A new review (pdf) of how women fare in the sciences has provoked a lot of online response, thanks mainly to a pugnacious newspaper op-ed concluding “Academic science isn’t sexist” penned by the review co-authors Wendy Williams and Stephen Ceci.
The full review team – two psychologists and two economists (two women and two men) – focused on issues within the mathematical sciences comprising geo-science, engineering, economics, maths and computer science, and physical sciences, which they describe as “GEEMP” subjects. The 54-page paper, summarising findings from previous studies and presenting new calculations, is far-ranging, so I will focus on the claim that there are now far fewer obstacles for women getting and succeeding in jobs in these sciences.
The reviewed evidence suggests rates of graduate entry into PhD, and movement from PhD into Assistant Professor (equivalent to “lecturer” position in the UK), are comparable for women and men. Women still make up only 30 per cent of assistant professors (up from zero per cent in the 1970s), and rates are much lower in tenured roles, but the authors argue that this reflects the legacy of past barriers. Once women get into the field, Williams et al argue, “their progress resembles that of male GEEMP majors.”
The authors also point out that women are invited to interview and offered these sorts of jobs at a slightly higher rate than they are found in the applicant pool; women are also promoted within jobs at a similar rate to men. This would imply fair hiring practices in academic science, which contradicts plenty of psychology experiments that suggest identical candidates are often treated differently due to their gender. Among reasons for this lab-real-world disparity, Ceci and Williams report their own finding: in a lab, no gender discrimination was observed when a candidate presented as extremely able – as shortlisted candidates in academia invariably are – instead of having a mixed pedigree, as is common in other studies. The authors suggest that only when recruiters have insufficient relevant information will they start making assumptions based on stereotypes.
Making like-for-like comparisons, the authors also argue that today there is considerable gender parity in salaries, but differences remain particularly for full professor roles. This could reflect the often-longer tenure for male professors, differences in negotiation strategy, or enduring discrimination. One popular explanation for pay gaps is that women become penalised due to investment into child-rearing, but this doesn’t seem to adequately account for the differences. Another possibility would be if women face hurdles that prevent them performing in their jobs, and so find it harder to get the same compensation that men do.
One such hurdle would be the unfair awarding of research grants by funding bodies. However, analysis using the largest and most systematic data sets, such as the series of studies on awards made by the American National Institution of Health, strongly suggests gender neutrality. In one example, the large funder EMBA found its rate of awarding to female investigators unchanged when applicant identity was blinded. (Contrast this with the plight of black researchers, who are one third less likely to receive funding from the NIH.) There is some evidence of higher dollar rates of award to men, but this may reflect the higher proportion of men being very senior applicants.
Publication output is another performance measure that differs between the genders. Men on average publish more and are far more likely to be highly prolific, defined as three or more publications in two years. This gap has closed since the 1970s but remained steady in recent years. One hurdle here may be that women spend more time in mentoring and support; unpicking why will be tricky – women report enjoying this work more than men, but it is also a lower-status activity relative to research. Home-life pressure could be a factor: although women with or without children publish at a similar rate, men actually publish more after children. It’s also possible that women are impeded by less access to institutional support, such as facilities and scarce equipment. It’s also important to note that in the data presented, the publication review process appears fair, with little evidence of gender discrimination.
Overall, the review argues that once women enter a mathematically intense field, their chances resemble their male peers – not perfectly, but certainly closer than one might expect. Of course, in these fields just a quarter of bachelor degrees are awarded to women. The authors consider some reasons for this underrepresentation, including differences in interest, cognitive profile, teaching encouragement and stereotype threat.
Ceci and Williams clearly feel that their evidence is so at odds with the received wisdom of “academic sexism” that they were justified in penning their rejoinder. But with such a politically and culturally mediated concept as sexism, others could accept all this data and still reach different conclusions. My own would be to cautiously welcome the reviewed data as good news for women engaged in the mathematic sciences: institutionally there appear to be fewer barriers than one might imagine, giving women’s talent and effort a greater chance to shine. But the review is largely silent on whether the modern scientific workplace is free from informal, everyday sexism, or if women in science careers have to cope with this additional hassle. Mathematical sciences have clearly made progress towards equality, but it seems premature to say sexism is done.
Ceci, S. J., Ginther, D. K., Kahn, S., & Williams, W. M. (2014). Women in Academic Science A Changing Landscape. Psychological Science in the Public Interest, 15(3), 75-141.
Women’s true maths skills unlocked by pretending to be someone else
Girlie scientist role models could do more harm than good
Understanding why girls underperform at science
Hey girls: Science helps people!
Investigating the role of genes in boys’ and girls’ science ability
Post written by Alex Fradera (@alexfradera) for the BPS Research Digest.