Gender differences in research areas and topics: An analysis of publications in 285 fields

Although the gender gap in academia has narrowed, females are underrepresented within some fields in the USA. Prior research suggests that the imbalances between science, technology, engineering and mathematics fields may be partly due to greater male interest in things and greater female interest in people, or to off-putting masculine cultures in some disciplines. To seek more detailed insights across all subjects, this article compares practising US male and female researchers between and within 285 narrow Scopus fields inside 26 broad fields from their first-authored articles published in 2017. The comparison is based on publishing fields and the words used in article titles, abstracts, and keywords. The results cannot be fully explained by the people/thing dimensions. Exceptions include greater female interest in veterinary science and cell biology and greater male interest in abstraction, patients, and power/control fields, such as politics and law. These may be due to other factors, such as the ability of a career to provide status or social impact or the availability of alternative careers. As a possible side effect of the partial people/thing relationship, females are more likely to use exploratory and qualitative methods and males are more likely to use quantitative methods. The results suggest that the necessary steps of eliminating explicit and implicit gender bias in academia are insufficient and might be complemented by measures to make fields more attractive to minority genders.

💡 Research Summary

The paper investigates gender differences in academic research fields, topics, and methodological preferences by analysing a massive corpus of US‑authored scholarly articles published in 2017. Using Scopus, the authors extracted first‑author papers across 285 narrow fields nested within 26 broad disciplinary categories. For each article they harvested titles, abstracts, and keywords, tokenised the text, and applied TF‑IDF weighting to generate field‑specific word‑frequency profiles. A logistic regression model with LASSO regularisation identified words that were statistically associated with male or female authors after correcting for multiple comparisons.

The results reveal three major patterns. First, the classic “people‑thing” dimension—where men gravitate toward thing‑oriented topics and women toward people‑oriented topics—holds for many domains. Male‑associated terms include “patient,” “clinical,” “treatment,” “model,” and “simulation,” whereas female‑associated terms feature “care,” “family,” “education,” “qualitative,” and “interview.” Second, notable exceptions to the people‑thing framework emerge. Women are disproportionately represented in veterinary science and cell biology, suggesting that societal values attached to animal welfare and microscopic life may attract female scholars beyond the simple people‑thing dichotomy. Conversely, men dominate fields characterised by abstraction, power, and control—politics, law, and related sub‑areas—where terms such as “abstraction,” “power,” and “control” appear more frequently. Third, methodological preferences differ by gender: female authors tend to employ exploratory, qualitative approaches (e.g., case studies, interviews), while male authors favour quantitative, model‑driven techniques (e.g., statistical analysis, simulation). Effect sizes for these associations are generally medium (Cohen’s d ≈ 0.3–0.5), indicating substantive practical relevance.

The authors acknowledge several limitations. Focusing on first authors excludes contributions from co‑authors, potentially under‑estimating mixed‑gender collaborations. Text‑based inference cannot capture the full nuance of researchers’ motivations, and the analysis is confined to a single year, precluding longitudinal insights. Despite these constraints, the study provides robust empirical evidence that gendered patterns in academic publishing are only partially explained by the people‑thing theory. Additional factors—such as perceived status, societal impact, and the availability of alternative career pathways—appear to shape field selection and methodological choices.

Policy implications are drawn from these findings. Merely eliminating explicit and implicit bias may be insufficient; institutions should also work to make traditionally male‑dominated fields (e.g., politics, law, engineering) more inclusive for women and to sustain the appeal of fields where women already excel (e.g., veterinary science, cell biology). Initiatives could include promoting gender‑balanced mentorship, offering equitable funding for qualitative research, and redesigning evaluation criteria to value diverse methodological contributions. By addressing both cultural climate and structural incentives, academia can move toward a more gender‑balanced distribution of scholars across the full spectrum of scientific inquiry.