Recent years have seen increasing evidence of systematic gender bias against women working in science, technology, engineering, and mathematics (STEM). Whether unconscious or otherwise, this bias can see a woman’s gender negatively affect the evaluation of their academic performance when traditional metrics are used. In a Nature Index blog, Bjarne Bartlett and colleagues discuss their study, which examined how alternative metrics, or altmetrics, could help to rebalance this inequality.

In academia, decisions regarding hiring, promotion, and tenure are often based on traditional evaluation metrics, such as citation counts and grant allocations, which have been found to disproportionately favour men. Altmetrics, on the other hand, measure the amount of online attention a paper receives on digital platforms, such as news sites and social media, and many journals now publish the Altmetric Attention Score (AAS) for individual journal articles.

Thanks to the broad nature of online audiences, altmetrics could effectively democratise research evaluation. To test this theory, the study by Fortin et al, published in Scientometrics, analysed the AAS of over 200,000 articles published between 2011–2018 in 7 major journals/preprint servers:

• Science
• PNAS
• PLOS One
• Nature
• Cell
• NEJM
• bioRχiv.

The authors found that articles by male and female authors generally received equal online attention one year after publication. The exception to this was in 2018, when marked gender bias against women was found in Science, the reasons for which are unknown.

By using altmetrics together with traditional citation-based metrics, a fairer, more equitable system for performance evaluation could be established for academics, whatever their gender.

The authors state that further research is required to determine why altmetrics did not show evidence of substantial gender bias for most journals. Understanding the underlying reasons may help to identify mechanisms that could be applied to eliminate bias from other performance assessment tools. The authors conclude that by using altmetrics together with traditional citation-based metrics, a fairer, more equitable system for performance evaluation could be established for academics, whatever their gender.

—————————————————–

—————————————————–