Abstract:

This study evaluated whether gender-related behaviors in a sample of 192 young children (M = 9.58 years; 98 boys, 94 girls) predicted peer acceptance and victimization. It was hypothesized that gender normative behaviors would lead to greater acceptance and less victimization while gender non-normative behaviors would have the opposite effect. In addition, a new Bayesian procedure of measuring gender, the Iterative Probability Procedure (IPP), was tested and validated.

As predicted, IPP scores were less influenced by random error than 3 alternate classification methods: Gender Diagnosticity, Discriminant Function Analysis, and scores derived using a single iteration of Bayes' theorem. This was evidenced by comparing gender scores generated using the 4 methods across an original dataset, and a modified dataset in which a third of the participants were randomly selected, and then incorrectly classified as the other sex. In every case, correlations between IPP scores generated from the original and modified data were significantly greater than those created using the other classification methods. This verified that IPP gender scores better indicate actual response patterns of masculine and feminine behaviors than scores derived from other methods.

Structural Equation Modeling confirmed that all gender behaviors predicted either acceptance or victimization; however, sometimes in ways contrary to the hypotheses. For boys, participating in masculine activities, greater reports of interacting with same-sex peers, and behaving in communal manners predicted greater peer acceptance and/or less victimization. For girls, participating in both masculine and feminine activities and the expression of both agentic and communal traits in girls were indicative of greater acceptance and/or less victimization. Greater degrees of unmitigated communion and unmitigated agency resulted in less acceptance and/or increased victimization in boys, whereas in girls, same-sex peer affiliation, greater levels of communion, unmitigated agency, and unmitigated communion predicted lower acceptance and/or increased victimization. Explanations of these results and their implications are discussed and future directions for research are presented.