Saturday, October 13, 2012

Biggest Barrier to Women in Science: Women

There it is. The New York Times has discovered it AGAIN! Bias against women in science. Here's the nub of the study that PROVES it:

All of the professors received the same one-page summary, which portrayed the applicant as promising but not stellar. But in half of the descriptions, the mythical applicant was named John and in half the applicant was named Jennifer.

About 30 percent of the professors, 127 in all, responded. (They were asked not to discuss the study with colleagues, limiting the chance that they would compare notes and realize its purpose.)

On a scale of 1 to 7, with 7 being highest, professors gave John an average score of 4 for competence and Jennifer 3.3. John was also seen more favorably as someone they might hire for their laboratories or would be willing to mentor.

The average starting salary offered to Jennifer was $26,508. To John it was $30,328.

The bias had no relation to the professors’ age, sex, teaching field or tenure status. “There’s not even a hint of a difference there,” said Corinne Moss-Racusin, a postdoctoral social psychology researcher who was the lead author of the paper.

"The bias had no relation to the professors’ age, sex, teaching field or tenure status"! The article highlighted this fact:

Female professors were just as biased against women students as their male colleagues, and biology professors just as biased as physics professors — even though more than half of biology majors are women, whereas men far outnumber women in physics.

But why would women, whom one would expect to be more aware of the bias against them and take mental counter-measures, have the same bias as men?

I suspect the answer is that the "bias" matches their experience. (Some "biases" represent actual information. The bias we should be rooting out is the bias against reality.1)

There are plenty of possible explanations for why this might be so. For example, it could be that women have a higher propensity to leave the so-called professional ("real") world for family2, so the most notable scientists are men.

It could also be that women of a given level of proficiency in science show it to a much higher degree than men of the same level—in other words, that men are less articulate or expressive than women. But then that would be a bias against men.

But further, why do we automatically assume that women and men have the same aptitudes in everything, and are in fact virtually identical except perhaps physically? Has it yet been demonstrated that men and women in general have equal skill at science or communication? There's nothing to say that men aren't in fact more or less skilled than women.3 Perhaps the ridiculous assumption that sex is simply an accidental "add-on" is the reason the mother of twins might be asked whether her daughter and son are identical twins.

We already know that the supposed applications in the experiment were not identical in one way significant to the reviewers: the apparent sex of the applicant. Why should we automatically assume that the sex of the applicant is not an important piece of information? Such an assumption would seem to represent a decided prejudice.

But that leaves an important question: why should prejudice justify the institutionalization of discrimination?


1. But in our postmodern age, we're told that there is no truth and truth claims are reflections of power differentials. Of course this idea means this article and the study it's based on are simply part of a raw exercise of power... as is the idea itself.

2. Making a home and raising a family are supposed to be insignificant. Perhaps this explains why the U.S. fertility rate has fallen below replacement level. Apparently matters of (national) life and death are insignificant.

3. Lest I needlessly raise the hackles of roaming thought-policepersons, I should point out an important but often-overlooked paradox. Just because the majority of mumbley peg players are male, doesn't mean that the majority of males play mumbley peg. Similarly, if all terrorists in a certain place are Muslim, it doesn't follow that all Muslims there are terrorists. Nor would a disproportionate number of Blacks among criminals mean that most Blacks are anything but law-abiding citizens.

Ideally we would judge every person on his own individual merits. But given our limited information about a given individual, we inevitably group people among others with the same apparent characteristics. It would seem that such prejudice is inevitable among created intellects.

Kenneth Chang, "Bias Persists for Women of Science, a Study Finds," New York Times (September 24, 2012).

Corinne A. Moss-Racusin, John F. Dovidio, Victoria L. Brescoll, Mark J. Graham, and Jo Handelsman, "Science faculty’s subtle gender biases favor male students," Proceedings of the National Academy of Sciences.


Osman Siddiqi said...

I can't help but feel this analysis of bias would have held more weight if there was a group of professors who were provided an applicant without a gender-biased name or if the applicant was presented without a name altogether.

Could also be interesting to see what the responses would be based on the type of name. I've heard for many years there is discrimination against difficult to pronounce names. So a comparison between John and Xiao could be pretty interesting to see as well.

Lawrence Gage said...

Thanks for your provocative thoughts!

How professors would evaluate an applicant in the absence of clues to his/her sex is an interesting question: would the a sex-neutral applicant be scored closer to male or female? I suspect the former.

But I'm wondering how that would that effect the weight of the study.

As to difficult names, that might be difficult to test. I'm guessing that national biases/predispositions would come into play--you'd need to come up with a way to neutralize that. And of course, the difficulty of a name depends very much on the reader.


noahluck said...

Women could, on average, be superior to men, equal to, or inferior to men at any given task. But we don't know a priori which it is. Before we look at the evidence, we might reasonably consider all three options to be equally likely. So the presumption of equality is the median possibility and therefore a good choice for a presumption -- at least until we see the evidence.