Meta-Analysis of Factor Structure Studies

Abstract

The importance of replication is widely recognized in several research traditions. Empirical synthesis of seemingly diverse results may be particularly important when factor structure similarities are being evaluated because apparent differences in findings may be an artifact of different locations of factor axes, even though variable positions in factor space may be virtually identical. The present study empirically evaluated the comparability of seven sets of factor structures reported in previous research with the Bem Sex-Role Inventory. Results indicate that the inventory’s primary factors were reproduced in the studies, although the Masculine scale was somewhat less invariant than its Feminine counterpart. Thus, the results support the validity of this androgyny measure across several variations in sample types. The study represents a heuristic example of a methodology for the meta-analysis of factor structure studies.