No surprise—women are undervalued and under represented in the sciences: an overview of the presentation on gender bias in the sciences by Corinne Moss-Racusin, Ph.D.
Dr. Moss-Racusin completed her Ph.D at Rutgers, and did post-doctoral work at Yale. She is currently an Assistant Professor at Skidmore. Her work focuses on understanding and ameliorating inequality within institutions. She focuses on the ways in which stereotypes shape behavior, social judgments, and self-regulation, and how these in turn impact intergroup relations and the equitable treatment of stigmatized group members within institutions.
Dr. Moss-Racusin was the lead author on the landmark paper describing a study documenting gender bias among male and female science faculty at research institutions (Moss-Racusin, C. A., Dovidio, J. F., Brescoll, V. L., Graham, M., & Handelsman, J. Science faculty’s subtle gender biases favor male students. Proceedings of the National Academy of Sciences, 109, 16474-16479). This paper was the first documented study that demonstrated that science faculties have a gender bias when evaluating CVs. The findings have generated a great deal of discussion in the scientific community, and prompted an opinion piece by Stanford’s Dr. Jennifer Raymond in Nature as part of a special women in science issue.
It also was the basis for a special symposium on gender bias at the 2013 Society for Neuroscience Annual Meeting.
Dr. Moss-Racusin has been visiting a number of institutions to discuss her findings, including a standing-room crowd at Stanford’s Clark Auditorium on Wednesday, January 15. Her presentation largely focused on her research findings, which clearly demonstrate the existence of gender disparity in the STEM (Science, Technology, Engineering and Math) fields. The last 10 Nobel laureates in physics were male, and overall only 2 women out of 193 have achieved this honor in physics. Four of 164 Nobel laureates in Chemistry have been women, an average across both fields of less than 2%. In addition, faculties in STEM fields are predominately male across higher education.
Her research set out to isolate the factors that might cause this disparity. She listed some possible factors:
- Intrinsic Ability—are women less capable?
- Lifestyle Choices—do women prefer other fields, or choose family over the rigors of science?
- Gender Bias—are male and female science students evaluated, mentored and advanced equally, or are there gender stereotypes in STEM?
Her research found little to support the first hypothesis, and while women might prefer other fields to some extent, it’s hard to separate this choice from possible underlying bias.
Dr. Moss-Racusin’s research focused on the third possibility, that Gender Bias is the main factor driving gender disparity in the STEM fields. In general, she believes Americans hold these cultural attitudes:
- We think of women as “communal,” that is: emotional, modest, supportive, and focused on family.
- In contrast we think of men as “agentic,” analytic, risk taking, self-promoting, career-oriented. She mentioned a study in which seven-year olds were asked to “draw a scientist.” Overwhelmingly, the drawings depicted older white men.
This cultural bias, Dr. Moss-Racusin believes, is at the root of a subtle but pervasive gender bias in evaluating individuals. For example, letters of recommendation for female candidates in relation to their male counterparts tend to be shorter, express more doubts, stress teaching and mentoring skills, and minimize accomplishments. This leads to a self-perpetuating minimization of female accomplishments, as they receive lower entry-level positions, smaller labs, less mentoring, etc.
The 2012 Study
To test this hypothesis, Dr. Moss-Racusin developed a blind study. She recruited 127 faculty members across the STEM fields to evaluate resumes. Participants read and rated lab manager applications. Each participant saw the same resumes, but half of the resumes were given female names, half male names. The names were reversed for the other group. That is, the same CV might have the name “John X” or “Jennifer Y.” Participants were asked to rate the resume as to competence, how likely they would be to hire the candidate; what salary the candidate should be given; and how likely they would be to mentor the candidate. Without gender bias, the same resume should show approximately the same rating regardless of the gender of the applicant.
The study revealed a pervasive underlying bias shared by both male and female faculty participants. In fact, the results showed that the candidates designated as female were rated 67% as competent as the same candidate designated as male. The starting salary differential was $4,000/year. Below is a chart that highlights these differences:
These results were the first scientific evidence of broad gender bias across the study. All the faculty (male and female) preferred the male student relative to the identical female student. The female students were simply viewed as less competent, despite the identical CV.
While these results are perhaps not surprising, they do point to a serious and often disregarded problem in the sciences, especially as recent studies by the Bureau of Labor Statistics, the National Science Foundation and the Bay Area Economic Council Economic Institute (csedweek.org/files/CSEdWeekPitch.ppt) predict a shortage of approximately 1,000,000 candidates for future STEM jobs.
How Can We Address This Bias?
Dr. Moss-Racusin then discussed what measures might address and ameliorate the persistent gender bias that her research has highlighted. Her first suggestion was to discuss the issue of gender bias openly. She noted that bias is often strongest among people who believe they are unbiased. If we acknowledge that we share a cultural bias against women in the sciences, this can start the process of change.
In addition, changes to hiring and mentoring policies can acknowledge and attempt to adjust for bias. This includes posting transparent advisory schedules that guarantee equal access to mentor’s time, assignment to secondary mentors when necessary, and changing hiring policies. For hiring, Dr. Moss-Racusin suggested developing clear evaluation criteria: what are we looking for and how will we know if we found it? Then ask individuals to rate CVs according to these criteria independently first, before committee discussion. She also discussed funding linked to bias training, just as is currently done with research ethics. Dr. Moss-Racusin has developed a “Scientific Diversity” course, which shows some promise. Those interested in offering this course should contact Dr. Moss-Racusin, firstname.lastname@example.org.
In conclusion, Dr. Moss-Racusin noted that women’s participation has been flattening or dropping in STEM fields since 2000, especially in computer science. She believes that gender bias is subtle and pervasive, and that the problem won’t fix itself. Diverse groups often do better work, and gender parity is in the best interest of national competitiveness. She hopes that her research can assist in raising awareness and starting amelioration in the STEM fields.
My Comments on the Presentation
My personal reaction to Dr. Moss-Racusin’s presentation was that it was terrific to have definitive evidence of gender bias through the cleverly designed blind resume study. But addressing the problem seems a lot trickier. One questioner asked Dr. Moss-Racusin if she had a lot of money to throw at the problem, how would she address it. Her reliance for the most part would be on discussion and diversity training. While I think discussion, particularly, is important, the fact that the study showed that CVs themselves were viewed differently based solely on the gender of the candidate remains problematic, as does the cultural perception that STEM fields are more suited to men.
My background includes almost 30 years in education and training, largely in the corporate sector, but also at Hass Business School, UC Berkeley Extension and Cal State Sacramento. Based on this experience, I feel pretty skeptical about the effectiveness of the Scientific Diversity course Dr. Racusin mentioned. This course was piloted at seven universities in 2013 as a summer course, and was evaluated through pre- and post-tests. That is, participants were given a questionnaire asking them about their level of awareness, hopefulness about change, etc. about gender bias both before and after the course. Their answers to these questions on a scale of 1-10, and were compared before and after the course. The improvement in scores in these areas, even by self-reporting, was minimal (I think fewer than 5 percentage points).
I know from my own work that measurement by pre- and post-tests is designed to show improvement. In my field, we were evaluated (and often paid) based on the degree of improvement as measured in pre- and post- tests, and our tests were carefully crafted to always show improvement after taking a course. A much more meaningful evaluation, and one that we often used if funds permitted, is to sample performance before and at intervals after the course. For example, would the participants, evaluating blind resumes coded by gender (as in Dr. Moss-Racusin’s 2012 study) before the course and then six-months after the course, demonstrate less gender bias? My intuition is that improvement would be even smaller by this measure. My own experience with corporate sexual harassment training is that it had very little effect on underlying, unconscious bias.
I also worry about well-meaning academic measures to mitigate disparity, and suggest as a discussion point, Kurt Vonnegut's short story, "Harrison Bergeron," which provides an imaginary but chilling glimpse of what institutionally mandated parity might look like.
My Alternative Suggestion
Rather than courses or mandates, if I had a mandate and funding to counteract pervasive cultural bias, I would use it to design a large-scale advertising campaign that showed women and girls solving problems, working in parity with male counterparts, highlighting the achievements of women in science. I would develop a set of TV dramas with women scientists, programmers, surgeons, and physics prodigies as protagonists (just as we now have women FBI agents and police). The shows would feature smart, capable (and of course, beautiful—it’s still TV after all) women excited about their work, making key discoveries, glamorizing the STEM fields. I think this would have a real impact on mitigating the underlying bias against women in the sciences. Women have made great strides in law and we now have many images of successful women lawyers as part of the cultural landscape. We need to do the same for the STEM fields.