Birds of a feather flock together!

Researchers from Göttingen, Marseille and Yale show gender bias in science

March 21, 2017

Science aims to be objective; its validity universal, its result validated without regard to the scientist’s personal characteristics. Researchers from the Max Planck Institute and the Bernstein Center for Computational Neuroscience in Göttingen (Germany), the CNRS in Marseille (France) and Yale University analyzed whether reality in science lives up to this ideal. Their study has been published in the journal eLife. Based on a large data set of more than 40.000 articles published in the last 10 years they conclude that during the peer review process especially male editors select male reviewers to assess an article’s merit. Hence, female scientists are participating even less in the scientific publishing process than expected by their numerical underrepresentation alone.

Do women review?

The researchers analyzed a core aspect of science called “peer review”. During the publication process every manuscript is assessed by other scientists, which are unknown to the authors. Reviewers are selected manually by editors of the journals in which publication is sought. The article will only be published after a positive evaluation regarding quality, relevance and methodology by typically several independent reviewers and its content will then be considered a scientific fact. Scientists therefore consider peer review the lifeblood of research in academia and it is imperative that reviewers are assigned exclusively based on their scientific expertise, without regards to a scientist’s personal characteristics. This is, unfortunately, not always the case as a group of researchers discovered in a multi-disciplinary data set of more than 40000 articles published in the last 10 years. Using tools from statistics and graph theory the researchers could show for the first time that across disciplines women are underrepresented in the assessment of research articles. Beyond that, they discovered that both male and female editors have a same-gender preference when appointing reviewers. This phenomenon, known in sociology as homophily, is well known from daily experience. Finding homophily in peer review shows novel aspects of gender bias in science. This form of discrimination undermines the integrity of science and goes far beyond numerical underrepresentation of female scientists.

Hope on the horizon

The team of researchers also found that the mechanisms of homophily differ between male and female editors. While this same-gender preference is restricted to a small number of female editors, it is wide-spread among male editors. The large majority of female editors, thus, already select reviewers without gender-bias. If the remaining editors become aware of their potentially unconscious biases, homophily might be eliminated completely. Which social enterprise, if not science, would have a better chance to reach this ideal?

New tools for more objectivity

If the suggestions of the scientists were implemented, their success could be assessed in a few years. “Our tools are open access. We invite all decision makers to assess and improve their own methods and policies”, highlights Demian Battaglia, CNRS scientist at Aix-Marseille University, who supervised the study. The scientists all agree that there is no place for unreasonable unconscious attitudes as homophily in science; and only with the joined effort of all participants the problem of homophily can be solved. After all, it is known in science that diverse working groups function better and that biases are detrimental. 

Go to Editor View