Women in International Relations
1 Daniel Maliniak, Amy Oakes, Susan Peterson, Michael J. Tierney
Women now receive political science degrees in record numbers, but female representation among political science faculty still lags behind that in many other disciplines. Only 26 percent of the 13,000 political science professors in the United
2States today are women. According to our recent survey of international relations (IR) faculty in the United States—the 2006 Teaching, Research, and International Politics
(TRIP) Survey—women comprise an even smaller proportion of IR scholars: 77 percent
3of the IR faculty respondents are men, while only 23 percent are women. Even more
than their counterparts in the wider field of political science, women in IR tend to be more junior and less likely to hold tenure than their male colleagues. Women comprise a minority at every level of the profession, but they are most scarce at the full professor level: only 17 percent of political science professors and 14 percent of IR professors are
Women may be underrepresented in the profession and trail their male colleagues because they see the world differently; they may see the world differently because of their minority status within the discipline; or the causal arrow may run in both directions. Many feminist scholars contend that gender subordination explains significant differences in worldview between men and women. Other scholars suggest that the content of women’s scholarship contributes to their marginalization within the profession:
female political scientists adopt methods and choose topics that are not considered to be
5the best or most rigorous types of research by the editors of leading journals. As a result,
―women’s publishing opportunities may be restricted, or ghettoized, to specific and
6 gendered domains.‖
Regardless of the cause, the 2006 TRIP survey reveals important dissimilarities between men and women in their status in the profession, approaches to teaching and scholarship, and views on the discipline. In addition to highlighting women’s minority
status within the profession, and particularly within its upper ranks, our survey of IR faculty indicates that women research and teach different topics and in noticeably different ways than their male counterparts. Women are more likely to study transnational actors, international organizations, and non-governmental organizations, while men are more likely to study U.S. foreign policy and international security. Women study Sub-Saharan Africa and Latin America; men focus on the United States and Europe. Women are more likely to describe themselves as constructivists who focus on the role of ideas and identities as explanations of world politics, while men are more likely than women to be realists who assume an anarchic international system and focus on the effects of military capabilities held by states.
To a significant extent, women are still second-class citizens within the IR profession, and their research and teaching differ in important ways from that of male IR scholars. At the same time, however, our survey reveals important similarities between men and women. Men and women graduate from many of the same Ph.D. programs. Both male and female IR specialists overwhelmingly describe their research as positivist and qualitative, although women employ qualitative methods slightly more often and are less likely than men to describe their research as positivist. Large percentages of both men and women also report subscribing to liberal IR theory.
In short, while women’s research and teaching tend to focus more heavily on topics and regions outside the mainstream of the field and use non-traditional theoretical tools, male and female IR scholars also have much in common. Some apparent differences, moreover, are better explained by age or life experiences than by differences in gender. In the following pages, we paint a picture of women in IR at this particular moment in the history of the discipline. Although our data do not enable us to explain every divergence, throughout the paper we turn to existing literature to speculate on some of the reasons for the patterns we observe. The data used to discern these patterns is drawn from the 2006 TRIP survey of 1,112 IR faculty throughout the United States. For the survey, we attempted to identify and survey all faculty members in four-year colleges and universities in the United States who do research in the sub-field of international relations or who teach courses on international relations. Forty-one percent of those identified ultimately responded to our 83-question survey on teaching, research, the discipline, and foreign policy. Below, we highlight some of those patterns in professional status, scholarship, and teaching.
WOMEN AND THE PROFESSION
Women occupy a different place within the profession than their male colleagues beginning with their education and continuing throughout their careers. Women are more likely than men to have attended and then teach at a liberal arts college. They favor different Ph.D. programs. Women IR faculty members tend to be younger and more junior than the men in the profession, and their work is less recognized and valued by the discipline as a whole.
Harvard leads the list for undergraduate training of both men and women, but the similarities end there. Forty-eight percent of women received their B.A.s from American Ph.D.-granting institutions, compared with 61 percent of men who received B.A.s at similar institutions. More women received their B.A.s at liberal arts colleges (20 percent), Master’s-granting institutions (12 percent), and foreign universities (17 percent). A smaller percentage of men got degrees from each of those categories (16 percent, 9 percent, and 11 percent, respectively).
Where do women get their Ph.D.s? They attend many of the same institutions as their male colleagues, but in different numbers. Five percent of men, but only two percent of women received their Ph.D.s at Harvard, the top ranked program in our
7 As the following list shows, female IR scholars are more likely to have attended survey.
Columbia, Virginia, Cornell, Stanford, Ohio State, UCLA, UCSD, Chicago, or Maryland, while male colleagues are more likely to have attended Harvard, Michigan, Berkeley, MIT, Wisconsin, or Yale.
[Figure 1 about here]
Gender differences persist as women move through the profession. The average female IR scholar is almost four years younger and received her Ph.D. five years later than her male colleagues. This reflects the higher percentages of women at the assistant and associate professor ranks compared to men. Forty-one percent of the women who participated in the survey are assistant professors, compared to only 27 percent of the men. Fifty-three percent of women hold tenure, compared to 62 percent of their male
colleagues. Thirty-three percent of female respondents were associate professors, compared to 25 percent of men. This means that strikingly few women have reached the top of the profession—only 20 percent of women are full professors, compared to 37
percent of men, and women comprise only 14 percent of all full professors of IR.
As a group, women do not have the same status within the profession as their male colleagues. First, they are less likely to teach at research universities. More women IR scholars (48 percent) than men (39 percent) teach at liberal arts colleges or universities without political science Ph.D. programs.
[Figure 2 about here]
This finding is somewhat at odds with results for the field of political science as a whole. The 2001-2002 American Political Science Association (APSA) Survey of Departments found that 30.1 percent of female political scientists in the United States are employed in B.A. political science programs, 25.8 percent in Master's departments, and 44.1percent in
8 Ph.D. programs.
Second, scholarship by women is generally not as well recognized as that of male IR scholars. Only two women—J. Ann Tickner (21) and Kathryn Sikkink (25)—make
the list of the top 25 scholars whose work has had the greatest impact on the field of IR in
9the last 20 years. This list also reveals other interesting gender differences. Female respondents were less likely than men to include leading realist scholars, such as Kenneth Waltz (-8 percent), John Mearsheimer (-4 percent), and Hans Morgenthau (-4 percent), or influential formal modelers, such as James Fearon (-7 percent) and Bruce Bueno de
Mesquita (-5 percent). They were more likely than their male colleagues to value the contributions of female scholars, including Ann Tickner (+8 percent), Cynthia Enloe (+3 percent), and Martha Finnemore (+3 percent).
The differences are more striking when one compares rankings of scholars whose work is considered ―the most interesting.‖ Of the top ten scholars identified by women as
producing the most interesting research in recent years, five are women. In the list of the top ten scholars whose work is most interesting to male scholars, in contrast, there was only one woman. Martha Finnemore was the only female scholar to be ranked in the top ten by both male and female scholars, but fully 10 percent more women than men
10 It is possible that these results included her in their list, making her their top choice.
could be explained simply by differences in paradigmatic focus and methodology. As Figure 3 shows, however, even after controlling for age and paradigm, gender is a significant predictor of whether a respondent finds the work of women scholars interesting and influential in his or her own research.
[Figure 3 about here]
These results appear to mirror findings from the wider discipline of political science. Natalie Masuoka, Bernard Grofman, and Scott Feld find that women are significantly underrepresented on the list of the 400 most frequently cited political scientists, relative
11to their representation in the profession.
Why do women IR scholars appear to lag behind their male colleagues in rank and recognition? The research that we have conducted to date cannot answer this
question definitively, but existing literature provides a wealth of potential explanations. One oft-cited reason is the differential effects on men and women of family responsibilities. Women are more likely than men to interrupt their careers to raise
12children, and they are more likely to become primary caregivers for children. Women
who have children as graduate students may be more likely to drop out of school, take longer to complete their studies, fail to get a job or succeed at a job than are men who
13have children while they are still in school. This hypothesis is supported by Viki Hesli
and Barbara Burrell’s finding that women have higher rates of attrition than their male
A number of other explanations have been offered. One reason women lag behind men in terms of promotion and tenure may be differences in productivity. Women are more likely to hold the ranks of assistant and associate professor, according to this argument, because they produce fewer publications on average than their male
15counterparts. Although the differences are narrower than they once were, men still out-
16publish women in political science journals by a ratio of two to one. This gap persists
even when controlling for reputation of Ph.D.-granting institution, rank, institutional type,
17professional age, and marital status. And it may be larger in international relations than
in other subfields of political science. Cheryl Young finds that, between 1983 and 1994, women were less likely to publish in journals that focused on international relations (they were largely absent, for instance, from Journal of Conflict Resolution) than in journals
18that focus on political institutions (like Legislative Studies Quarterly). In Marijke
Breuning, Joseph Bredehoft, and Eugene Walton’s study of articles published in three top
tier IR journals, women were the first authors of 20.2 percent of articles published,
19 although they make up 31.8 percent of International Studies Association members.
