The chart above is from a study just published in Nature, showing that articles published with women as first authors are cited less frequently than articles with male first authors. Since citations are an important currency by which scientists are rewarded, this matters a great deal.
Earlier this fall, we organized a gender gap symposium (see here for all posts) here at The Monkey Cage, which discussed similar findings in political science based on the research of Barbara Walter, Daniel Maliniak, Ryan Powers, and Sarah Mitchell. In some ways, the evidence in those studies was stronger as the analyses were able to compare citations to articles accepted in the same year and the same journal (the graph above is just raw numbers). On the other hand, the Nature study is based on 5,483,841 research papers, making it by far the largest study of its kind.
The Nature study has lots of other interesting data, including an interactive map of where gender disparities in research output are larger or smaller. Most of the world still fails to fully take advantage of the intellectual capabilities of half its population.