Citation

Glynn & Sen (2015) — Daughters & Decisions (AJPS): daughters make judges modestly more pro-women.https://scholar.harvard.edu/files/msen/files/daughters.pdf

Research Question

Does having at least one daughter causally affect how U.S. Courts of Appeals judges vote in civil cases involving women’s rights and gender-related legal issues?

Key Takeaways

Judges with at least one daughter vote more often in a feminist or pro-women’s direction in civil cases involving gender-related issues, conditional on the total number of children; the effect is concentrated among Republican male appellate judges and does not appear among Democratic judges at statistically significant levels; the relationship is issue-specific, emerging in gender-related civil cases such as employment and pregnancy discrimination, reproductive rights, and Title IX, but not in criminal cases or in civil cases without a gender dimension; the effect appears once a judge has at least one daughter and does not increase with additional daughters, weighing against simple lobbying or pressure-based explanations; the findings show that personal experiences can exert a measurable influence on judicial voting in narrowly defined substantive contexts, beyond party affiliation, ideology, or judge gender.

Dataset Description

The study combines newly collected biographical data on the number and gender of biological children of 224 U.S. Courts of Appeals judges with a hand-coded dataset of 990 civil gender-related appellate cases decided between 1996 and 2002, producing 2,674 individual judge votes. Gender-related cases include employment and pregnancy discrimination, abortion and reproductive rights, and Title IX claims. To test whether any observed effects generalize beyond gender issues, the authors also analyze a separate random sample of approximately 3,000 published appellate cases across all issue areas from the Kuersten and Haire dataset covering the same period. The jurisdiction studied is the U.S. federal courts of appeals.

Methodology

Statistical and quantitative analysis using quasi-experimental identification based on the near-random assignment of a child’s sex, conditioning on the total number of children.

Key Findings

Conditional on family size, judges with at least one daughter are significantly more likely to vote in a feminist or pro-women’s direction in civil gender-related cases than judges with only sons. The estimated effect corresponds to a roughly seven- to nine-percentage-point increase in feminist voting and is driven primarily by Republican male judges. No comparable effect is found in criminal cases or in civil cases without a gender dimension. The results are robust across multiple model specifications, controls for ideology, demographics, circuit, and time, and subsample analyses including judges with only one child, reducing concerns about fertility stopping rules or selection effects.

Summary

This study examines whether judges’ personal family experiences influence judicial decision-making by exploiting the quasi-random assignment of a child’s sex as a natural experiment. Glynn and Sen examine whether having daughters affects how federal appellate judges vote in civil cases involving women’s rights, using newly assembled data on judges’ family composition and a comprehensive, hand-coded dataset of gender-related appellate decisions from 1996 to 2002.

The authors show that judges with at least one daughter vote more often in a feminist or pro-women’s direction in gender-related civil cases, even after controlling for ideology, judge gender, race, religion, court, and time period. The effect is substantively meaningful, statistically robust, and concentrated among Republican male judges, while no similar shift appears in criminal cases or in non-gender-related civil disputes.

By demonstrating that personal experiences can systematically influence judicial behavior in a constrained, issue-specific manner, the study extends prevailing models of judicial decision-making beyond ideology and legal doctrine alone. The findings suggest that judicial outcomes in certain domains reflect structured and empirically detectable influences tied to life experience, while remaining bounded by institutional context and subject-matter relevance.

How the Study Advances Empirical Understanding of Legal Outcomes

The study finds that appellate outcomes in gender-related civil cases reflect structured and repeatable patterns linked to judges’ life experiences, demonstrating that decisions in this domain are not solely the product of abstract doctrine or partisan identity. By isolating a quasi-random personal attribute and showing its effect only within a defined class of cases, the analysis shows how institutional decision environments can systematically condition outcomes in issue-specific ways without implying randomness or universal effects. This methodological approach aligns with Pre/Dicta’s emphasis on rigorous, case-based empirical analysis of legal outcomes and decision contexts as a necessary foundation for high-level strategic litigation practice, while remaining grounded in observable data rather than conjecture.