This past semester I took a course about women in science and engineering and read a number of scholarly works about women in STEM academia. One of the most interesting was a 2001 paper published in the “Gender and Science Reader” titled Nepotism and Sexism in Peer-Review by Christine Wenneras and Agnes Wold. If you are interested it is available on Google books, here.
In the paper, the authors investigate the scores given, in peer reviews, to men and women applying for fellowship positions in the Swedish Medical Research Council.
This quote sums up the paper perfectly: “Our study strongly suggests that peer reviewers cannot judge scientific merit independent of gender. The peer reviewers over-estimated male achievements and/or under-estimated female performance, as shown by multiple-regression analysis of the relation between parameters and scientific productivity and competence scores.”
The authors found that women were scored lower than equally qualified men and that both men women who had a relationship with their reviewer received a “friendship bonus” but, for women, it did not compensate for the gender penalty.
This paper is not anecdotal. It is not a collection of stories about sexism. It is hard data comparing quantifiably equally qualified candidates. Gender is shown to be a statistically significant factor for evaluation. The study takes place in Sweden, the land of gender equality (and fjords) and there is still a problem. Reports like these make it much more difficult for sexism skeptics to argue that women are not treated differently or that women are only treated differently because of a difference in the quality of their work. This type of study requires the system to address its own bias and make changes, while also supplying a metric by which change can be measured.
What makes this paper even more interesting is that the authors had to go to court to get the peer-review scores for these fellowships. In the past they have not been public. There was really no reason for the reviewers to place their scores into a larger context that would reflect back on them. If you don’t think that anonymity leads to bad behavior, just check out any comment section on the internet.
In the end, the authors suggest an open system with oversight, which makes perfect sense. We humans are terrible at recognizing our own biases. What systems would you like to see change to reduce sexism?