Skip

Bite me, Larry Summers.
July 24, 2008 10:42 PM   Subscribe

No gender differences found in math performance. None. Not on average, at least in countries where the sexes are treated equally. And no, not at the highest, outlier levels of mathematical ability, either, despite what some believe. And not in number of undergrad math degrees earned. And not in terms of complex problem solving ability. Just plain not.
posted by kyrademon (103 comments total) 24 users marked this as a favorite

 
I still believe I can think up some plausible explanations for the gender differences I imagine.
posted by bluejayk at 10:46 PM on July 24, 2008 [2 favorites]


Can we all go shopping together, now?
posted by Blazecock Pileon at 10:50 PM on July 24, 2008


If only these results if they didn't come from a woman struggling to do statistics.
posted by Hello Dad, I'm in Jail at 10:51 PM on July 24, 2008


Uh, or a man trying to write grammatically.

If only these results didn't come from a woman struggling to do statistics.
posted by Hello Dad, I'm in Jail at 10:53 PM on July 24, 2008


Just for the record.

"[C]ountries in which men and women have access to similar resources and opportunities" doesn't necessarily mean "countries where the sexes are treated equally."
posted by Solon and Thanks at 10:55 PM on July 24, 2008


If only these results didn't come from a woman struggling to do statistics.

Dr. Hyde's results will appear in Science, which means peer review. Likely that review process would pick up problems in methodology or interpretation. Can you qualify the difficulties you are referring to, as they relate to these results?
posted by Blazecock Pileon at 10:58 PM on July 24, 2008


(Ack. I should clarify - the "treated equally" phrase seems to come from the article's summary of the study.)
posted by Solon and Thanks at 10:58 PM on July 24, 2008


No... none... not... no, not... not... not... not. Women are so negative.
posted by pracowity at 10:59 PM on July 24, 2008


PINKER: Look, the truth cannot be offensive. Perhaps the hypothesis is wrong, but how would we ever find out whether it is wrong if it is “offensive” even to consider it? People who storm out of a meeting at the mention of a hypothesis, or declare it taboo or offensive without providing arguments or evidence, don’t get the concept of a university or free inquiry.

This bears repeating.
posted by vernondalhart at 10:59 PM on July 24, 2008 [20 favorites]


Although I should add that I'm glad to hear that such a large study has been done with these as the results.
posted by vernondalhart at 11:00 PM on July 24, 2008


Okay, before we go down this road again, Summers was mentioning one in a series of proposed theories for gender difference while introducing a panel on the topic. This is far from professing an agenda or even some kind of personal viewpoint. To borrow from the Pinker piece the FPP links:
Good grief, shouldn’t everything be within the pale of legitimate academic discourse, as long as it is presented with some degree of rigor? That’s the difference between a university and a madrassa.
posted by l33tpolicywonk at 11:01 PM on July 24, 2008


wow.. Pinker said that? seriously, madrassa? at least pick something with a little bit less loaded connotations... say like a pub.
posted by Sam.Burdick at 11:08 PM on July 24, 2008


Okay, before we go down this road again, Summers was mentioning one in a series of proposed theories for gender difference while introducing a panel on the topic. This is far from professing an agenda or even some kind of personal viewpoint.

That's total bullshit. Here's what he said:
There may also be elements, by the way, of differing, there is some, particularly in some attributes, that bear on engineering, there is reasonably strong evidence of taste differences between little girls and little boys that are not easy to attribute to socialization. I just returned from Israel, where we had the opportunity to visit a kibbutz, and to spend some time talking about the history of the kibbutz movement, and it is really very striking to hear how the movement started with an absolute commitment, of a kind one doesn't encounter in other places, that everybody was going to do the same jobs. Sometimes the women were going to fix the tractors, and the men were going to work in the nurseries, sometimes the men were going to fix the tractors and the women were going to work in the nurseries, and just under the pressure of what everyone wanted, in a hundred different kibbutzes, each one of which evolved, it all moved in the same direction. So, I think, while I would prefer to believe otherwise, I guess my experience with my two and a half year old twin daughters who were not given dolls and who were given trucks, and found themselves saying to each other, look, daddy truck is carrying the baby truck, tells me something. And I think it's just something that you probably have to recognize. There are two other hypotheses that are all over. One is socialization. Somehow little girls are all socialized towards nursing and little boys are socialized towards building bridges. No doubt there is some truth in that. I would be hesitant about assigning too much weight to that hypothesis...
If you subtract the clearly disingenuous hemming and hawing, it's pretty clear that what he's trying to say is "women are biologically fitted to be nurses and men are fitted to be engineers." If you read through his remarks, his agenda and viewpoint are pretty clear. He's a sexist and a weasel.
posted by nasreddin at 11:16 PM on July 24, 2008 [10 favorites]


Not in number of undergrad math degrees earned? Really?

I don't mean to disparage, but there were three women and thirteen guys in my Linear Algebra course, and five women to twelve guys in my Differential Equations course. And I went to a school that had close to a 3:1 ratio of women to men.

Of course, the girls did significantly better than I did at Linear Algebra. Christ I hated that class.
posted by Caduceus at 11:28 PM on July 24, 2008


@Blazecock Pileon: See, the stereotype that she's saying is hard to change, the one in which women and girls are thought to be worse at math, lets me claim that we can't trust her statistics because she's a woman. It's a lame little joke.

I don't actually believe the higher variance plus slight skew for men hypothesis that Dr. Summers (foolishly) brought up. The Rosenthal and Jacobson "self-fulfilling propehcy" or "Pygmalion effect" is a much better explanation of previous data on differences in math ability. Note that there has been some criticism of the "Rosenthal Effect" (which name leaves out the woman, Lenore Jacobson, who co-authored the research, hah!) that the wikipedia article ignores.
posted by Hello Dad, I'm in Jail at 11:30 PM on July 24, 2008 [1 favorite]


Wow. He's incredibly clueless, huh? "I didn't buy them dolls so clearly no socializing happened!"

Somehow little girls are all socialized towards nursing and little boys are socialized towards building bridges.

"Somehow", as if our culture were some big mysterious process.
posted by Solon and Thanks at 11:33 PM on July 24, 2008



I don't mean to disparage, but there were three women and thirteen guys in my Linear Algebra course, and five women to twelve guys in my Differential Equations course. And I went to a school that had close to a 3:1 ratio of women to men.


OMG my anecdote totally disproves your data!!!
posted by nasreddin at 11:33 PM on July 24, 2008 [7 favorites]


From the Guardian article:

The research also found a striking gender gap in reading skills. In every country girls perform better than boys in reading but in countries that treat both sexes equally, girls do even better.

Why does this gap exist? Is it nature or nurture? Would it be sexist to hypothesize that it might be nature?
posted by ssg at 11:39 PM on July 24, 2008 [1 favorite]


I better wake up my wife and get her to explain some of these stats to me.
posted by Pater Aletheias at 11:41 PM on July 24, 2008 [1 favorite]


BTW, according to this, in 1995 women earned 6491 bachelor's degrees in math, while men earned 7360. Not a surprise that it could be 50/50 now.
posted by Pater Aletheias at 11:47 PM on July 24, 2008


Why does this gap exist? Is it nature or nurture? Would it be sexist to hypothesize that it might be nature?

I think the selection you quoted -- "in countries that treat both sexes equally, girls do even better" lends itself to the sexist (yup) and yet "logical conclusion" that it's simply nature (if all else is equal, than what else could it be?) The scare quotes are there because I hope someone doesn't take that conclusion seriously.

Because like I pointed out before, the article turns 'similar access to opportunities' into 'countries that treat both sexes equally.' Which, and someone please point out if I'm missing something and this is what the study says, is ridiculous. What countries treat both sexes completely equally?
posted by Solon and Thanks at 11:51 PM on July 24, 2008


Yeah, but did they test Asian women?
posted by fleacircus at 11:59 PM on July 24, 2008 [1 favorite]


I think the selection you quoted -- "in countries that treat both sexes equally, girls do even better" lends itself to the sexist (yup) and yet "logical conclusion" that it's simply nature (if all else is equal, than what else could it be?)

I don't know whether nature or nurture is more responsible for this difference and I'm not really sure how you could design a study that settled that question definitively, but I see no reason to dismiss the nature hypothesis completely. Male and female bodies are different, so the brains in those bodies may be different. If they are (and I mean different on average, rather than in some binary way), why couldn't males and females have naturally different mental aptitudes on average? I refuse to believe that being willing to consider nature as a factor makes a person sexist.

And, yes, I realize that the nature argument has been, and continues to be, used to justify all kinds of bad things, but that doesn't mean that it should be taboo.
posted by ssg at 12:15 AM on July 25, 2008 [3 favorites]


@nasreddin: I challenge your call of "total bullshit." Far be it from me to defend Larry Summers, although I think the question he (unwisely) raises is a fascinating one, and he might in fact be a sexist (but don't put weasels down -- they're underappreciated).

I'd summarize his remarks as:
Here are three possible explanations of the observed sex differences in professor (and other top) jobs.

1. In our society, men have an easier time and are more likely to make the (possibly unreasonable) all-waking-hours commitment to a job than women. This is the biggest reason for the observed difference.
2. His speculative and shaky calculations of some unclear and divergent data suggest that three and a half or four standard deviations from the norm, in some unclear measure of ability, there are likely to be more men than women. [This is where I think he might have chosen a different forum to present this idea, and only after strengthening it]
3. It doesn't require socialization for little girls to be more interested in dolls, and little boys to be interested in trucks. [See Gerianne Alexander's monkey study for support of this point.]
He goes on from there, but l33tpolicywonk's description as "one in a series of proposed theories" is pretty fair.
posted by Hello Dad, I'm in Jail at 12:17 AM on July 25, 2008


I refuse to believe that being willing to consider nature as a factor makes a person sexist.

I'm not dismissing it completely and I never said it should be taboo. I simply pointed out that the way in which the article changes the report's wording strongly implies that the difference is natural, something to keep in mind and be wary about before making conclusions.

I don't think considering a study about the subject is sexist at all, but I do think making a jumping to conclusion without really conducting a relevant study could possibly lead to a sexist hypothesis. My purpose, anyway, was to try and avoid the whole "reverse sexism" argument that always seems to make its way into these discussions. I wasn't thinking to make a greater point in my use of the word.
posted by Solon and Thanks at 12:24 AM on July 25, 2008


I fail at previewing, once again. "I do think jumping to conclusions."
posted by Solon and Thanks at 12:26 AM on July 25, 2008


Yeah, but did they test Asian women?

No, all the data is based on American high school students. So, essentially they've found that in a group that can't do maths at all, there are just as many girls as boys who can do maths.
posted by jacalata at 12:26 AM on July 25, 2008 [2 favorites]


How did they compare complex problem-solving activity?

Given the near-parity among boys and girls in the number of undergraduate math degrees, it is surprising that Putnam Fellows tend to be overwhelmingly male.
posted by Gyan at 12:29 AM on July 25, 2008


Here's a better description of Gerianne Alexander's "Male monkeys prefer toy cars, females like dolls" study, plus a pointer to independent reproduction of her results (by a guy named "Kim"). It includes a cute picture of vervet monkeys.
posted by Hello Dad, I'm in Jail at 12:30 AM on July 25, 2008


3. It doesn't require socialization for little girls to be more interested in dolls, and little boys to be interested in trucks. [See Gerianne Alexander's monkey study for support of this point.]

But the monkey study actually says that while male monkeys are more interested in trucks, female monkeys are equally interested in trucks and plush toys. I'm not arguing that there are no possible mental-biological differences between men and women, but I do think we should be precise about what exactly we're attributing to each gender and why.

Summers isn't arguing the same thing as the monkey study, either. He says, "I guess my experience with my two and a half year old twin daughters who were not given dolls and who were given trucks, and found themselves saying to each other, look, daddy truck is carrying the baby truck, tells me something." He's certainly very vague about what the situation "tells him", but he seems to imply that little girls, given trucks, will treat the trucks like dolls and therefore must innately prefer dolls. Which seems to fit with your point three.

Neither your point nor Summers' actually matches up to the monkey study, where the females neither prefer the plush toys nor the trucks (unless we're assuming the male monkeys are the norm and monkeys should always be solely interested in trucks.)

This, I thought, was also important: "Wallen cautions against over-interpreting the results. The plush and wheeled categories served as proxies for feminine and masculine, but other toy characteristics, such as size or colour, might explain the male's behaviour, he says. Or the male monkeys might seek out more physically active toys, he says."
posted by Solon and Thanks at 12:39 AM on July 25, 2008


I simply pointed out that the way in which the article changes the report's wording strongly implies that the difference is natural, something to keep in mind and be wary about before making conclusions.

I haven't read the paper in question, so I can't comment about how accurately the article reflects the research, but the more important point, to me, is that females performed better on the reading test in all 40 countries. I'm not about to jump to any conclusions, but that does at least seem like a pretty significant finding.
posted by ssg at 12:43 AM on July 25, 2008


HDIiJ's description of the study differs somewhat from the earlier posted one, in that one claims "the female vervet monkeys showed significantly greater interest in the feminine toys" and one says "the females played with both plush and wheeled toys equally."

So, I looked up the actual abstract, which says:

"Male monkeys, like boys, showed consistent and strong preferences for wheeled toys, while female monkeys, like girls, showed greater variability in preferences. Thus, the magnitude of preference for wheeled over plush toys differed significantly between males and females. The similarities to human findings demonstrate that such preferences can develop without explicit gendered socialization. We offer the hypothesis that toy preferences reflect hormonally influenced behavioral and cognitive biases which are sculpted by social processes into the sex differences seen in monkeys and humans."

It definitely and clearly supports the idea that these things can develop to a without socialization. But it doesn't necessarily support the "girls automatically prefer dolls to trucks" idea, which in my mind is a harmful one that can evolve to "girls automatically prefer traditional women's work to men's" in the wrong hands. And is the reason I took issue with the particular framing of this idea in that way.

I'm not about to jump to any conclusions, but that does at least seem like a pretty significant finding.

Absolutely. I agree. And I've also, clearly, been making up too many of the posts in this thread and am going to step away. I simply thought the monkey study was fascinating and wanted to make sure the abstract was well-represented.
posted by Solon and Thanks at 12:51 AM on July 25, 2008


This argument is very like a religious argument. All people (taken individually or in groups) are naturally equal: that's the basic tenet. Anything that contradicts that tenet, or that could be used by those who might wish to contradict that tenet, must be downplayed, ridiculed, discredited.

Which, given the common behavior of people based on assumptions about one group being naturally better than another, perhaps is understandable. It's not good science, but it's understandable.

What countries treat both sexes completely equally?

The undiscovered country, from whose bourn no traveller returns. For one, anyway.
posted by pracowity at 12:55 AM on July 25, 2008 [3 favorites]


Well, generally American chicks are dumb, but only because Americans in general are dumb.


GO TEAM ANYONE BUT US!
posted by sourwookie at 1:04 AM on July 25, 2008


but only because Americans in general are dumb

Allow me a pre-emptive refutation of this: Have an of you actuallly tried growing up in SW Missouri?
posted by sourwookie at 1:07 AM on July 25, 2008


Oh, right, the old nature VS nurture argument: We actively try to breed and raise dumbshits in these parts, gender be damned.

A litter of obese semi-literate shoppers is a badge of honor here. Fuck math.
posted by sourwookie at 1:13 AM on July 25, 2008


From the Guardian article (2nd link in the FPP):

Sapienza said: "The research also found a striking gender gap in reading skills. In every country girls perform better than boys in reading but in countries that treat both sexes equally, girls do even better.

On average, girls have reading scores that are 32.7 points higher than those of boys (6.6% higher than the mean average score for boys). In Turkey, this amounts to 25.1 points higher, and in Iceland, girls score 61.0 points higher.

Sapienza said: "Our research indicates that in more gender equal societies, girls will gain an absolute advantage relative to boys."


As someone who found all of a sudden in 2005 that it was increasingly difficult to hold the opinion that there were real innate cognitive differences between the sexes, it pleases me that even this latest research still suggests that differences do exist irrespective of socialization. How fitting for Larry Summers that the most striking difference appears to be that compared to girls, boys are quasi-literate brutes intellectually incapable of grasping the most basic precepts of effective communication.
posted by [expletive deleted] at 1:58 AM on July 25, 2008


The very worst thing about the shortage of mathematicians and hard scientists (of either gender) is that there's no shortage of well-meaning passionate critics who, perched behind every conceivable title and philosophy except that of hard scientist, are content to voice speculations from the sidelines. If you want to see more girls/boys/Americans succeed in a subject, then for all our sakes stop writing that paper on postmodern flimflam and just roll up your sleeves and do it yourself, or it will gnaw at you forever.
posted by kid ichorous at 1:59 AM on July 25, 2008 [4 favorites]


I don't see why anyone would find this particularly shocking or contentious?
posted by mary8nne at 2:17 AM on July 25, 2008


mary8nne: because you are operating under the assumption that Humans are Humans--silly girl.
posted by sourwookie at 2:27 AM on July 25, 2008


Perhaps the hypothesis is wrong, but how would we ever find out whether it is wrong if it is “offensive” even to consider it? People who storm out of a meeting at the mention of a hypothesis, or declare it taboo or offensive without providing arguments or evidence, don’t get the concept of a university or free inquiry.

Because the investigation is uninteresting and the results not useful. Who cares if there are genetic cognitive differences between gender (or race, or ethnicity, or any genetic grouping)? How does that knowledge help anyone?

If say there was a DNA test that could prove without any scientific doubt that men are better than women, whites are better than hispanics, and asians better than whites in math, what would you do with the results?

I mean other than to decline the invitation to speak about your scientific results at the annual KKK BBQ.
posted by three blind mice at 2:30 AM on July 25, 2008 [1 favorite]


If say there was a DNA test that could prove without any scientific doubt that men are better than women, whites are better than hispanics, and asians better than whites in math, what would you do with the results?

I'd use it as more data in the effort to understand how people think and do math at all, how that was related to whatever the DNA encoded, and how this could be helpful in replicating human intelligence in artificial agents.

Duh. Why else do you study people?
posted by jacalata at 2:36 AM on July 25, 2008 [5 favorites]


Interesting, though if we know that women earn 48 percent of all mathematics bachelor's degrees, doesn't that pretty much say already that their performance is as good as men's?

There's a sense in which it shouldn't even matter, since two groups of people don't need to achieve at an equal level in order to deserve equality of opportunity.
posted by Phanx at 2:44 AM on July 25, 2008 [1 favorite]


Wait, so they did the research? So the question ISN'T inherently sexist and evil and degrading and mean?

HUH
posted by DU at 3:06 AM on July 25, 2008 [1 favorite]


He's certainly very vague about what the situation "tells him", but he seems to imply that little girls, given trucks, will treat the trucks like dolls and therefore must innately prefer dolls.

One wonders if he's ever heard the kind of dialogues that those trucks have with each other, when played with by little boys.

(But alright, my sample set might be somewhat tainted by the fact that they've all seen the movie Cars).
posted by effbot at 3:31 AM on July 25, 2008


though if we know that women earn 48 percent of all mathematics bachelor's degrees, doesn't that pretty much say already that their performance is as good as men's?

If you can "prove" that women don't perform as good as men, you can write that off as proof of affimative action and political correctness. There's no shortage of pundits who claims to have proved that to themselves, usually by observing one or two women in their immediate surroundings.

(Or themselves; we had a small brouhaha over here when a high-profile political pundit claimed that it was directly related to breast size, based on the observation that she herself had larger than average tits and knew absolutely nothing about math, while some math chick she'd seen had smaller than average tits. Why bother with real science when you can extrapolate from yourself?)
posted by effbot at 3:50 AM on July 25, 2008


After reading some of the comments, I feel like there's some context missing here.

Very few people feel that the subject of gender differences should not be investigated at all. I will admit that a few such people have, much to my surprise, tried to make that case in this thread, but it is not a common one in the research world. There is a strange perception that the subject is "taboo" in psychological research circles, but in fact, it is a very well-researched and well-funded area of investigation.

There is, however, a certain amount of frustration that every single piece of research which even vaguely hints at the remotest tentative possibility that there might be gender differences is jumped on by large numbers of people shouting, "See?!!! We were right all along! We hire fewer female math professors because it's genetically appropriate!" -- for example, see the monkey study referenced above, which if I recall correctly might prove as little as female monkeys preferring the color red a little more, but which has been used by some to justify the idea of "women-as-nurturers, men-as-engineers".

Another example I have seen come up with some frequency on metafilter is the concept that more men are IQ "outliers" in both directions, and therefore there will be more hyper-science-geniuses among men, while women will stick a little closer to the mean. This has been offered as an explanation of the greater number of male science professors, not just with a straight face but as if it were a known fact. Even leaving aside the extremely dubious assertion that outlier-high-IQ correlates directly to hyper-science-genius correlates directly to being-hired-by-a-university, this was always based on tentative and frankly unlikely evidence which has been flatly contradicted by a number of studies, including, now, the one in the main FPP link. And yet, somehow, for a while it was "fact".

So it's not so much no one thinks the research shouldn't happen, as it is tiresomely predictable how often this kind of research is misinterpreted, misused, and misapplied ... and in which direction, almost always.

As for Summers, one thing that people never seem to remember when they defend his "brave stand against PC" or whatever is that his words came in response to his being called out because women hires dropped precipitously on his watch ... not because he changed the requirements for, say, achievement or IQ, but because he changed the age range for preferred hires, to one that is generally acknowledged to make it much more difficult to hire women in current American society. The age range, mind you -- he wasn't trying to hire "better" professors by any sensible metric.

And when he spoke in his own defense about it, he blathered on about toy trucks and made inane points that had already been brought up and soundly rebutted at the very same conference he was speaking at, by people with far more knowledge of the field than he had. He was, frankly, being a condescending ass and attempting to give his prejudices a veneer of scientific validity before an audience that knew better, and I'm not surprised they got pissed off -- not at his daring to suggest that "taboo" research be investigated, but because he was trying to cover his own ass with nonsense unrelated to the problem. And Pinker, I think, damn well knows that, and is only trying to frame the debate, not promote pure research or anything nearly so noble.
posted by kyrademon at 4:01 AM on July 25, 2008 [36 favorites]


Dr. Hyde's results will appear in Science, which means peer review.

From what I have heard Science, which has an awesome impact factor overall, has a pretty poor reputation for quality in psychology. Not all peer reviews are equal.

Also, for those who don't know..peer review usually means the editor and three reviewers, hopefully qualified and motivated area experts, have examined the article. Peer review is only the first hurdle for a finding and it is sometimes too easy. Replication and alternate corroboration are what you should hold out for before believing something.

