During the first months of the COVID-19 pandemic, school closures and severe lockdown measures put family support out of reach in many countries. The result was an increased childcare burden that was shouldered more by women than men. An international study1 coordinated by Flaminio Squazzoni, a sociologist at the University of Milan, found that this applied also in the research industry.
The scientists analysed journal submissions and peer-review activities from January 2018 to May 2020 for more than 2,300 journals by the Netherlands-based publisher Elsevier, which shared its database with them under a confidentiality agreement. Overall, nearly 5 million scientists were included in the analysis either as authors or reviewers. While the absolute number of journal submissions increased between February and May 2020 over the same period of the previous year (mostly due to COVID-related papers), the increase was larger for male scientists than for females, especially junior female researchers.
Previous studies had found similar results, but they were mainly focused on preprints or specific scientific areas. Instead, this study focuses on a larger set of peer-reviewed journals and covers more research areas, especially those where preprints are not as common. Moreover, Squazzoni and his team were able to track the productivity of individual scientists before and during the pandemic. “Our unique database allowed us to consider the impact of lockdowns on individual researchers, and then aggregate those impacts, instead of making assumptions on the male-female proportion in normal times for each scientific area”, says Squazzoni.
Each researcher in the database was defined as male or female based on their name, using gender guessing algorithms. Age was inferred by their profile in the Scopus database, and the country of residence was used to estimate lockdown length and timing through the Google Mobility Report. Squazzoni and his collaborators are aware that gender guessing algorithms have some limits. “They disregard non-binary identities and fail more often in some geographic areas because they are trained on datasets that contain mainly western names,” he says. “But, we excluded data where the algorithm predictions were not highly reliable”.
Results show that between February and May 2020, the number of submissions to Elsevier journals increased by 30% compared to the average of the same months in 2018 and 2019. The trend was even stronger in health and medicine journals, with a 63% increase. However, the increase for women, and in particular younger women who are more likely to have young children, was significantly smaller than men. Health and medicine journals showed the biggest difference in the increase, followed by physical sciences and engineering, life sciences, and social sciences and economics. The only group that did not show such a gap, was that of social sciences and economics scholars in the older age cohort. On the other hand, the total number of accepted peer-review invitations increased by 29% from 2019 to 2020, but at the individual level, researchers declined more frequently in 2020 than in 2019 with no considerable gender differences.
“Writing a referee report requires less creative energy than drafting a manuscript, which could explain why women managed to maintain the peer-review activity despite the fragmentation of working time due to family care duties” says Laura Inno, astronomer and post-doctoral researcher at the Parthenope University of Naples, who authored an early analysis2 of the gender gap in submissions of astronomy and astrophysics papers to the arXiv preprint server during the first wave of the pandemic in Italy.
“We are monitoring data on the following period of the pandemic, but independently of what happened later, the gap we see in those first months could have long-term consequences”, says Squazzoni. More submissions mean more publications and thus more citations, he notes, some of the most important metrics used when awarding permanent positions. “Funding agencies could reverse this trend by applying special policies for this extremely anomalous period.”