David Smerdon

Is the Gender Equality Paradox Real? A machine learning test across multiple domains

Over the past decade, a number of studies in social psychology have reported a “gender equality paradox” in different domains. The general format of the main result in these papers is similar: A significant, negative correlation is reported between the gender equality of a country and the share of women in a stereotypically men-dominated field or trait (for example, the share of STEM graduates who are women). These results are often reported as paradoxical because it might be assumed that countries with stronger rights and opportunities for women would contain more equal representation in typically men-dominated fields. Domains in which gender equality paradoxes have been found include educational choices, occupational choices, personality traits, personal values, academic publishing, participation in tennis and chess, and even the naming of babies.

The gender equality paradox has thus been used as evidence for biological differences in preferences and skills, most notably by popular commentators such as Jordan B. Peterson. However, most studies that report a paradox have come under some form of criticism, usually because of the reliance on cross-country correlational data or the statistical methods employed.

The proposed project will:

• Replicate the results of the main gender equality paradox studies, subject to data availability
• Check the robustness of the claimed results using cross-country panel data and standard regression techniques, testing the sensitivity of the findings to the inclusion of country-level control variables and specification choices
• Further check the robustness using machine-learning regression methods
• Summarise which gender equality paradoxes hold, under which conditions, and how large the effects are (if any)

This project would suit a student who:

• is competent in coding and using regression packages in either STATA or R, and is comfortable with panel regressions
• has a sufficient level of understanding of econometric theory (a good test is whether you can broadly follow along with the first few paragraphs of this blog post: https://towardsdatascience.com/double-debiased-machine-learning-part-1-of-2-eb767a59975b )
• is interested in the gender equality paradox literature.

Readings:

Breda, T., Jouini, E., Napp, C., & Thebault, G. (2020). Gender stereotypes can explain the gender-equality paradox. Proceedings of the National Academy of Sciences, 117(49), 31063-31069.

Falk, A., & Hermle, J. (2018). Relationship of gender differences in preferences to economic development and gender equality. Science, 362(6412), eaas9899.

Marsh, H. W., Parker, P. D., Guo, J., Basarkod, G., Niepel, C., & Van Zanden, B. (2021). Illusory gender-equality paradox, math self-concept, and frame-of-reference effects: New integrative explanations for multiple paradoxes. Journal of Personality and Social Psychology, 121(1), 168.

Stoet, G., & Geary, D. C. (2018). The gender-equality paradox in science, technology, engineering, and mathematics education. Psychological Science, 29, 581–593. doi:10.1177/0956797617741719

Stoet, G., & Geary, D. C. (2020). The gender-equality paradox is part of a bigger phenomenon: Reply to Richardson and colleagues (2020). Psychological science, 31(3), 342-344.

Vishkin, A. (2022). Queen’s gambit declined: The gender-equality paradox in chess participation across 160 countries. Psychological science, 33(2), 276-284.