The Shalala Report on Women in Science and Engineering

Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering, is a book-length study published last month by the National Academy of Sciences. The study was conducted by a committee appointed by the NAS (along with the National Academy of Engineering), and it concludes that women's underperformance in academic science and engineering relative to men is caused not by any innate differences between men and women but by subtle biases, and by barriers in the form of refusing to make science jobs more "woman friendly." The study is available online at http://darwin.nap.edu/books/0309100429/html/R7.html.

The study will, one hopes, be carefully dissected by experts, but I will be surprised if it stands up to expert scrutiny. Of the 18 members of the authorial committee, only one was a man, and only five were members of the National Academy of Sciences and only one was a member of the National Academy of Engineering. The one man, Robert J. Birgenau, although a distinguished physicist, happens to be the Chancellor of the University of California; for him to have dissented from the report would have condemned him to the same fate as Lawrence Summers, and swiftly too. The composition of the committee shows remarkable insensitivity. The theme of the report is the importance of unconscious bias with respect to issues of gender; did it not occur to the members and to the NAS and NAE that women might have unconscious biases regarding the reasons for the underperformance of women in science and engineering relative to men?

Economists, foremost among them Gary Becker, have done a great deal of work on issues of sex discrimination and women's career choices. The only economist on the committee, however, was Alice Rivlin, a specialist in the federal budget. Her Brookings website lists works such as "Restoring Fiscal Sanity," but lists no book or paper relating to gender issues.

The problem of the committee's biased makeup would be less serious if the report itself were transparent, but it is not. Although it cites a great many academic studies, it does not give the reader enough information about them (the methods used, the robustness of the findings, the quality of the journal in which the study was published, the professional standing of the authors, the reception of the study in the relevant professional community, etc.) to enable an evaluation. Some of the observations in the report suggest a distinct lack of academic rigor, as when it reports that Japanese schoolgirls do better on math tests than American schoolboys. Since there is much more job discrimination against women in Japan than in the United States (see, e.g., http://www.pbs.org/nbr/site/research/educators/060106_04c/), one would expect Beyond Bias and Barriers to predict that Japanese girls would do very poorly on math exams.

The report expresses particular concern with underperformance of black women in science and engineering, who underperform not only white men and women but also black men, even though black women generally outperform black men in educational attainment. This suggests that maleness rather than race explains differential performance in science. Other obvious objections to findings favored by this biased report are ignored. For example, there is a large difference in the average research output of male and female scientists. However, that difference is greatly diminished when the comparison is between male and female scientists in leading research universities; the obvious but unmentioned reason is that these universities are not discriminating in favor of women but merely applying the same high standards to both sexes. No one thinks that no female scientists are comparable to excellent male scientists; the issue is why there are so few female scientists in those top-tier universities. Another example: from the fact that the gender gap in science has diminished in recent decades one cannot reason, as the report does, that there are no genetic or otherwise innate differences in preferences or aptitudes for a scientific career. If a gender or racial gap is due partly to discrimination and partly to innate factors, then eliminating discrimination will narrow the gap, but will not eliminate it.

The study is notably deficient in comparisons between women in science and in other demanding occupations. Women do better, relative to men, in academic law than they do in academic science, mathematics, and engineering yet law is a highly demanding field. And how to explain their domination of primatology, a scientific field? The problems that women in science face, particularly in highly mathematized fields such as physics, in combining family and career seem no different from the problems they face in other fields inside and outside of science. If the report's ambitious program of making science woman-friendly, for example by more financial aid, day care, and the stretching out of degree programs, were extended--and why shouldn't it be?--to other demanding fields, there would be no basis that I can find in the study for predicting that more women would enter science rather than the fields that they appear to prefer.