GENDER BIAS IN LEGISLATIVE STUDIES?

I was fortunate to begin my career as a PhD student at the University of Rochester. (I was actually lucky that my first-choice school offered to put me on its waiting list, stating that I would be accepted if any of their admits were drafted and unable to attend.) Bill Riker was establishing a serious graduate program at Rochester, and he looked for applicants that the top-ranked schools might overlook. Hence, at Rochester, often underrepresented groups (e.g., women) were overrepresented. Bill’s merit-based attitudes were shared by the faculty—women and other underrepresented groups were not treated as second-class citizens. Consequently, Rochester produced several well-known women scholars. Dick Fenno’s presence in the department ensured that some of them would be Congress scholars, including Barbara Sinclair, Wendy Schiller, Linda Fowler, Diana Evans, and Christine DeGregorio, and in comparative legislatures, Gail McElroy and Tanya Bagashka.

These Rochester alumnae all started their careers as legislative scholars and generally continued to publish exclusively or primarily in that subfield. However, many of us either have interests in more than one subfield or we started publishing in one subfield and moved to another. I went to Rochester to study game theory and was Dick McKelvey’s first PhD student; we learned the dissertation process together—he received his PhD the year I was writing mine. I became more deeply interested in substantive questions by coauthoring with colleagues in my first job. Although I wrote occasionally on Congress, it was not until 1995 that all of my work was on legislatures or legislators, and it has become increasingly institutional. As a Rochester graduate, the legislative field was welcoming to me. Several of my cohort were legislative scholars, and after I returned to teach at Rochester, many of my students became legislative scholars. I have served on paper, book, and now career award committees for the section. It seems to me to be an exemplary section in terms of openness to women scholars, including in leadership selection.

Thus, I was surprised when Gisela Sin and Laurel Harbridge told me that our section was the third lowest in female membership among the 43 APSA sections (Roberts 2018). It would be interesting to break down female–male representation in the sections by PhD-year cohorts and to construct an individual-level dataset including variables to show sections to which APSA members submitted papers and presented in by year, along with their section memberships. That might allow us to distinguish among various explanations for our relative gender imbalance. One possibility is that younger scholars are more likely to be female and to publish on topics that fall into more than one APSA section. If so, are legislative-“plus” scholars disproportionately choosing to be involved in other sections rather than ours? That is, are we losing the competition to attract younger scholars, which incidentally is showing up in our overall male–female ratio? If so, this is important because young scholars, male as well as female, are the future of our section. Or are we losing more young female scholars than young male scholars? Or is the problem less about age cohorts and more about gender? The answer to these and other questions may lead us to different solutions about addressing the problem.

I asked several untenured female scholars whose work fits in our section and at least one other section about their perception of our section and the other(s). One said that the Congress field and, hence, the legislative section had a reputation as a “boys club” more than the other. Another said that the other field had an open yearly conference to present papers and seemed to be an especially friendly and mentoring section. Congress and History has been a wonderful resource for Congress scholars, but our section does not have an annual legislative conference. Moreover, of course, Congress and History does not include comparative legislatures and US state legislatures. I think both the annual methods and the state politics conferences have been helpful in attracting and mentoring young scholars. The latter, in particular, seems to have more female participation. The opportunity to attend a section conference is a valuable selective benefit, whereas the other benefit—receiving the section journal—has decreased in value because of the easy availability of online articles through university subscriptions.

Also, although the section has tried in terms of panels, officers, and awards to include comparative legislative as well as state legislative scholars, researchers in those fields are still less likely to belong to our section than they should be. Both of those groups, I suspect, have a higher proportion of women. I think that as a section, we would be enriched by their participation, both male and female.

Gisela and Laurel also pointed out that in the past two years of Legislative Studies Quarterly articles, 18% were authored by women (solo or team), 59% by men, and 23% by mixed-gender teams. I compared that with the report I received for the American Political Science Review (APSR) editorial board: the APSR’s comparable percentages are 12%, 69%, and 19%. Therefore, we do not fare as badly as the APSR in women’s representation. Yet, I think a better comparison would be with other section journals. Of course, what is truly important is the equal treatment of identifiable subgroups, such as women, and the publication percentages cannot speak directly to that. Although our reviewing is blind, many of us sometimes can guess the authorship or have actually seen the manuscript presented at a conference. Furthermore, I suspect that if a reviewer googled the title, the paper often would come up as a previously presented conference paper. So, it is an interesting question about how “blind” our review process really is.

Identifying bias is typically difficult. Gisela and Laurel asked me to comment on my early experiences in a male-dominated field. Of course, bias is easily identifiable if it is overt. For example, I asked a colleague and good friend why a new hire had been assigned to teach a grad course, whereas I had taught there (not Rochester) for several years but had not yet been given a grad seminar. He said that they did not think that, as a woman, I would be able to handle a grad course filled with male students. I did start teaching grad courses shortly thereafter.

The problem is identifying bias that is not overt—did bias play a role in a journal rejection or not being invited to a conference? When I had a manuscript rejection, I assumed the fault was mine—that my actual work would be evaluated only for its content, not in relationship to my gender. It is only recently that experiments have been used to identify the effects of gender bias in academia. It is especially worrisome to find notable gender bias in evaluating the quality of research, which in turn affects publication and career prospects. Experimental research, for example, has found that randomizing the gender of the “author” on an abstract affects—negatively for women—a PhD student’s evaluation of the research and their interest in collaborating with the author. Knobloch-Westerwick, Glynn, and Huge (2014) found this to be true for “masculine” topics—that is, what we study, unless the research is on women, families, and children. Two articles by economists in the American Economic Review found that women are disadvantaged in attaining tenure in top-rated economics departments because men are given significantly more credit than women for coauthorship on a mixed-gender team publication (Sarsons 2017). Research also found that women are about as likely as men to manifest gender bias disadvantaging women—so, the problem is “us” collectively. Addressing the problem is difficult. One solution designed to take childbearing out of the equation had the opposite effect of what was intended. Gender-neutral tenure-clock stopping increased the likelihood that a man gained tenure while decreasing the likelihood that a woman would (Antecol, Bedard, and Stearns 2018). With regard to Sarsons’ work on team authorship, it has been suggested that changing our authorship pattern from alphabetical last names to an ordering that reflects each author’s contribution to the work might be helpful. We need to think creatively about ways to reduce gender and other forms of bias in our profession. Top orchestras used to be overwhelmingly male, because—we were told—men were simply “better.” This changed when auditions using screens that concealed the gender of the musician showed otherwise and created orchestras that now are close to gender parity.

The problem is identifying bias that is not overt—did bias play a role in a journal rejection or not being invited to a conference? When I had a manuscript rejection, I assumed the fault was mine—that my actual work would be evaluated only for its content, not in relationship to my gender.