Traditional Focus on Sex Discrimination in Science Is Misplaced

• By Ars Technica
• February 15, 2011  | 

By Kate Shaw, Ars Technica
Today, more than half of all PhDs in the life sciences are awarded to women, compared to a measly 13 percent bestowed upon women in 1970. However, women still lag far behind men in full professorships and tenure track positions in math-intensive fields.
Despite claims that this disparity is due to discrimination against women in the processes of publication, grant review, interviewing, and hiring, a review in PNAS last week, written by Stephen Ceci and Wendy Williams of Cornell University, finds that there is actually little evidence for sex discrimination in these areas, and concludes that women’s underrepresentation stems from other causes.

Is it harder for women to publish?

Getting research published is a must for scientists, and is essential for getting hired and moving up the ranks in all scientific professions. Critics have claimed that men have an advantage in the reviewing and publishing processes, and that this bias may account for the dearth of females in tenured positions. However, after reviewing several studies in this area, Ceci and Williams conclude that this just doesn’t seem to be the case. Studies of publication rates in Nature Neuroscience, Cortex, and Journal of Biogeography, among others, found no evidence of sex discrimination.
When men and women with similar resources are compared, there is no evidence for publication differences between the sexes. However, there are some factors that affect women at a disproportionately high rate and may cause biases in the publication process.
For example, women are more likely to work at teaching-intensive colleges and therefore lack the time and resources to produce frequent and high-quality research for publication. When the type of institution, the scientist’s funding, the teaching load, and the research assistance are taken into account, there is no difference in publication rates for men and women. It seems that the critical factor isn’t sex, but instead access to resources, an area in which women may lag behind men.

Are women at a disadvantage when applying for grants?

Another commonly-cited issue is that it is harder for female scientists to get funding for their work than it is for male scientists. Some studies, such as a very influential 1997 Nature publication by Wenneras and Wold, have found that grant review panels are more likely to fund males over females. This study found that women needed to be “2.5 times more productive” than men to be funded by the Swedish Medical Research Council in 1995.
However, this study has been challenged based on methodological and conceptual issues, and further studies have mostly been unable to replicate this level of bias. In fact, where biases have been found, advantages often go to the women; a 1996 study of females funded by a UK panel found they have published just 11.2 papers on average, while the funded males had published an average of 13.8 papers. Large scale analyses of grant review in the past 30 years at NSF, NIH, the US Department of Agriculture, and the Australian Research Council have not found any evidence of sex discrimination.
Research does suggest that, before the 1980s, it was more difficult females to get grants than it was for males. However, most research agrees that the playing field has evened out in terms of funding in the last few decades.

Does hiring occur without regard to gender?

Finally, Ceci and Williams examined the interviewing and hiring process at research institutions. A very famous 1999 study distributed mock CVs to 238 psychologists who were reviewing possible hires for assistant professor and tenure track positions. For the assistant professor job, the reviewers tended to rate CVs with male names more highly than identical CVs with female names (although this effect disappeared for the tenure track job). This finding, as well as similar results in other studies, suggested to many that females may be at a disadvantage when applying for some scientific positions.
However, since 2000, there is little evidence that females face more hurdles than men in the hiring process. In fact, some studies have shown that women get interviewed and are offered tenure-track jobs at a slightly higher rate than males. What is evident, however, is that females tend to make different choices than men when applying for and accepting jobs. Here, Ceci and Williams argue, is where much of the scientific gender gap arises.
About 80 percent of both male and female graduate students believe that working full-time is “important” or “extremely important.” However, nearly a third of women believe that working part-time for a period is “important” or “somewhat important,” compared to just 9 percent of men. In the UK, females are almost twice as likely as men to work part-time for at least some length of time. So, while females have similar aspirations as men, they seem to make different choices when it comes to employment.

Asking different questions

Instead of debating whether women are being discriminated against in publication, grant review, and hiring, Ceci and Williams argue that we should concentrate on more current causes of female underrepresentation in the sciences. The important question to address now is whether women on the academic track are making personal choices that promote happiness and satisfaction, or whether their careers are constrained by biology and societal obligations.
The authors suggest that there are three main contributors to this unde-representation: career preferences, family choices, and ability differences. Women tend to chose careers in which they are “helping people,” such as working at teaching-intensive institutions, at a greater rate than men do. Additionally, fertility and lifestyle choices affect women’s careers at a disproportionately high rate when compared to men.  In terms of ability differences, far fewer women than men appear in the top one percent of standardized math results, such as SAT and GRE scores.  The authors cite this last fact without venturing into the complicated reasons behind the phenomenon, except for stating that it is “potentially influenced by both socialization and biology.”
So, women in scientific fields — especially math-intensive ones — seem to be at a disadvantage due both to free choices and various constraints.  According to the authors, however, discrimination in publication, grant review, and hiring do not seem to be among those constraints.
As a start, Ceci and Williams suggest exposing young women to successful role models in science and math-intensive fields and giving them more information about career opportunities. Furthermore, they advocate changing the tenure system slightly so that there are fewer disincentives for women to have families and children. As a model, the authors cite UC Berkeley’s “Family Edge” program, which provides child care and urges reviewers to ignore family-related gaps in employment and productivity.
While more work clearly needs to be done on clarifying the causes of women’s underrepresentation in the sciences (especially in math-intensive fields) the review strongly suggests that discrimination isn’t the cause, and that we need to “redirect our energies” toward more pertinent questions.
Image: Flickr/juan.barredo.
Source: Ars Technica.
Citation: “Understanding current causes of women’s underrepresentation in science.” Stephen J. Ceci and Wendy M. Williams. Proceedings of the National Academies of Science, published online Feb. 7, 2011. DOI: 10.1073/pnas.1014871108