Source-linked AI summary
Gender Inequality in Research Productivity During the COVID-19 Pandemic
Ruomeng Cui, Hao Ding, Feng Zhu
TL;DR
The paper examines how pandemic-related work-from-home disruption affected research productivity and gender inequality among social science academics. Using data on new research papers, it finds that male academics significantly boosted productivity while female academics’ productivity changed little, widening the gap.
Problem
Understanding how work-from-home disruption changes productivity and related gender equity is an important operations and societal question.
Method
The study uses a large dataset of female and male academics’ new research papers from the world’s largest open-access social science repository.
Results
13.2–13.9 percent: women produced less research relative to men in the United States after lockdown began.
Takeaways & Limitations
The findings bring longstanding inequities faced by women to the forefront and identify the pandemic shock’s impact on gender inequality in academia.
Takeaways & Limitations
The findings may not generalize beyond social science disciplines, and the study has limited information about additional factors.
Abstract
from arXiv · showhide
We study the disproportionate impact of the lockdown as a result of the COVID-19 outbreak on female and male academics' research productivity in social science. The lockdown has caused substantial disruptions to academic activities, requiring people to work from home. How this disruption affects productivity and the related gender equity is an important operations and societal question. We collect data from the largest open-access preprint repository for social science on 41,858 research preprints in 18 disciplines produced by 76,832 authors across 25 countries over a span of two years. We use a difference-in-differences approach leveraging the exogenous pandemic shock. Our results indicate that, in the 10 weeks after the lockdown in the United States, although the total research productivity increased by 35%, female academics' productivity dropped by 13.9% relative to that of male academics. We also show that several disciplines drive such gender inequality. Finally, we find that this intensified productivity gap is more pronounced for academics in top-ranked universities, and the effect exists in six other countries. Our work points out the fairness issue in productivity caused by the lockdown, a finding that universities will find helpful when evaluating faculty productivity. It also helps organizations realize the potential unintended consequences that can arise from telecommuting.
1. Introduction
The paper examines whether COVID-19 lockdowns disproportionately affected female academics’ research productivity and provides systematic evidence on gender inequality in this disruption. Using SSRN data and a difference-in-differences design, it finds a widened productivity gap, with implications for academic evaluation and telecommuting policies.
- Women were expected to face greater productivity losses because lockdowns increased domestic responsibilities and interruptions alongside research work.The paper links unequal childcare and household duties to less research time and lower efficiency for female academics.
- The study addresses limited systematic evidence on how the pandemic shock affected gender inequality in academic research productivity.
- A difference-in-differences analysis compares weekly gender-specific paper production before and after lockdowns and checks parallel pre-lockdown trends.The authors also test whether a comparable seasonal period in 2019 produced a similar productivity-gap change.
- The intensified gap was stronger among assistant professors and researchers at top-ranked universities and also appeared in six other countries.
- The findings highlight fairness concerns for faculty evaluation and potential longer-term career risks associated with pandemic-related productivity disparities.The authors argue that institutions and organizations should consider these disparities when designing evaluation policies and operational choices.
2. Literature Review
The paper connects pandemic-related gender inequality to research on productivity, socially responsible operations, gender bias, and telecommuting. It emphasizes that working from home can create fairness concerns through unequal domestic burdens and altered work conditions.
- The paper contributes to productivity, social operations, and telecommuting research by examining pandemic-related fairness and equity in research productivity.
- Prior productivity research identifies workload, multitasking, task sequences, fatigue, technology, and peer effects as drivers of worker productivity.
- Working from home can alter productivity through multitasking, workload, task sequence, peer effects, and reduced interaction with colleagues.
- Gender research documents unequal domestic labor, care responsibilities, evaluation outcomes, promotion opportunities, and professional recognition for women.
- A survey of 4,500 principal investigators reported sharper reductions in female scientists’ research time, primarily due to childcare needs.
- The authors argue that pandemic-related productivity inequality may create long-term career risks and should inform organizational operations and performance evaluation.
- The paper extends telecommuting research by identifying an unexpected fairness issue arising from this operating model.
3. Theory Development
The theory explains pandemic productivity through changes in researchers’ available time and efficiency. It predicts larger disadvantages for women because domestic responsibilities and multitasking are distributed unequally.
- Research productivity is conceptualized as depending on both time available for research and research efficiency.
- Lockdowns reduce some in-person activities while increasing home responsibilities, so researchers may gain or lose research time depending on their allocation of domestic duties.
- Childcare and household work intensified because schools, daycare centers, and other support services were closed or constrained.
- Working from home can reduce research efficiency through multitasking, task switching, fatigue, altered task sequences, and weaker peer effects.
- Women were expected to lose more research time and efficiency because they generally shoulder more childcare and household responsibilities.
- The theory hypothesizes that female researchers’ productivity would be disproportionately affected relative to male researchers during the pandemic.
4. Data and Summary Statistics
The study constructs a large cross-country SSRN dataset and measures weekly gender-specific preprint production around country lockdowns. US trends show a post-lockdown divergence, with male productivity rising while female productivity changed little.
- Author names, webpages, affiliations, addresses, and external rankings identify gender, academic rank, country, and university ranking.
- Weekly productivity counts assign each author 1/n of a multi-author preprint when the paper has n authors, then aggregate counts by gender.
- Figure 1 marks March 11, 2020 as the US lockdown week and compares female and male academics’ preprint trends.
- Before lockdown, female and male productivity followed parallel trends; afterward, male academics significantly increased submissions while female productivity changed little.
- 35 percent: US research productivity increased after lockdown, with male authors contributing most of the increase.
- The discipline-level response varied substantially: some fields increased productivity, while Anthropology, Cognitive, and Information Systems experienced declines.
5. Identification
The study treats the COVID-19 lockdown as an exogenous shock and estimates its effect on gender differences in research productivity using difference-in-differences models.
- The identification exploits the COVID-19 lockdown as an exogenous shock to academic activities.
- The approach assumes the shock was unrelated to researchers’ anticipated responses, although strategically accelerating paper completion could confound the treatment effect.
- The analysis compares female and male researchers’ productivity before and after the lockdown using discipline-level panel data.
- The outcome is the logged number of preprints uploaded by each gender group within disciplines and weeks.
- The models include weekly time effects and discipline fixed effects to control for time trends and time-invariant disciplinary characteristics.
- The aggregate specification estimates gender differences in logged weekly preprint counts across disciplines.
6. Empirical Results
The empirical results show that the lockdown widened the gender productivity gap, with effects varying across academic ranks, university rankings, disciplines, and countries.
- Main Results: Female academics produced fewer preprints generally, and their productivity declined significantly relative to male colleagues after the lockdown.
- Main Results: 13.2% reduction in female academics’ productivity relative to male academics’ productivity occurred over the 10-week post-lockdown period.
- Effects across academic rank: Female assistant professors experienced the most significant productivity drop compared with male junior faculty after the lockdown.
- Effects across university ranking: The pandemic effect was more pronounced in top-tier universities and generally decreased as lower-ranked universities were included.
- Effects across countries: 21 out of 25 countries experienced declines in female researchers’ productivity, with statistically significant declines in six additional countries.
- Effects across countries: The country comparisons may understate changes because SSRN is used primarily by US researchers, limiting preprint coverage elsewhere.
7. Robustness Checks
The robustness checks examine identification assumptions, seasonality, research quality, coauthorship, and author-sample composition. Across these tests, the reported gender-productivity gap remains supported, while research-quality measures generally do not change significantly.
- Parallel trends: No pre-treatment differences in female and male productivity trends support the parallel-trends assumption.The analysis tests week-by-week effects before the shock, and visual inspection shows parallel pre-shock evolution.
- Seasonality and falsification: Insignificant placebo-treated average treatment effects in 2019 indicate that women’s productivity did not decline significantly in the previous year.The falsification test uses the same time window as the 2020 analysis to assess whether seasonality explains the result.
- Research quality: Insignificant average treatment effects for abstract views and downloads suggest that research quality did not change significantly after lockdown.The study uses abstract views and downloads as quality indicators and reports that the findings are unlikely to be driven by shifts in quality.
- Research quality: Female researchers’ downloads per preprint decreased relative to male researchers’ after lockdown.This result appears in the download-based quality measure even though the general quality effects are reported as insignificant.
- Coauthorship: All-female preprints dropped significantly after lockdown, with larger and more significant gender-interaction coefficients than in the full sample.The authors interpret this pattern as indicating more pronounced inequality when research teams have only female authors.
- Author-sample composition: The main productivity result remains consistent when extending the US post-treatment window to 40 weeks and using a balanced author panel.Within the same group of authors, female researchers’ productivity dropped significantly after lockdown compared with male researchers’ productivity.
8. Conclusions
The study finds that the COVID-19 lockdown widened gender inequality in social-science research productivity, despite increasing overall productivity. The findings highlight implications for faculty evaluation, institutional support, and telecommuting policy.
- 13.2–13.9 percent fewer papers were produced by women than men in the United States after lockdown.
- The productivity inequality was especially significant for junior faculty and researchers at top-ranked universities, and was significant in seven countries.
- The results remained robust in same-gender-author samples, balanced panels, and an extended data sample.
- If lockdowns persist too long, junior women and women at top-ranked universities may face greater unemployment or career risk.
- 35 percent overall productivity growth coincided with a 13 percent increase in the gender productivity gap.
- Universities could provide additional support, including childcare, and account for unequal productivity when evaluating professors for promotion.
- The findings suggest telecommuting can have unintended consequences for gender inequality, so institutions should consider gender equality when designing telecommuting policies.
- The study is limited because it focuses on social sciences and lacks information needed to identify the mechanism behind the observed patterns.
Appendix
The appendix documents supplementary figures, tables, robustness checks, placebo and parallel-trends tests, and analyses across countries, university rankings, academic ranks, and outcome measures.
- The appendix includes figures and tables covering time trends, summary statistics, university rankings, academic ranks, and author-composition restrictions.
- The reported analyses include parallel-trends and falsification tests, with weekly time fixed effects and significance thresholds documented for regression results.
- Supplementary regressions examine lockdown effects on gender inequality, abstract views, and downloads across specified post-lockdown weeks.