Home // FSTM // News // New research on how race and gender shape science

New research on how race and gender shape science

twitter linkedin facebook email this page
Published on Monday, 10 January 2022

Researchers from the University of Luxembourg together with scientists from the University of Montreal, Georgia Tech and Berry College have recently demonstrated that racial and gender identities shape the way science is produce, with Black and Latinx authors being underrepresented in terms of participation and citation counts, affecting scientific advancement. Their findings have been recently published in the prestigious journal Proceedings of the National Academy of Sciences (PNAS) of the United States of America.

Analysing diversity and knowledge advancement 

Strong disparities are observed in the composition of the scientific workforce. At the global level, women account for less than a third of scientists and engineers. In the United States, women represent 28.4% of the scientific workforce, and this percentage varies by domain, with a high of 72.8% in psychology and a low of 14.5% in engineering. Disparities are also observed at the intersection of race and gender, with White men comprising a disproportionate amount of the US scientific workforce. 

Barriers to entry and participation have been well-studied; however, few have examined the effect of these disparities on the advancement of science. Furthermore, most studies have looked at either race or gender, failing to account for the intersection of these dimensions. “Our analysis uses millions of scientific papers to study the relationship between scientists and the science they produce. We found a strong relationship between the scientists race and gender identity and their research topics, suggesting that diversity changes the scientific portfolio with consequences for career advancement for marginalized individuals”, explains Diego Kozlowski, doctoral student within the Doctoral Training Unit on Data-driven Computational Modelling and Applications (DRIVEN) at the University of Luxembourg and first author of the study.

Computational methods 

Researchers have examined 5.4 million articles indexed in the Web of Science database between 2008 and 2019. They focused on first authors as they are generally those who have contributed the most to an article and represent the most visible name in bibliographic references. Using computational methods based on census data and country-specific lists of names, they identified first authors’ race and gender and showed that White and Asian populations are overrepresented among US authors, while Black and Latinx populations are underrepresented.


“We found that women tend to work more on topics related to education, nursing and gender-based violence. Black authors publish more on racial-discrimination, while Latinx authors publish more on migration, and Latinx bodies. These research topics have a different impact in their fields. Topics in Social Science and Health where White and Asian men are over-represented, are also those that have more citations. There is a distributional bias by topics. But even if we see the distribution within each topic, no matter if they are highly cited or not, marginalized groups tend to have lower number of citations. There is also a within topic bias.” 

Some recommendations 

The authors explain that marginalised authors tend to publish in scientific disciplines and on research topics that reflect their gendered and racialised social identities. The study recommends that scientific institutions recognise the existence of knowledge gaps related to author race and gender and promote topics where gendered and racially marginalised authors are more present. “To enhance the robustness of science, research organisations should provide adequate resources to historically underfunded research areas while simultaneously providing access for minoritised individuals into high-prestige networks and topics”, concludes Diego. 

About the Doctoral Training Unit DRIVEN 

Launched in 2018 by the University of Luxembourg in collaboration with the public research centres LIST and LISER, the Doctoral Training Unit on Data-driven Computational Modelling and Applications (DRIVEN) trains cohorts of doctoral candidates who develop data-driven modelling approaches to tackle complex data-intensive problems in all sectors of the economy. The project is funded by the Luxembourg National Research Fund (FNR) and coordinated by Prof. Andreas Zilian from the Department of Engineering at the University of Luxembourg. More information: https://driven.uni.lu 

Publication: “Intersectional inequalities in science”, PNAS, January 2022 

Picture: Copyright Lina Castellanos