Abstract
Science attempts to be a meritocracy; however, in recent years, there has been increasing evidence for systematic gender bias against women. This bias is present in many metrics commonly used to evaluate scientific productivity, which in turn influences hiring and career success. Here we explore a new metric, the Altmetric Attention Score, and find no evidence of bias across many major journals (Nature, PNAS, PLOS One, New England Journal of Medicine, Cell, and BioRxiv), with equal attention afforded to articles authored by men and women alike. The exception to this rule is the journal Science, which has marked gender bias against women in 2018, equivalent to a mean of 88 more tweets or 11 more news articles and a median of 20 more tweets or 3 more news articles for male than female first authors. Our findings qualify Altmetric, for many types and disciplines of journals, as a potentially unbiased measure of science communication in academia and suggest that new technologies, such as those on which Altmetric is based, might help to democratize academic evaluation.