Sex Isn't Gender: New Measure Shows About a Quarter Do Not Align

What the Study Asks: Amanda Bittner and Elizabeth Goodyear-Grant investigate whether conventional survey practice—using biological sex as a stand-in for gender—accurately captures the social and identity dimensions that matter for public opinion and political behavior. The paper addresses a core measurement problem in research on gender gaps in opinion and partisan choice.

Why This Matters: Political science research often treats gender as a simple male/female dichotomy by using respondents' reported sex. That conflation overlooks non-binary, androgynous, and varying degrees of masculine or feminine identification, which could reshape findings about how gender relates to policy preferences, party attachment, and voting behavior.

How the Authors Tested It: The authors compare a newly introduced measure of self-reported gender identification—designed to capture degrees and non-binary identities—with the conventional survey item that records biological sex. Their approach evaluates whether sex is a reliable proxy for gender and whether a finer-grained gender measure yields different analytical leverage when studying public opinion.

Key Findings:

• Sex is a reasonable proxy for gender for many respondents, but not for everyone: about a quarter of the sample show misalignment between biological sex and self-reported gender identification.
• Analyses that include the more nuanced gender measure capture variation that the sex proxy misses, offering greater descriptive and analytic precision when examining gendered patterns in opinion and partisan choice.
• The results imply that relying solely on sex can obscure meaningful heterogeneity in how people understand and express gender, with potential consequences for conclusions about gender gaps.

What This Means for Future Research: Bittner and Goodyear-Grant argue that these findings should prompt scholars to rethink standard survey items. Measuring gender directly and with greater nuance—allowing for non-binary categories and degrees of identification—can improve the validity of research on political behavior tied to gender. The paper frames this work as a starting point for broader methodological development rather than a definitive solution, and calls for additional studies to refine gender measurement in political science surveys.