A psychometric study of the Bem Sex Role Inventory with an older adult sample.

The factor structure of the Bem Sex Role Inventory (BSRI) for an older adult sample was investigated. Confirmatory factor-analytic procedures failed to yield an adequate fit for the orthogonal two-factor model proposed by Bem for either men or women. Maximum-likelihood exploratory factor analyses were conducted separately for each sex. Eight- and nine-factor solutions for men and women, respectively, emerged as the most adequate models representing the observed variance/covariance matrices. Issues are raised regarding the usefulness of median-split-based procedures with the BSRI for purposes of sex role classifications and for assessing age-related changes in personality functioning among older adults. An alternative conceptual and methodological approach is proposed.