Gender in the International Studies Quarterly Review Process

We have good reasons to believe—not only from our own experience and observation, but also from recent studies—that women are underrepresented in syllabi (Colgan 2017), citations (Maliniak, Powers, and Walter 2013), faculty rosters (Shen 2013), and, most relevant to this forum, the pages of our leading journals (Mathews and Andersen 2001; Breuning and Sanders 2007; Evans and Moulder 2011; Williams et al. 2015). While we see considerable variation in the number of female authors in each issue of International Studies Quarterly (ISQ), women publish in the journal at a rate lower than we would expect given the number of research-active female scholars in international studies. The December 2017 issue has 26 authors, six of whom are women. Only two articles are single authored by a woman and only one is coauthored only by women. The pooled average across the last three years is a little less lopsided, but not by much. Only about 36% of articles accepted in 2013–16 had at least one female author, only around 11% were authored solely by women and about 5% were coauthored by all-female teams. Looking only at ultimate outcomes, however, only provides part of the picture. For example, are there gaps in representation or biases at work in other stages of the editorial process that contribute to the underrepresentation of female authors in “top” journals? Is this pattern reflected in the pool of submissions? Or both? In this article, we use data from our online submissions system—supplemented by hand coding conducted by a variety of editorial and undergraduate assistants—to examine patterns that might suggest bias in ISQ’s peer-review process. Only one model suggests evidence of any such bias (though in the opposite direction), which implies that the “gender gap” in publication rates is mostly—if not entirely—a function of differential rates in submissions. That is, that larger structural factors (and perhaps expectations of bias at ISQ) account for the “gender gap.” But we caution that such analysis is preliminary, based on blunt aggregate data, and therefore of limited value.