Answers are grounded in this paper and cite their sources.
Research samples have historically overrepresented men, resulting in worse outcomes for women. Even when women are well represented, inequity might persist due to underlying differences in how people reason about gender. Theories of androcentrism argue people think of men as prototypical “people,” resulting in a corresponding tendency to emphasize gender more for women than men. Based on this theory and building on prior archival work, we conducted four preregistered experiments ( N = 1,844). Participants read identical scientific findings based either on all-male or all-female samples (between-subjects design). In either case, the gender homogeneity of the sample is a key limitation that should be noted when describing results. However, participants were more likely to pick women as a keyword to represent findings based on all-female samples (compared to leaving gender unlisted or listing both women and men ) than to pick men for findings based on all-male samples. In an internal meta-analysis, we found these effects among U.S. community samples, a multinational student sample, and a journalist-trained sample. Effects were weakest and not significant among journalists, suggesting that expertise might help to encourage more precise and consistent descriptions. Our work has practical implications for understanding, recognizing, and addressing gender bias in communication.
No takes yet. Share an insight, caveat, or question.
Bailey et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: