To become a professor, one must first spend years working to earn a PhD. The coursework that students pursue as part of their doctoral programs serves as the foundation for their academic careers. Those courses don’t just convey knowledge; they also socialize students into the culture and norms of academia.
Given the importance of coursework, we were curious about the readings that professors in our field of political science assign to their graduate students. We were particularly interested in what the proportion of female-written readings might be. Across academia, studies show that academic articles written by women are less likely to be cited in other published work.
In our research with co-authors Hannah June Kim and Philippe Meister, we found that professors assign few readings written by women. We found that to be true both for syllabi, which outline a course’s required readings, and for reading lists for comprehensive exams. These exams are a key requirement for many PhD programs and often require students to demonstrate competence in and intellectually engage with core readings in their area of research.
Our findings build on and confirm the results of a few recent studies of PhD syllabi in the political science subfields of American politics and international relations. However, our project is by far the largest and most comprehensive data set of graduate readings to date, to the best of our knowledge.
How did we study assigned readings?
We created the Graduate Assignments DataSet (GRADS) by collecting 840 course syllabi and 65 mandatory comprehensive exam reading lists from 94 political science PhD programs in the U.S. in the fall of 2016.
Turning syllabi into analyzable data involved many steps. Citations in syllabi and reading lists are often incomplete and poorly formatted. By using a combination of manual and automated coding methods and by overseeing a large and dedicated team, we extracted 88,673 assigned readings from syllabi and reading lists.
We then coded the authors’ genders based on their first names. First, we developed an extensive list of the names and genders of specific people with hard-to-code names — for instance, “Terry” and “Lee” each has about a 50 percent chance of being female. We checked all of our authors against that list. Then, we coded the rest of the authors using several data sets of the genders typically associated with first names.
The work of female political scientists is underrepresented in U.S. PhD programs’ assignments
Our research indicates that works by women are underrepresented in PhD courses and comprehensive exam reading lists in political science. In every subfield of political science, the percentage of readings with a female first author is lower than several different benchmarks we considered. Women’s names appear less frequently in syllabi and reading lists than in the tables of contents of top scholarly journals — where women are already underrepresented. The gap is even bigger if we consider the difference between the percentage of assigned readings written by women and women’s representation in faculty rosters or the rosters of graduate programs.
Who assigns and who doesn’t assign research written by women?
The instructor’s gender matters. While about 15 percent of the readings assigned by men have female first authors, about 24 percent of the readings assigned by women do.
Other instructor characteristics matter, too. Among women, age matters; older female faculty assign more female-written work than do younger women. Male faculty of color assign substantially more female-written work than do their white male peers. Also, we find evidence that more female-written work is assigned by instructors in environments where women make up a larger proportion of the faculty.
A faculty member’s colleagues matter. When departments hire more women, all faculty become more likely to assign readings written by women. Moving from a department that has 10 percent female faculty to a department that has a 50-50 gender balance nearly doubles the rate at which men assign work written by women. We suspect this phenomenon occurs because in gender-balanced departments, both men and women are more likely to hear about new research published by women.
An opportunity for change
Political scientists now have access to many tools that allow scholars to be more inclusive, both in teaching and beyond the classroom. The Women Also Know Stuff initiative maintains a database of female scholars who are open to sharing their expertise with other scholars, policymakers, journalists and others. Instructors can search our GRADS data set to see some of the female-authored research that other faculty have assigned. Instructors can also assess the gender representation of their syllabi using Jane Lawrence Sumner’s Gender Balance Assessment Tool.
Previously in this series: