The measurement of masculinity, femininity and androgyny: A reply to Rowland (1977)

In a recent article, Rowland has presented proportions of males and females scoring masculine, feminine and androgynous on the Bem Sex-Role Inventory (BSRI). Among others, her conclusion that fewer Australian males are masculine sex-typed than their American counterparts is fallacious because of limitations in the design of the study, the analysis and reporting of the data, and the inferences drawn from those results. In the present study the BSRI was administered to first-, second-, and third-year Macquarie University students, postgraduate Master of Business Administration students, and married couples with children. Means, standard deviations, and intercorrelations for the masculinity, femininity, and social desirability scales are reported separately for males and females, together with proportions falling into the 5 Bem sex-role categories and 4 Spence et. al. categories. The generality of the sex-role proportions over the present disparate samples, and their similarity to Bem's, highlight the atypicality of Rowland's findings.