Our own research provides mixed evidence for this argument. We supplement our faculty survey data with a new database of all international relations articles published in the top twelve peer-reviewed IR and political science journals from 1980 to
202007. Over the time period from 1980 to 2007, female IR scholars comprise only 17 percent of the author pool. However, if we limit our analysis to articles published from 2004-2007, a time period that corresponds more closely to our survey, then females make up 22 percent of the author pool – almost identical to the 23 percent of IR faculty who are
women. While the most recent time period is relatively short, these data suggest that the gender gap in publishing may be shrinking. At the same time, however, women are more likely than men to co-author articles. From 1980 to 2007, men published 87 percent of the single-authored articles appearing in these journals, while women authored only 13 percent of the articles. In the most recent years, 2004-2007, the gap narrowed, however: 80 percent of single-authored articles were published by men and 20 percent by women.
Women are significantly more likely than their male colleagues to collaborate in their research and they are disproportionately likely to co-author with men, rather than other women; however, they are far less likely to be listed as the first author on these articles. As Figure 4 illustrates, in fact, our database of IR articles shows that women’s
names appear in increasing percentages the farther removed one gets from the first author
21in multi-authored articles.
[Figure 4 about here]
Some analysts argue that the gender gap in publishing is the result, not of differences in productivity, but of differences among male and female political scientists
22in ambition, reputation, and merit. Others argue that women care more about advising,
23administrative work, and departmental committees than research. Women also are
more likely (65.8 percent) to devote eight or more hours per week to class preparation
24than men (59.1 percent). Some of this may be explained by the fact that fewer women teach at Ph.D.-granting institutions, and faculty at research institutions may spend less time preparing for class than faculty at baccalaureate- or master’s-granting institutions.
Nevertheless, higher percentages of men (41 percent) than women (33 percent) who were on the job market in 1996 said that scholarship was the most important attribute in getting a job, and higher percentages of women (42 percent) than men (32 percent) said that
25teaching was the most important requirement.
Another explanation may be isolation or discrimination. In this view, women are more likely to be excluded from social networks in departments dominated by men, to
26receive less institutional support, and to suffer from subtle or even overt discrimination.
Hesli and Burrell find that many more women than men moved jobs because of hostility among faculty members or superiors and that they found the work climate to be
27―chilly.‖ There are fewer professional networks for women or opportunities for
28mentoring relationships. One piece of evidence for this claim may be the lower
percentages of women who publish in edited volumes, since professional contacts are a key part of participating in these volumes. Indeed, edited volumes on international politics had among the lowest percentage of female authors—only 14 percent —even
though 26 percent of IR scholars in APSA are women, according to Lanethea Mathews
29 Another observable implication of the social network hypothesis and Kristi Anderson.
is that women should have fewer opportunities to co-author than men, all else equal; and that when they do co-author, they should be less likely to co-author with men. However, our analysis of 2,806 journal articles in the twelve leading peer reviewed IR journals over the past 28 years demonstrates that women are not isolated academically and are actually more likely than men to co-author articles. In fact, women are more likely to publish a co-authored article (285) than a single authored article (249). Men, on the other hand, are much more likely to publish single-authored articles (1657) than co-authored pieces (796). If gendered social networks are as insidious as these scholars suggest, then women should be expected to have more difficulty engaging in collaborative opportunities to publish. Of course, it is possible that women are simply co-authoring with other women and continue to be excluded from gendered social networks. To test this claim we calculated the probability that a co-authored article with two authors would have two females, two males, or a male and female co-author. If left to random chance, one would expect 3 percent of articles to be co-authored by two women, 69 percent by two men, and 28
30percent by a female and a male co-author. The social network hypothesis suggests that
the actual count of male-female collaboration should be lower and that male-male should be higher than random chance suggests. The evidence suggests exactly the opposite. Men are disproportionately likely to co-author with women (32 percent rather than 28 percent) and men are less likely to co-author with each other (65 percent rather than 69
31percent). So, while female faculty members may be excluded from golf and
departmental poker games, there is little evidence that they are excluded from