Normally I don’t like to just parrot content from other blogs, but Alex Tabarrok on Marginal Revolution does such a good job of summarizing the latest talk on gender differences and math ability that I had to include it here. Basically, the research shows that males, while having the same mean in terms of math ability, have a higher variance in ability than females. This means that you get more outliers (on both ends of the spectrum) that are male. (This is fair to say in absolute numbers do to the roughly 1:1 ratio of males to females in the population.) If you believe that a supremely high (read, off the charts) math ability is a significant driver of a career such as, oh I don’t know, physics or math professor, it wouldn’t be surprising that there are more males than females in these professions.
That in itself is nothing particularly new or surprising. What is a bit suprising is that, again quoting MR, “Now the study authors clearly wanted to downplay this finding so they wrote things like ‘our analyses show greater male variability, although the discrepancy in variances is not large.'” This is surprising, and frustrating for that matter, because the study was published in the July 2008 issue of Science, which is from what I can tell a respected academic publication. Isn’t one of the perks of academia the fact that researchers can be objective in reporting their findings (and in fact, one could argue, are obligated to do so) rather than have to couch results in politically correct terms? Alex goes on in his post to explain that this “not large” discrepancy actually makes a big difference for the tail ends of the distribution, so the downplaying is not particularly justified.
Anyway, enough of quoting others…my (probably unpopular) question is why is it so damn important for women to be like men in this regard? I am in no way anti-feminist, but the desire to “prove” equality rather than try to objectively understand similarities and differences truly perplexes me. What perplexes me even more is I’ll bet that no one ever got hung (hanged?) in effigy (or fired, for that matter) for saying that, I don’t know, the top male runners can run faster than the top female runners, even though this, too, is mainly a matter of innate ability. Or, to give a more direct analogy, why aren’t people trying really hard to understand why there are more female librarians than male ones? Maybe there are more hyper-organized females in the upper tail of that distribution? My guess is that the attention would shift if librarians were brought to a level of compensation and prestige equivalent to that of university science professors.
It seems as though the real impetus for a lot of this research is to try to make the argument that women are discriminated against in at least some parts of the science and engineering disciplines, or that they are somehow discouraged from pursuing their true love for the quantitative early in life. (If we rule out innate differences, discrimination must be the explanation, right? I’ll come back to this below.) Unfortunately, the authors of this research end up grasping at questionable straws rather than admitting that the statistical evidence is not what they want it to be. Isn’t it a bit odd that researchers appear to want to find evidence for a discrimination situation? Personally, I would prefer that people of my type are not discriminated against, thank you very much. My guess is that the discrimination situation is more palatable, since we can “fix” that problem through policy, whereas it’s really difficult to make people innately smarter. Come to think of it though, the U.S. isn’t really stellar at writing anti-discrimination laws or at teaching math. Hm.
It’s not the greatest outcome from an equity perspective if females overall end up consistently sorting into lower paying and lower prestige occupations, but, given that the trend is for women to be more educated on average than men, I doubt that the sorting issue is a severe concern going forward. (See the CPS data for details.) For example, of 25-34 year olds in the U.S., .75% of women as compared to .56% of men hold doctoral degrees, and that trend persists down the academic food chain:
masters degree or higher: men 7.1%, women 9.4% (though men have more professional degrees, which could tip wages in their favor)
bachelor’s degree or higher: men 28%, women 34% (men also show a decreasing trend in this rate)
If we see education as a proxy for intelligence (which we can’t really do, but go with me here), the .56% and the .75 are roughly consistent with the idea that those holding doctorates are about 3 standard deviations away from the mean in that regard, with females actually appearing to have a fatter tail. While this doesn’t get at mathematical ability specifically, it just shows that in the aggregate it must be the case, if men outnumber women in the upper ranks of science and engineering, that women are more (academically) accomplished than men in other areas that are not being studied as fervently.
I’m almost done, I swear. My last point is, despite the outcome on the intelligence/ability debate, there are explanations for the observed occupational data that have little, if anything, to do with discrimination. A couple of my favorites:
— Preferences. Even if nobody primed them with said toys, it is still likely the case that, on average, more boys than girls would like trucks, and more girls than boys would like dolls. (Psychologists, please correct me if I am wrong- that’s what the comments feature is for.) Why is it so hard to believe that similar preference patterns exist for things like career selection? Note that this doesn’t even have to do with the idea of females choosing careers for their flexibility aspects.
— Competitiveness, or rather taste for competition. Economists Muriel Niederle and Lise Vesterlund argue that “women as a group dislike competition more than men, even if they are of the same ability. If women seek to avoid competition, then they may be less successful in obtaining promotions and more lucrative jobs.” Okay, so I suppose that this should technically go under the “preferences” category. I also find it interesting that these reseachers also find that men improve their performance more in competitive environments than do women, and this could also explain the skew of promotions and status towards men, since promotions and the like can be thought of as winning a competitive tournament against one’s coworkers. (To relate this to the example at hand, academia is more competitive than it seems from the outside, especially in the sciences, and in a lot of ways it resembles a winner-take-all environment.)
Furthermore, the statistical discrimination explanation doesn’t even make sense. According to Wikipedia, statistical discrimination occurs when individuals from different groups are treated unequally because these groups, on average, differ in behavior. (Statistical discrimination is basically a fancy synonym for stereotyping.) Therefore, the incentive to discriminate in this way comes about when there is a lack of information on the measure of interest. In the hiring market for science and engineering professionals, there are plenty of available measures that proxy an individual’s intelligence and ability, so there is little reason to resort to statistical discrimination based on gender in this context.
Ok, I am going to play with my Barbies now.