Since the 1980s, there has been robust real-world evidence of a preference for hiring women for entry-level professorships in science, engineering, technology and math (STEM). This evidence comes from hiring audits at universities. For instance, in one audit of 89 US research universities in the 1990s, women were far less likely to apply for professorships only 11%-26% of applicants were women. But once they applied, women were more likely to be invited to interview and offered the job than men were.

But what went on behind the scenes with these hiring decisions? Did women applicants give better job talks than men, publish more or in better journals, or have stronger letters of recommendation? Were hiring committees trying to address the faculty gender balance that typically skews more male than female?

To find out why academic faculty preferred women, an experiment was needed, and we recently conducted one.

Collecting Hypothetical Hiring Data

Previously, in five national experiments, we asked 873 faculty from 371 colleges and universities in all 50 US states to rank three hypothetical applicants for entry-level professorships, based on narrative vignettes about the candidates and their qualifications. We told participants our goal was to collect information about what faculty looked for in job applicants when hiring, so we could advise our own graduate students.

We asked them to imagine that colleagues in their department had already met these hypothetical applicants, evaluated their CVs, attended their job talks, read their letters of recommendation and rated the applicants as 9.5 out of 10 (very impressive) or 9.3 (still impressive, but just less so).

One of the applicants was an outstanding woman, pitted against an identically outstanding man. Because men and women were depicted as equally talented, any hiring preference had to be due to factors other than candidate quality. We included a third, male, foil candidate as one of the many ploys we employed to mask the gendered purpose of the experiment. In this previously published research, we found that both female and male faculty strongly prefer (by a 2-to-1 margin) to hire an outstanding woman over an identically outstanding man. The sole exception to this finding was that male economists had no gender preference.

Faculty of both genders exhibit 2-to-1 preference for hiring women applicants with identically outstanding qualifications, with the exception of male economists.

Even when we gave faculty only a single applicant to evaluate, those given the woman rated her more hireable than did those given the identical applicant depicted as a man. Not surprisingly, this finding caused a media frenzy, as it contradicted what many believe to be sexist hiring in academia.

Note that these experiments were not designed to mimic actual academic hiring, which entails multi-day visits, job talks and so on. The purpose of our experiments was not to determine if women are favored in actual hiring but rather to determine why data suggest they are in real-world conditions. To answer this question, one needs a controlled experiment to equate applicants.

Remember that our experiment looked at typical short-listed candidates who are extremely qualified at the point of hiring, and did not address advantages or disadvantages potentially experienced by women, girls, men and boys throughout their development. It is worth acknowledging, though, that a 2-to-1 advantage enjoyed at the point of tenure-track hiring is substantial and represents a pathway into the professoriate that is far more favorable for women than men.

Finding The Limit To A Preference For Women

We wondered how deeply the faculty preference for women that wed previously identified ran. Do faculty prefer a woman over a slightly more qualified man? How about a much more qualified man?

Our most recent experiment, just published in the journal Frontiers in Psychology, examined this question.

Using the same methods from our earlier study, we presented 158 STEM faculty with two male applicants and one female applicant for a tenure-track assistant professorship in their specific field. We presented another 94 faculty with two female applicants and one male applicant. In one contest, the female applicant was slightly less outstanding than her two male competitors, although still impressive; in the other, the male applicant was slightly less outstanding than his two female competitors.

It turned out that faculty of both genders and in all fields preferred the applicant rated the most outstanding, regardless of gender. Specifically, faculty preferred to hire slightly more outstanding men over slightly less outstanding women, and they also preferred to hire slightly more outstanding women over slightly less outstanding men.

Reconciling With Other STEM Sex Bias Research

These results show that the preference for women over equally outstanding men in our earlier experiments does not extend to women who are less accomplished than their male counterparts. Apparently, when female and male candidates are not equally accomplished, faculty view quality as the most important determinant of hiring rankings.

This finding suggests that when women scientists are hired in the academy, it is because they are viewed as equal or superior to males. These results should help dispel concerns that affirmative hiring practices result in inferior women being hired over superior men.

The absence of preference for a less outstanding man does not necessarily imply that academic hiring is meritocratic under all conditions. It is possible that with different levels of candidate information (or if the candidates were somewhat less competent, as opposed to being stellar), results might differ. Discrimination may be a concern when candidate qualifications are ambiguous, but, based on our study, not when candidates are exceptionally strong. Thus, our interpretation of our results is that women who are equal to or more accomplished than men enjoy a substantial hiring advantage.

These findings may provoke concerns. If affirmative action is intended to not merely give a preference to hiring women over identically qualified men, but also to tilt the odds toward hiring women who are slightly less accomplished but still rated as impressive, gender diversity advocates may be disheartened. Those whove lobbied for more women to be hired in fields in which they are underrepresented, such as engineering and economics, may find the present findings dismaying and argue that extremely well-qualified female candidates should be given preference over males rated a notch higher.

One claim finds no support in our new findings: the allegation that the dearth of women in some fields is the result of superior female applicants being bypassed in favor of less accomplished men. If excellent women applicants were given short shrift, the slightly less qualified man would have been chosen frequently over more qualified women. But this scenario occurred only 1.2% of the time similar to the number of times a slightly less accomplished woman was chosen over a more accomplished man.

None of this means women no longer face unique hurdles in navigating academic science careers.

Evidence shows that female lecturers teaching ability is downrated due to their gender, letter writers for applicants for faculty posts in some fields use more standout (ability) words when referring to male applicants, faculty harbor beliefs about the importance of innate brilliance in fields in which womens representation is lowest, and newly hired women in biomedical fields receive less than half the median start-up packages of their male colleagues to mention a few areas in which women continue to face challenges.

Nor do the present findings deny that historic sexism prevented many deserving women from being hired, or that current implicit stereotypes associating science with men are not related to lower science course-taking.

All of these studies suggest areas in need of further work to ensure equality of opportunity for women.

On the other hand, based on hundreds of analyses of national data on the lives of actual faculty women and men across the United States, we and economists Donna Ginther and Shulamit Kahn found that the overwhelming picture of the academy since 2000 is one of gender fairness. Our analyses examined hiring, remuneration, promotion, tenure, persistence, productivity, citations, effort and job satisfaction in every STEM field. The experiences of women and men professors today are largely comparable, as is their job satisfaction.

Our new experimental findings call into question unqualified claims of biased tenure-track hiring. Sex biases and stereotypes might reduce the number of women beginning training for the professorial pipeline, but when a woman emerges from her training as an excellent candidate, she is advantaged during the hiring process.

Stephen J Ceci, Professor of Human Development, Cornell University and Wendy M Williams, Professor of Human Development, Cornell University

Read more: