Measuring gender as a system using confirmatory factor analysis

Jill Williams, University of Colorado at Boulder

This paper develops a new method for constructing measures of gender and women’s empowerment with cross-sectional survey data. I re-conceptualize gender and women’s empowerment for measurement purposes and argue that gender and women’s empowerment are best measured as a system of interrelated dimensions derived from context specific gender norms. Qualitative research on women’s empowerment is used to guide the development of a theoretical model of women’s empowerment in rural Bangladesh which is then tested using confirmatory factor analysis of data from the 1996 Matlab Health and Socioeconomic Survey (MHSS). The results of the confirmatory factor analysis are then used to construct weighted measures of women’s empowerment that are compared to simple scale measures. This analysis advances the research on women’s empowerment by testing many of the theoretical assumptions found in demographic research on women’s empowerment, and, most importantly, makes sophisticated measures of gender and women’s empowerment accessible to demographers.