Gender bias in science: The unfairer sex?

(Economist.com) - SCIENCE, popular prejudice often has it, is a man’s world. These days that is not actually true of many disciplines, particularly biological ones. But some, though not all, recent research has suggested women are indeed still discriminated against by the processes of recruitment and advancement on which scientific careers are built—especially in fields such as engineering, mathematics and physics that remain male bastions.

The results of an experiment conducted by Wendy Williams and Stephen Ceci of Cornell University are therefore intriguing. As they report in the Proceedings of the National Academy of Sciences, Dr Williams and Dr Ceci have searched for sex bias in one of the most important steps on the scientific career path, recruitment to a so-called tenure-track position. If the successful candidate does well in such a post it can lead to a job for life. And the two researchers did indeed find bias—against men.

Dr Williams and Dr Ceci conjured up trios of hypothetical candidates for tenure-track jobs in various fields. In each case two of the three were fantastically qualified and one, there to act as a foil, slightly less so. They sent the three candidates’ CVs, together with mocked-up interview comments about them, to 873 high-level academics in the departments of biology, economics, engineering and psychology at 371 American universities. They tweaked the particulars of each trio to match the relevant discipline, and randomised which of the two outstanding candidates was referred to as “he” and which as “she”. Respondents were asked simply to pick the best of the three.

As the chart shows, professors of biology, engineering and psychology all chose female candidates over equally qualified male ones, and did so by an overwhelming margin (as high as three to one in the case of psychology). Moreover, they made this choice regardless of whether they, themselves, were men or women. The sole exception to this pattern was economics. In this discipline male professors showed a slight preference for men, though females had a strong one for women.

When Dr Williams and Dr Ceci carried out further experiments, looking in more detail, they found that the pattern they had discovered held up regardless of whether or not hypothetical candidates were married, had children or had taken a period of parental leave. These factors, often cited as damaging to women’s academic careers, seemed to weigh little with the professors in question.

A criticism of the researchers’ method is that the professors knew they were involved in an experiment (though they did not know its purpose). They may therefore have chosen the female applicant simply because they knew they were being scrutinised and wanted to show their feminist credentials, knowing that they would not have to live with the consequences. To control for this possibility, Dr Williams and Dr Ceci also sent out 127 identical CVs—half purporting to be of women and half of men—to 127 other academics, asking them simply to rate the candidate. Their idea was that an absence of applicants for comparison would reduce any pressure to be politically correct.

In this case, too, the women triumphed. Notional female candidates scored a full point higher than male ones on a ten-point scale. Presented with identical track-records, respondents seemed simply to think more highly of women.

Whether the preference revealed by this experiment translates into the rough and tumble of the actual tenure-track recruitment process is hard to say—as is whether similar preferences exist at other milestones on the academic career path. But this result is so surprising that it should at least give pause to those who believe the academic deck is stacked against women. It might also give pause to those who think male economists are rational.