Women are more likely to start a research career now than they were 20 years ago, reveals a longitudinal study of the publishing records of millions of researchers around the world. But they are less likely to continue their academic careers than are their male contemporaries, and in general publish fewer papers.
Ludo Waltman, a quantitative scientist at Leiden University in the Netherlands, and his colleagues took a deep dive into the huge Scopus citation and abstract database, hosted by Elsevier. They looked at the publication careers of some six million researchers globally who had authored at least three papers between 1996 and 2018. The team posted its findings on the preprint server arXiv.org1.
The authors found that the proportion of women starting a career in science rose over time. In 2000, 33% of researchers starting their publishing career were women; that grew to 40% in recent years (see ‘Gender gap’). Waltman says that although the results are not surprising, it’s important that we now have concrete statistics confirming the trend for many countries and scientific disciplines.
In the physical sciences, mathematics and engineering, male authors still made up a much higher proportion of authors than did women, even in more recent years (see ‘Differences by discipline’).
Waltman and his team took their work further by tracking the researchers’ publication records to see whether they continued authoring scientific papers — a proxy for continuing a career in science. They discovered that women were less likely to continue publishing papers than were men, whatever year they began their careers.
But Waltman says his team was surprised — given that there is a well-known problem of fewer women than men progressing to senior roles in science — that women were only slightly less likely to continue to publish papers than were men. They found that 54% of women who started publishing in 2000 had dropped out 15 years later, as opposed to 52% of men.
Nevertheless, “while the length of the scientific careers of men and women is quite similar, there are important differences between men and women in the way in which their scientific careers develop”, he says. In general, men seemed to progress to senior roles — roughly judged by appearing as the last author on a paper — more quickly. On average, they also published 15–20% more papers than did women over the time span of the data, though there is wide variation across different fields.
One important limitation of the study was that it excluded data from India and China — which together account for around one-third of the world’s population — because the authors’ algorithms struggled to assign gender unambiguously to names from these nations. It also did not account for non-binary authors.
Flaminio Squazzoni, a sociologist at the University of Milan in Italy, agrees that the lack of data from India and China is a big gap. “It is probable that if we would have data on these two countries, the perspective on women academics could be slightly worse,” Squazzoni predicts.
Nevertheless, “it’s an honest study” and the Scopus database is a rich source of information, says Squazzoni. His own team published work this year showing that there is little gender bias in peer-review processes2. That, combined with this latest study, suggests that investments in gender-diversity schemes in science could be paying off, he says.
Waltman says that although the overall picture looks positive for more women entering science, it’s important to remember that the increase is an average over many countries and scientific disciplines, and that there is huge variation within the numbers. He is also keen to point out that scientific careers are about more than publication numbers. “The bibliometric lens through which many studies, including our own, look at diversity in science is a very narrow one,” he says.
Nature 596, 177 (2021)
Boekhout, H., van der Weijden, I. & Waltman, L. Preprint at https://arxiv.org/abs/2106.12624 (2021).
Squazzoni, F. et al. Sci. Adv. 7, eabd0299 (2021).