Masculinity and femininity: The role of desirability in the relationships among measures based on self-concept and personality traits

With data gathered from 96 female college students, intercorrelations were examined among self-concept ratings of masculinity and femininity, and scales from two personality-trait measures of masculinity and femininity—the Adjective Check List (ACL) and the Extended Personal Attributes Questionnaire (EPAQ). The sign and magnitude of the correlations between pairs of measures were successfully predicted on the basis of the congruence between measures of both the gender relevance and the desirability of the self-rated traits comprising each measure. The desirability of the trait items employed in self-report personality-trait scales of masculinity and femininity evidently introduces methods variance that can either inhibit or potentiate the relationship with other scales composed predominantly of either desirable or undesirable traits. When balanced for the desirability of trait items, personality-trait scores of masculinity and femininity were substantially inversely correlated with each other, and only marginally correlated with self-concept ratings of masculinity and femininity, but these correlations remained in the anticipated directions. This result is consistent with the findings of Pedhazur and Tetenbaum (“Bem Sex Role Inventory: A Theoretical and Methodological Critique,” Journal of Personality and Social Psychology, 1979, 37, 996–1016), and supports Spence's (Masculinity, Femininity, and Gender-Related Traits: A Conceptual Analysis and Critique of Current Research,” In B. A. Maher and W. B. Maher (Eds.), Progress in Experimental Research, Vol. XIII, Normal Personality Processes, New York: Academic Press, 1984) position that personality traits should be viewed as only one of many possible referents for gender/sex role identity.