Measuring Gender and Sex in Surveys: Lessons Learned from 50 Years of Cross-National Survey Data and Nonresponse Patterns

Abstract

While gender and sex play a pivotal role in public opinion surveys, their measurement often draws criticism for being imprecise and incomplete. Despite increasing attention, the scope, challenges and changes in the gap between survey practices and contemporary gender theory remain unclear. This paper examines the problems that arise when aligning gender/sex research developments with 50 years of cross-national survey data. Specifically, it analyses survey documentation on respondent’s sex and/or gender as well as nonresponse trends across 23 international survey projects, 174 waves and 3,329 national surveys, covering the period from 1966 to 2017 and including 156 countries. The results of this large-scale historical analysis of sex and gender survey measures uncover persistent challenges of inadequate documentation, limited sex/gender self-identification options, and lack of nonbinary categories. It also shows increasing popularity of the term “gender” since it appeared in survey documentation in the 1990s. Nonresponse patterns reveal country effect in prevalence of missing codes for sex/gender items, with an overall decreasing trend. The paper discusses observed challenges and addresses survey practitioners' possible counterarguments against introducing change. It contributes to the literature on empirical data collection on sex and gender and offers practical insights to survey producers and survey data users.