As You Were

There's been much hubbub recently about a recent speech by Larry Summers, the President of Harvard and former Clinton Treasury Secretary.

He certainly has a talent for pissing people off (and getting headlines while doing so — he even managed to snag the lede of a MoDo column [not one of her best, though]).

My friend Collin's mother was so incensed that she fired off an angry email when she heard about it on the news:

From: Sharon Poczatek
Subject: Boo on you!!
To: Lawrence Summers

You are neither a scientist nor a mathematician. You are merely an economist.
You cannot undo the damage done by your comments. You're an idiot.

Dr. Sharon Poczatek, a woman and a scientist.

This is clearly a highly emotional issue. It's an unpleasant topic that forces us to examine subconscious prejudices and question our fundamental assumptions about gender. Women don't like to be told they aren't as good as men at certain types of math and men don't like to feel guilty about their tendency not to take women seriously. Gender roles are instilled and reinforced every day and picking them apart feels weird and icky. We'd like to understand what's going on here, but the truth might be unpleasant, so we tend to stick our heads in the sand or offer up weak platitudes ("Some of the best scientists have been women . . .").

According to the New York Times, Summers is said to have cited two reasons why he thinks that women are under-represented in the sciences: 1) the variation in math ability among men is higher than among women (i.e. a man is more likely to to either be very good or very bad at math) and 2) a very large time commitment is required in the early stages of academia and that few married women with children are willing to accept such sacrifices.

His first reason is a complete canard but his second, that societal pressures and cultural factors select against women in academia, seems more plausible.

It's been my experience that the difference between the mean ability of men to do math and the mean ability of women to do math is much, much less than the variability within those subgroups. It's one thing to say men are better at math — it is quite another to use that assumption to say a particular man is better than a particular woman at math. Hairstyle, wardrobe choices, knowledge of Star Trek arcana — these stereotypes are much probably better predictors of math success than gender.

That said, we can't be fundamentally opposed to the notion that men are better than women at math — attempts to enforce ideology through epistemology quickly devolve into absurdity (see Soviet genetics or Bush Administration science policy).

There does seem to be a difference between the math abilities of men and those of women. Most (but not all) studies have shown that in childhood and college years men consistently outscore women on standardized math tests. Why this occurs and what it means have been the subjects of intense research. Men are consistently better at solving spatial problems and one theory is that this helps in visualizing word problems. Another area of research is in the ability to quickly recall mathematical facts (like 7 * 3 is 21). Again, most (but not all) studies show that the best men are better than the best women at retrieving these facts. In early math education the memorization of these "facts" is emphasized much more than reasoning about them — it is not hard to imagine boys and girls developing significant differences in their attitudes towards math solely as a result of the drilling of multiplication tables.

We would like to be able to blame this completely on that stupid Barbie doll that exclaimed "Math is a hard!" Some studies have attempted to quantify the effect that attitudes towards math have on math ability. They show that although spatial skills are the most important mediator of overall math skills, math self-confidence is also a significant mediator. It is worth noting that spatial skills increase with practice, though, and that our society has historically pushed boys towards activities that hone such skills.

There is an emerging consensus that, in general, boys are better than girls at math. My contention is that these gender differences don't matter. In academic science, fast math-fact retrieval and spatialization skills are much, much, much less important than higher-order reasoning, the ability to look at old problems in new ways, and, of course, hard work. You might argue that these skills do matter in high school and that a gender gap in science interest developed then could carry over into college, but this doesn't seem to be the case.

The historical trendline of women in the natural sciences (Biology, Chemistry, Physics, Earth Sciences, and Agricultural Sciences) is really quite remarkable.

(Data from the Association of Women in Science).

This trendline shows two important things:

1. The percentage of science majors who are women has risen dramatically over the last three decades. This is clearly a cultural transformation, not a biological one.


2. The gender ratio in undergraduate science has exceeded 1: more women get Bachelor's degrees in the natural sciences than men.

This is a really big deal. It shows that women are just as capable of doing undergraduate science (and perhaps more so) as men. If we accept the assumption that spatial reasoning and math-fact retrieval skills are much more important in early, mechanical classes than in a real research setting, then it stands to reason that the math skills differential becomes less important as a woman advances in her career. That we don't see parity at higher levels means something else is going on here. Math ability can't explain the dearth of women with science PhDs.

There are certainly biological factors in play here. Pregnancy, after all, is an option limited to women. But for most women the immediate effects of pregnancy — reduced mobility in the late stages, the physical stress of giving birth, postpartum depression — are relatively short-lived and can't account for the sharp dropoff in the number of women at masters and PhD level.

The lack of women in the higher positions of math-heavy scientific fields, then, must be primarily an artifact of culture. We need to understand why women who major in science don't choose to go to graduate school — and if they do make it to graduate school, why they don't choose to continue on.

It's all too easy to blame the current academic leadership for the problem, to portray them as a bunch of old white men who control the money and the hiring and the good-old—boy network, but I don't think that works any more. This is 2005. Active, fully conscious discrimination against women is more common than we would like, especially in the engineering departments, but it is not endemic.

The real problem, of course, the painful problem, the undeniably obvious problem, is that our society treats women not as second-class citizens but as second-class intellects. Every day we're told through sexist radio ads, through stale sitcom plots, through the unconscious prejudices of our parents, that men are more capable than women, that women can study science but that they can't be scientists. Ask a child to draw a scientist and he or she will draw a man. Probably with glasses, a labcoat, and some kind of explosion. In The Scientist as Hero, the scientist is always a man. Even The Mad Scientist is either a man or, rarely, a psychotic lesbian. There are plenty of non-scientist career paths one can take with only a bachelor's degree and many more available to someone with a master's. It seems that women choose these alternative options more than men. We'd like to know why.

Societal pressures are important. We tend to think of women as mothers first and people second. There are still places in this country where a woman is considered a bad mother if she doesn't stay at home while her kids are young. There are still places where teaching is the only acceptable career option for a woman with school-age children. Even in the most progressive of families and with the most enlightened fathers (assuming, of course, that he is still around), the mother still gets stuck with more of the housework (according to National Council of Women’s Organizations, women contribute an average of 35.1 hours a week to domestic chores, while men contribute 17.4 hours). Economic pressures are also important. All too often it is cheaper for a women to stay home and not work than to put her kids in daycare and find a job. This would be funny if it weren't so sad.

Academia presents its own particular set of challenges. It is an unforgiving occupation that demands tough choices and unpleasant sacrifices. In its early stages, especially, it offers low wages and often requires a lot of mobility. In a household with two working parents, packing up and leaving for that postdoc or first tenure-track job on the other side of the country is not a decision to be made lightly — and in a society that values the contributions of men more than the contributions of women it's not not hard to guess who gets the short end of the stick. Compounding the problem is that it is damn hard to get back into academia after having left it for any significant period of time. This so-called "leaky pipeline" effect is the primary cause of the gender gap in academia because at most steps along the way more women choose to drop out than men. The good news is that we seem to have plugged the leak, in the sciences at least, through high school and the undergraduate years, The bad news is there is still a significant disparity at the graduate level and the trendline for PhDs doesn't look good.

So what is to be done? We can shame men in to bearing their full share of domestic chores, but we can't change biological imperatives. The costs of pregnancy will always be higher for women than for men. Still, a lot can be done to correct current imbalances. Ubiquitous cheap childcare on university campuses would make a really big difference here — the state of such programs is laughable at even top universities. What good is daycare if you have to spend six months on a waiting list to get your child into it? Every university should have a fully funded daycare and preschool on site. These programs add prestige for Schools of Education and are a godsend for students and professors with young children. More subsidized housing, too, can never hurt. These programs are expensive, but they make a big difference in quality of life. They probably don't hurt in university rankings, either.

Fundamentally, though, we need to value the minds of women as much as as we value the minds of men. We don't, now, but I think we're making pretty big strides in the right direction. There's a bootstrapping problem, of course. Before society can fully accept the notion of the woman as scientist, we need to have lots of women scientists. Visibility is very important. There need to be "Mrs. Wizard"s who go on local tv and make fools of themselves trying to explain scientific phenomena to confused news anchors and astounded elementary school kids. There need to be women scientists in movies and on tv shows and caricatured in Saturday morning cartoons. We need to teach our kids that women can, and should, be scientists too.

Most importantly, we need to convince ourselves that the gender gap is real problem, that this is something worth fixing, that ignoring it is a tremendous waste of talent. We can't just throw our hands up in the air and say that women aren't as good at math or that they don't want to work as hard as men. These are bogus excuses — that they came from the mouth of the President of Harvard means we still have a lot of work to do.