New findings should go in the maybe folder until they have withstood the test of time.
posted by srboisvert at 4:01 AM on July 25, 2008


On the other hand, my girlfriend has a theory, that boys in Iceland go off and fish, and the girls stay home and read. I like that theory; it's the right amount of Just-so Story to be perfectly at home with evolutionary psychology.
posted by [expletive deleted] at 4:12 AM on July 25, 2008


So, I think, while I would prefer to believe otherwise, I guess my experience with my two and a half year old twin daughters who were not given dolls and who were given trucks, and found themselves saying to each other, look, daddy truck is carrying the baby truck, tells me something.

It tells me something about the quality of Pinker's "science" -- that he hasn't really fucking done any ever since he figured pseudoscientific bullshit just-so stories gave better press.
posted by mobunited at 5:13 AM on July 25, 2008


> the investigation is uninteresting and the results not useful. Who cares if there are genetic
> cognitive differences between gender (or race, or ethnicity, or any genetic grouping)? How does
> that knowledge help anyone?

...as the audience left the hall, the Bishop of Worcester's wife was heard to voice her opinion of the Theory of Evolution to her companion: "Let us hope it is not true. But if it is true, let us hope it does not become generally known."
-- Walter Gratzer, Eurekas and Euphorias: The Oxford Book of Scientific Anecdotes

After all, dear... men related to apes! How could knowing that possibly help anyone?
posted by jfuller at 5:15 AM on July 25, 2008


compared to girls, boys are quasi-literate brutes intellectually incapable of grasping the most basic precepts of effective communication.

Well, we can certainly associate that finding with socialization, viz. the celebration of extreme physical brutality among young boys. Forget trucks: boys get guns and bats and are expected to use them to defend themselves or else labeled unmasculine. Girls are allowed to talk things out, developing those key communication systems and remaining relatively unbrutalized by mundane social interaction (until they start having to deal with the bullied-turned-bullying boys.)
posted by anotherpanacea at 5:24 AM on July 25, 2008


Sadly, this stereotype still pervades society. Right now I'm at a summer camp, the Ross mathematics program, in which interested students can spend eight weeks studying number theory. The ratio of males to females is approximately 3 to 1 here. This isn't unique to this summer camp, though: Every summer program I've participated in involving math has had roughly this same ratio. Every math team I have been a part of has had roughly the same ratio. The females are every bit as capable as the males in every possible scenario you could imagine - they're equally distributed throughout the ranks. One trait of the distribution of the females on Chicago's ARML team is especially amusing: Chicago sends four teams to ARML. Every single blonde was on the "A" or the "B" team.
posted by LSK at 5:33 AM on July 25, 2008


MeFi is such a boyzone with posts like this.
posted by smackfu at 5:35 AM on July 25, 2008


Not on average, at least in countries where the sexes are treated equally.

Huh.

1. Do a global study of math scores amongst boys and girls
2. Define the countries with the smallest differences between the sexes as "countries where the sexes are treated equally"
3. ???
4. Profit!

Yeah, I'd like to see that data.
posted by splice at 5:41 AM on July 25, 2008 [1 favorite]


Anotherpanacea, I was being somewhat flippant, and of course, I certainly see how such a discrepancy could be associated with socialization, which makes the remarks quoted pretty ridiculous. The discussed study clearly suggests that social factors play a profound role in achievement between boys and girls, and this researcher then uses these same limited findings to speculate that in some magical gender-blind world girls would have an absolute intellectual advantage over boys. It seems like a pretty specious comment, and the sensationalistic nature of such a premise makes me suspect that it might have been taken out of context by the reporter, not that anything like that ever happens.
posted by [expletive deleted] at 5:48 AM on July 25, 2008


I love the old-school zealous, religious ferver socially 'enlightened' individuals respond with to any suggestion of scientific findings of innate population differences amongst sexes, races, or other politically sensitive groupings of people. It lets me know that even though notions of political fashionability change, people are innately the same people that burned heretics and witches a couple hundred years ago.
posted by norabarnacl3 at 5:52 AM on July 25, 2008 [3 favorites]


Because the investigation is uninteresting and the results not useful. Who cares if there are genetic cognitive differences between gender (or race, or ethnicity, or any genetic grouping)? How does that knowledge help anyone?

If say there was a DNA test that could prove without any scientific doubt that men are better than women, whites are better than hispanics, and asians better than whites in math, what would you do with the results?


While I agree such investigations are largely uninteresting, any sort of scientific differentiation is useful. If asians are 'better' (in a scientifically operationalized sense natch) than whites at math, then we can look to see what are the genetic differences between asian and white populations. Say we find genes A,B,C statistically different between the groups. Hey, we just isolated genes likely directly involved in neurogenesis or some aspect of cognition! That, is fucking useful.

Other points to keep in mind: population genetics are not temporally static. As the flynn effect demonstrates, nobody knows how fast they change. All differences are necessarily generational; today's asians might be statistically quicker at math than whites, but not necessarily in our grandchildren. The other point to keep in mind is all such results are populational. They therefore predict macrosocietal trends, and do not in anyway justify sexual/racial/etc. discrimination. Just because Bob is Asian doesnt mean you should immediately hire him as your accountant.
posted by norabarnacl3 at 6:04 AM on July 25, 2008


[expletive deleted]: No worries. I guess I heard echoes of Christina Hoff Sommer's "War against Boys" in your comment. I'm certainly pleased that the achievement gap has closed for girls, but all this talk of unnurtured natural differences gets under my skin, as if there wasn't a complex interplay of social factors required to turn any mass of genetic material into a living, breathing, mathematical-competent human being.
posted by anotherpanacea at 6:15 AM on July 25, 2008 [1 favorite]


Because the investigation is uninteresting and the results not useful. Who cares if there are genetic cognitive differences between gender (or race, or ethnicity, or any genetic grouping)? How does that knowledge help anyone?

