Confirmatory Factor Analysis of the BEM Sex-Role Inventory

Abstract

Confirmatory maximum likelihood factor analyses of the long and short versions of Bern's Masculinity and Femininity scales were conducted using two samples. The long-form Femininity Scale contained a number of items that did not load properly on the hypothesized factor. The short form of the inventory was more homogeneous than the long form and provided a better fit to the data than both two- and four-factor models of the long form.