For our fourth installment in the gender gap symposium (see here, here, and here for the first three) I am delighted to introduce Sara McLaughlin Mitchell. Sara is professor and department chair at the University of Iowa. She researches international cooperation and conflict. She is a long time organizer of the Journeys in World Politics workshop, which brings together senior and junior women working in international relations. She has also published on gender differences in bargaining.
I am so pleased that the Monkey Cage is sponsoring this rich discussion of gendered citation issues in the academic profession. When I served as Chair of the Status of Women Committee for the International Studies Association (ISA), I organized a roundtable on the topic at the 2012 ISA conference. My motivation for this roundtable stemmed from my familiarity with Marianne Ferber’s work on gender, citations and networking. As a labor economist, her first study in 1986 examined citations in a leading journal in her field (Journal of Economic Literature); she found that scholars were more likely to cite the research of other scholars who shared their biological sex.
Ferber’s 1988 study expanded her data to include articles published in U.S. and Canadian journals in economics, mathematics, psychology and sociology. The expanded dataset confirmed her original findings that male scholars were more likely to cite the research of other male scholars than to cite work by female scholars. She also found important differences across disciplines; the gender citation gap was smallest in disciplines with few female scholars (mathematics) or in disciplines with a critical mass of female scholars (psychology). Gendered gaps in citation patterns emerge in disciplines where women scholars constitute a sizable minority. This is certainly relevant for political science, a field where 28 percent of faculty positions in the U.S. are held by women, despite 45 percent of bachelor’s degrees and 40 percent of doctoral degrees held by females (what is called the “leaky pipeline”).
Samantha Lange, Holly Brus and I replicated Ferber’s research design by focusing on publications in two political science journals: International Studies Quarterly (ISQ) and International Studies Perspectives (ISP). Ferber’s design was innovative because she paired the biological sex of the author(s) of a journal article with the biological sex of the authors of each article or book in the article’s bibliography. This design allows us to capture the process of individual authors’ decisions about who to cite, or the dynamics of “who is citing whom”. Analyzing all articles published in ISQ and ISP in 2005, my coauthors and I find that male authors are three times more likely to cite the work of other men than to cite articles authored by women (see also the findings of Maliniak, Powers, and Walter). Working with women coauthors does not improve this situation because when women coauthor with men, their articles adopt citation practices that are similar to all male authored teams.
While women constitute 28 percent of faculty in political science, there is considerable variation in female participation rates across different areas of the discipline. Within the International Studies Association, an organization with 40 percent female members, the percentage of women who belong to ISA’s 33 organized sections ranges from 20 percent to 97 percent. Michelle Dion and I explored this variance further by comparing four smaller organizations within political science: the Society for Political Methodology (Polmeth), the Peace Science Society (PSS), the State Politics and Policy (SPP) group, and the International Political Economy Society (IPES). The figure above shows gender differences in paper or poster presentations at the organizations’ annual conferences, revealing an 11 percent gender gap across these groups. The State Politics and Policy emphasis on substantive issues in American politics attracts a higher percentage of female presenters than conferences focused on methodological issues in the field (Polmeth).
There is also temporal variation for women’s participation in these four organizations (see figure below). Some organizations like Peace Science have more than doubled the number of female participants over the past two decades, while other organizations’ female participation rates are static. Michelle and I found that the number of women participating in the organizations was related to citation patterns in these research areas. Examining journals sponsored by two of these organizations (Polmeth: Political Analysis; PSS: Conflict Management and Peace Science, Journal of Conflict Resolution), we continued to find evidence for a gender gap in citations, with male authors much more likely to cite the work of male scholars compared with their female peers.
However, we also found that the total number of citations to work by women was higher as a proportion of the total citations for journals associated with the Peace Science Society (6 percent) compared with the Political Methodology journal (2.9 percent). Where women’s work is visible in smaller academic groups, this translates into a higher frequency of citations to work by female scholars. One thing we must consider in developing strategies for improving the gender gap in citations is to improve the descriptive representation of women in various disciplines and sub-disciplinary areas. When women have ample opportunities to present their work at conferences and when women achieve milestones within the organizations (e.g. career awards, the presidency, being featured speakers), this raises the community’s awareness about women’s research.
There are many potential causes of the gendered citation gap including implicit biases, gendered networking patterns, and the lack of self-citation by female scholars. Another cause that I have described here is the overall representation of women in an academic field. When women constitute a critical mass of scholars in a field, the gender gap in citations diminishes. The loss of women scholars at multiple stages of the pipeline needs to be understood and addressed. We need to collect panel data on scholars as they move through their academic careers. While aggregate data reveal a problem with the retention of women in political science and other fields, we do not understand the motives behind individual decisions for exit. These decisions have important consequences for the sociological underpinnings of our discipline. While we have made considerable progress, we need to continue to “lean in” and “push back” to address the citation gender gap and other gender issues in academia.