Effects of Validity and Bias on Gender Differences in the Appraisal of Children's Competence: Results of MTMM Analyses in a Longitudinal Investigation

Abstract

Third graders (N = 598) and 6th graders (N = 630) participated in a 3-year, 6-wave, 2-cohort study of perceived competence. As part of a multitrait-multimethod longitudinal study, teachers, peers, and students completed measures of perceived academic competence, social acceptance, athletic competence, physical appearance, and behavioral conduct twice per year. The measures of competence manifested significant levels of convergent and discriminant validity for most domains at all grade levels and for both genders. Validity differences emerged between sources of information, between age groups, and between domains of competence. Evidence of bias also emerged, such that self-report measures systematically underestimated girls' competence in all domains and teacher ratings overestimated boys' competence in athletics. Corrected for bias, latent mean differences between boys and girls were still significant. Girls were perceived as more competent than boys in most domains, particularly behavioral conduct. Boys were perceived as more competent than girls in athletic competence. Developmental changes in perceived competence between 3rd and 8th grades were unremarkable.