videotaped problem-solving interactions with groups of unfamiliar peers. Within 6-member peer groups, counterbalanced for gender, the adolescents participated in a series of problem-solving and discussion tasks. Peer groups met one hour daily for six consecutive days. The interactions were coded with the LIFE code and three behavioral constructs were assessed: facilitative, problem-solving, and aggressive. Same- and opposite-sex peer ratings of three factors, sociable, aggressive, and withdrawn, were obtained at the end of the second and sixth days. Hierarchical linear modeling analyses revealed that same-sex sociability ratings were predicted by prosocial behaviors of both girls and boys whereas boys' aggressive behavior also predicted these ratings across gender. In contrast, aggressive behavior displayed by girls predicted boys' ratings of girls' aggression. Withdrawal ratings were related to low levels of all behaviors. The results provide further evidence that peer perceptions may be driven by characteristics associated with gender that have been determined by culturally proscribed practices and socialization efforts. |
