When we talk about the most famous scientists, we’re often on a last name basis. For figures like Darwin and Einstein, first names and even titles like “professor” seem irrelevant. We know who they are, and a single name is enough to conjure up all they accomplished.
But can you think of any female scientists where the same is true? A new study suggests that using a scientist’s surname may be helping perpetuate a bias against female scientists.
True in politics and science
The studies were performed by Stav Atira and Melissa Ferguson of Cornell. The first set asked a relatively simple question: is there any evidence of a gender bias in referring to people by their last name?
To find out, the two took advantage of a couple of available sources of data. One is simply the transcript of various news programs from across the ideological spectrum (, , , , and ). References to individuals were identified, as were the genders of those being referred to. The researchers then determined whether the individual was referred to by only their last name, without any use of initials or titles like Doctor or Professor. On the radio, at least, people were over twice as likely to refer to males by using only their last name.
To see whether this might hold true in an academic context, Atira and Ferguson turned to Rate My Professors, a service that allows students to do just what its name implies. Data was aggregated for professors from five departments (biology, psychology, computer science, history, and economics) at 14 different universities, with the same search for references based on professors’ last names done. Students during the rating were over 50 percent more likely to reserve last name references for male professors. Atira and Ferguson checked whether gender was just acting as a stand-in for seniority of the professor, but that turned out not to be the case.
Shifting directly to science, the researchers created a bullet-point list of scientific achievements and randomly attached it to a male or female name. This was handed over to about 200 Mechanical Turk participants, who were then asked to describe the researcher. If the profile had a male name attached to it, people were over four times more likely to refer to the fictitious person using their last name only.
So there’s a definite trend here of treating the genders somewhat differently when we’re communicating about them, and it does seem to extend to science. With that established, Atira and Ferguson turned to whether this tendency has consequences.
To find out, the researchers prepared a set of research proposals and randomly substituted full names and last names for the researcher who would be doing the work. Participants from Mechanical Turk felt that researchers referred to by their last name were better known and more eminent (though, oddly, not more distinguished). They also rated researchers who were mentioned only by last name as more likely to earn an award for their work. When given the option of dividing funds for different research projects, the participants allocated more to any researchers referred to only by their last name.
It’s important to recognize the limits of this study. The people recruited on Mechanical Turk are unlikely to be the same ones that foundations recruit to review their grants, so the direct relevance of this to scientific acclaim and awards is probably limited.
But that’s not the same as saying this work is completely irrelevant. While the process of science as a whole tends to limit the impact subtle biases like this one, individual scientists are just as prone to their influence as any other human. Many scientists also use popular accounts to follow news outside their field and thus they could pick up a biased perspective from that. Finally, the popular perception of scientists also plays a role in who becomes a scientist and what fields they specialize in.
Given that there are still large gender disparities in science in countries that have made significant progress toward equality, subtle issues like this one are worth paying attention to. A follow-up where name usage among scientists was examined directly would certainly seem appropriate.
For their part, Atira and Ferguson are curious about why the bias comes about in the first place. They suggest possibilities such as the fact that some women change their last name after marriage or the fact that women’s first names seem more descriptive in fields that have been male dominated.
Regardless of its cause, being aware of the potential for this bias is critical for limiting its influence. That is, at least until the day it becomes habit to talk about people like Dresselhaus and Nüsslein-Volhard as regularly as we do Darwin and Einstein.