Because it would imply that we shouldn't expect identical cognitive "outcomes," as measured on a population basis, even on a level playing field. That would be good to know, because so much policy is driven by the notion that unequal outcomes imply unequal opportunities.
posted by Mr. President Dr. Steve Elvis America at 6:23 AM on July 25, 2008 [2 favorites]


Nasreddin: If you're dismissing Summers' obvious use of equivocation as some front for hemming and hawing, you must obviously have some evidence he felt more strongly, right? Something I'm missing? Weasel hairs perhaps? The comments themselves may not be acceptable on face, but absent other evidence, they're all we have to discuss and prejudging them seems to necessitate bringing a bias to them to begin with.
That being said, Summers hasn't stopped apologizing for this then or since, so why is he still getting wailed on and called names on MeFi?
posted by l33tpolicywonk at 6:51 AM on July 25, 2008


I love the old-school zealous, religious ferver socially 'enlightened' individuals respond with to any suggestion of scientific findings of innate population differences amongst sexes, races, or other politically sensitive groupings of people. It lets me know that even though notions of political fashionability change, people are innately the same people that burned heretics and witches a couple hundred years ago.

LOL PC is like the inkwisition amirite!!!
posted by nasreddin at 6:53 AM on July 25, 2008


And not in number of undergrad math degrees earned.

Um. Many American mathematics departments (including the one at my university) offer two very different undergraduate degrees: one called simply "mathematics" and one called "mathematics education". The first is overwhelmingly male; the second is overwhelmingly female. Of course, it's very much to the department's and university's benefit to lump them all together as "math majors" to make the genders look balanced, but they most certainly are not.

My wife is a mathematician. My most recent PhD student, who happened to be female, just got a tenure-track job in a math department. Barbie was right—math class is tough—but what she should have said next was "That's why I like it!"
posted by erniepan at 6:56 AM on July 25, 2008 [1 favorite]


Nasreddin: If you're dismissing Summers' obvious use of equivocation as some front for hemming and hawing, you must obviously have some evidence he felt more strongly, right?

Well, first of all, he was the asshole that suggested exporting hazardous waste to the Third World because "underdeveloped countries are vastly underpolluted."

But beyond that, the fact that a sexist clothes his words in the kind of bet-hedging language scientists tend to use doesn't make him less of a sexist. If you read the entire transcript, his message is, roughly, "Women aren't as good at math or engineering as men, and that's an innate difference, so they should just be nurses and just deal with it." Note how he didn't say "doctors" or "psychologists" or "politicians," even though all these things are compatible with women not having an aptitude for engineering, and are far less tied to a given social status.

Notice, too, that in his discussion of theory 1 (high-powered jobs) his conclusion is "I think in terms of positive understanding, the first very important reality is just what I would call the, who wants to do high-powered intense work?" Not that women are constrained by enormous social and cultural pressure to subordinate their work to family considerations, not that maternity leave policies are far too limited, but that women just don't want to do "high powered intense work."

These things reveal a pretty clear sexist mindset--that the status quo is unchangeable, and if women end up in lower-status positions it's because they prefer it that way, and it's pointless to try to change it.

Do you think the fact that he does hedge his bets mitigate this at all? I mean, there are people like J. Philippe Rushton, who are out-and-out racists, who talk in this periphrastic way in order to gain credibility (and it works). Obviously given the assembled audience he couldn't have just said "Yo bitches, cook me a steak." So he did the next best thing.
posted by nasreddin at 7:13 AM on July 25, 2008 [3 favorites]


Ernie, my student experience at two different schools was that the post-grad math courses were much more gender-balanced (and racially balanced) than undergrad. Once all the folks who wanted to be engineers and (I say it like a curse) actuaries were gone, the population that stuck it through looked a lot more like the random grab-bag of a city bus.
posted by kid ichorous at 7:14 AM on July 25, 2008


Also, looking over the coursework for the "teaching mathematics" major:

* Math 347 or 348 (Fundamental Mathematics)
* Math 415 or 418 (Abstract Algebra)
* Math 461 or 463 (Probability or Statistics)
* Math 402 or 403 (Geometry)
* Math 417 or 453 (Abstract Algebra or Number Theory)
* Math 444 or 447 (Real Analysis)
* Any three additional advanced math courses

I'm not sure what "27-28 hours" means, but if it means "pick 7-9," it doesn't sound too shabby. A couple semesters worth of Abstract Algebra and Real Analysis would set someone on pretty solid ground.
posted by kid ichorous at 7:24 AM on July 25, 2008


Because it would imply that we shouldn't expect identical cognitive "outcomes," as measured on a population basis, even on a level playing field. That would be good to know, because so much policy is driven by the notion that unequal outcomes imply unequal opportunities.
posted by Mr. President Dr. Steve Elvis America at 9:23 AM on July 25


This doesn't follow. It assumes that the system producing the outcome maximizes genetic potential, which it almost certainly does not. It is almost universal in American education, for example, that children take either pre-algebra or algebra in the 7th grade. Are we to conclude from that that it is genetically highly unlikely that any children in the 7th grade (roughly 12 years old) can handle pre-calculus, or even calculus.

Of course there are such children, but we only know about them because their parents have chosen to remove them from the system and accelerate their education commensurate with their ability. But by an large the system underestimates the mathematical ability of all students. If the genetic potential of two people is 150 and 100 respectively, but both are asked to perform at the level of 40, you should expect identical outcomes from the two people.
posted by Pastabagel at 7:37 AM on July 25, 2008


This FPP reports two different studies, which had different results. The results of the Hyde, et al., study, are actually consistent with Summers's famous hypothesis about greater variance in male math abilities. The explanations given are different, but this study doesn't differentiate those explanations.
What about at the very high end of the scale?


"While we did find more boys than girls above the 99th percentile at a 2-to-1 ratio, still, 33 percent of those kids who are above the 99th percentile are girls," she said.


Hyde's explanation for the fact that boys score higher on average on the math SAT's is that more girls take the SAT, so that the test includes more girls who are lower in the ability distribution.
Link.

The second study, Sapienza, et al., discussed in the second link (Guardian article), might have something to say about Summers's hypothesis, but the article is silent on the point, as it talks only about means and not variances.
posted by grobstein at 8:01 AM on July 25, 2008 [1 favorite]


That being said, Summers hasn't stopped apologizing for this then or since, so why is he still getting wailed on and called names on MeFi?
posted by l33tpolicywonk at 6:51 AM on July 25


Part of it was his misunderstanding of his job, which is also misunderstood by those quoting Pinker on the spirit of free inquiry in the academy, and by Pinker himself. A university president's position is traditionally rather hands-off. You don't support free inquiry and rational method by stirring up hostility to lines of inquiry in a field about which you know nothing. He wrote off a large body of painstaking research with a trivial, rambling, anecdotal and attention-seeking speech that put itself in front of the work it was meant to be merely introducing. It was an abuse of his position, and an assault on the mission of the university. It's justifiably an unforgivable sin within the academy, because it makes things more stupid - how can you carry out your job if its measured by a yardstick that makes no sense to it? Under the circumstances, walking out was a dignified and appropriate response.

More generally, Summers seems to have acted more like a certain kind of divide-and-rule CEO than a consensus-building university admin, and that's just harmful, particularly in a place like Harvard. He had few supporters left by the time he came out with that rant. One of the conspiracy theories is that he decided to go out in a blaze of controversy. If so, it worked: he's a hero to some, and no number of apologies will help them understand what it is that he is notionally apologizing for.
posted by GeorgeBickham at 8:18 AM on July 25, 2008 [2 favorites]


It tells me something about the quality of Pinker's "science" -- that he hasn't really fucking done any ever since he figured pseudoscientific bullshit just-so stories gave better press.

mobunited, you are quoting Summers, not Pinker.
posted by ssg at 8:28 AM on July 25, 2008


Just so we're clear, it's just good science to conclude that girls are better than boys at reading, but it's a disgusting revelation of a black soul to conclude (from other studies, etc.) that boys are better than girls at math. Do I have that right?
posted by callmejay at 9:50 AM on July 25, 2008 [1 favorite]


Why does it have to be nature OR nurture? I don't understand why the debate MUST be polarised like this. I believe our tendencies/preferences are shaped by a mix (yes, that's the keyword) of nature and nurture. Am I a heretic for believing this?
posted by FieldingGoodney at 10:16 AM on July 25, 2008


Where are you getting that from, exactly, callmejay? Did you miss that this whole discussion has been over whether-or-not such results are sexist and such studies should be done in the first place, regardless of the results?

And what (other studies, etc.) are you talking about?
posted by Solon and Thanks at 10:20 AM on July 25, 2008


Am I a heretic for believing this?

Um, no. Everyone agrees with you, and in fact that's the standard and obvious position. What's in question is the composition of the mix.
posted by nasreddin at 10:21 AM on July 25, 2008


And what (other studies, etc.) are you talking about?

Oh, I'm not saying I know of any, per se. I'm just reacting to this idea that Pinker and Summers are sexist douchebags. I don't know much about Summers, but I've read some of Pinker's work and he strikes me as a scientist who legitimately wants to know what's true, not as some guy with an axe to grind.
posted by callmejay at 10:34 AM on July 25, 2008


Gripe: if you're going to post about a journal article, at least post a link TO the article. Some of us can read the full text (others at least can read the abstract).

Hyde et al., Gender Similarities Characterize Math Performance, Science, 2008. DOI: 10.1126/science.1160364
posted by bonehead at 11:00 AM on July 25, 2008


It's funny how this study is being trashed in academia but lauded by the public and media. There is no treatment or analysis of higher level math skills (complex problem solving), due entirely to its absence in the data they used (standardized and state exams). Meaning the important skills for success in mathematics (calculus and the like) are not being considered, only low level problem solving skills or recall.

Summers is not a douchebag, by the way. He's an idiot who speaks before he thinks about something carefully. Forgetting Summers for a moment, I don't think anyone should take offense for suggesting that a potential exists for a genetic basis for differences in how young boys and girls learn abstract concepts such as mathematics. The worst part of the entire outcry was that anyone who later tried to pose a question like this got grouped up with him. In my experience from high school to graduate school, women are better at math, especially abstract concepts. Well, maybe thats because I only paid attention to the smart ones.
posted by available at 11:06 AM on July 25, 2008


"Here's a better description of Gerianne Alexander's "Male monkeys prefer toy cars, females like dolls" study, plus a pointer to independent reproduction of her results (by a guy named "Kim"). It includes a cute picture of vervet monkeys."

God, if that's the better description, I pity those who suffer with worse.

Unnoted regarding vervet monkeys: They're social animals too. That they have gendered differences in play does not mean that those differences are innate.

Unfortunately, many such questions about the relative levels of influence between environment and genetics are unanswerable without truly evil experimentation, especially given that environment shapes brain activity and growth. It's not like we can just lock children away to shelter them from any gendered socialization.
posted by klangklangston at 11:08 AM on July 25, 2008


And no, not at the highest, outlier levels of mathematical ability, either, despite what some believe.

From the article:
The variance ratio (VR), the ratio of the male variance to the female variance, assesses these differences. Greater male variance is indicated by VR [greater than] 1.0. All VRs, by state and grade, are [greater than] 1.0 [range 1.11 to 1.21]. Thus, our analyses show greater male variability, although the discrepancy in variances is not large. Analyses by ethnicity show a similar pattern.
I'd mark this still in question. I certainly don't think Pinker's hypothesis on sex difference of variencea1 is proven or refuted by this study. A difference of 10% variance is small but significant in the authors' sample. The effect may not be as large as Pinker supposes, but this study does not conclusively prove it false either.

1. It ain't a theory without proof. It's a hypothesis. Calling it a "theory" gives it far more weight than it should have at this stage. Very bad.
posted by bonehead at 11:11 AM on July 25, 2008


Larry Summers' argument was that there is greater variation in male intelligence, with more outliers at both ends of the curve. Of course, there's good evidence that there are more men with mental retardation, and we know the genetic reason why. Of the three major causes of retardation, one (fragile X) affects the 23rd chromosome, which also determines the difference between the sexes. Thus this genetic problem predominately affects men (women are carriers, but with some tragic exceptions do not exhibit lower intelligence), since women's second X chromosome masks the effects of the first "fragile X" chromosome. So there are more mentally retarded males, which speaking from the experience of my extended family, is really tragic. On the other hand there's no good evidence that there are more male than female geniuses. And even the bad anecdotal evidence doesn't have even a passably rational genetic explanation. Put succinctly: Just because there are more retarded men doesn't mean that there are more genius men.

This should highlight the public policy problem. If the President of Harvard University, speaking at an important conference can espouse a theory the the loserest of the loser humanities geeks like me can see is flawed . . . well then, we have a problem with the way we generate hypotheses.

Look, psychologists have shown us that we make mistakes in predictable ways. One of those ways is that we draw relationships between things based on "salience," i.e. qualities that (for us) stand out from the background. For most of us, gender and race are very salient. We notice them, and we're more inclined to attribute certain additional qualities to them, rather than to unnoticed or undernoticed other qualities. So when we see that there are more and less intelligent people in the world, we're inclined to wonder whether race or gender has something to do with it. And if white men hold most of the power, then the seemingly natural and objective hypotheses to test appear to be "are women bad at math?" and so on.

So what's the problem with having bad hypotheses floating around? First it leads to a whole bunch of Neanderthal "women are dumb, yuk, yuk" posturing, that does have an effect on behavior and community. Beyond that, it's a distraction from the hard job of discovering the nature of things -- which I take it is the job of science. I imagine it's hard to advance science, when you constantly have to conduct a rear guard war with everyone who gets the brave "new" idea that maybe gender (or race) and intelligence are linked. (See Kyrademon's great comment above)

These "hypotheses" are especially annoying when they pop up in any discussion of race (see this thread for the usual result on this website). You know, we all came from Africa and only a small portion of our ancestors left that continent. My memory is that the genetic diversity among Africans and the recent diaspora (as a result of slavery) is as extensive as the diversity among all the other peoples and nations of the entire world. Thus any sentence that starts with "Blacks are . . ." is almost certain to be wrong from the start.

This problem is especially big at Metafilter, because we take such pride in our massive, pulsing brains. In Woody Allen's words, it's our second-favorite organ. As a result, we have a LOT of posts about intelligence, and then a LOT of comments about why some people (us, tee hee) are perhaps more intelligent than some other people (jocks, people we hate, people we don't really know, etc.).
posted by ferdydurke at 11:14 AM on July 25, 2008 [3 favorites]


The study could also show:

1. Men are better at better at math and are being socialized against in math performance.
2. Women are better at math and are can hit equal performance despite being discriminated against.
3. Little about nature/nurture but simply that we're at a point where Math abilities are approaching equality between the genders on average.

Also, the 'dolls/trucks' example that comes up doesn't necessarily suggest innate ability but possibly an innate desire towards the things which nurture certain abilities. This seems to me to be how evolutionary psychology texts define it, but I'm not in the field so maybe my selection is off. Not sure which the spokespeople for or against it stand on this but the rhetoric seems to suggest that they believe there is some math gene(s) which some think is better in men than women and others think this is equal. This is of course fairly silly.
posted by kigpig at 11:54 AM on July 25, 2008


There seems to be a reticence to examine the natural side of our tendencies/preferences whenever this discussion comes up. Is it wrong for a little boy to love playing with trucks, no matter how much his concerned parents try to mold him into something more "neutral" via introducing dolls to his toys? If he rejects those dolls, is that a good or bad characteristic?

We tend to be good at what we love doing (an obvious thing to say, but it needs saying). If a girl finds math interesting just because of some natural preference that inclines her toward puzzle/problem solving, and she therefore excels at math, is it wrong to suggest it's in her nature that made her great at math? Similarly, a man may be great at math because of the very same tendencies.

Of course we can highlight and encourage these preferences through nurture, but why should be fear saying "it's simply his/her nature to love this"? That's what I'm getting from this thread.
posted by FieldingGoodney at 12:31 PM on July 25, 2008


The article referenced in the second link:

Gusio et al., Culture, Gender, and Math, Science, 2008., DOI: 10.1126/science.1154094

There is no direct discussion of variances in the article itself, only averages. Their main conclusions are:
...intra-gender performance differences in reading versus mathematics and in arithmetic versus geometry are not eliminated in a more gender-equal culture. By contrast, girls' underperformance in math relative to boys is eliminated in more gender-equal cultures. In more gender-equal societies, girls perform as well as boys in mathematics and much better than them in reading. [The factor is about 30% better in reading-bh]
There's a very interesting data set in the authors' supplemental (PDF). In this, the authors present their base facts and figures, which support their conclusion above. However, they also include data for ratios of girls to boys scoring above 95% (ie above 2 standard deviations from the means) and above 99% (3 standard deviations). In the supplemental, it's clear that there is a linear correlation between the mean sex differences and the sex differences at +2 sd. This implies that any "larger variance" explanation for sex differences must be less than their standard error (about 20%). So the largest variance difference that would still fit this data is less than 15%.

So there's no conflict between Hyde et al. and Gusio et al., but clearly the "greater male variance" hypothesis is a) at most a smallish effect, less than 10%, and b) much, much less than the social and cultural effects. (as can be seen in developmental index data in the supplemental of Gusio et al.).

So, if this greater variance in "nature" exists for the male population in mathematical ability (and it may, according to Hyde's data), it's overwhelmed by "nurturing" cultural effects (as shown by Gusio).

Is this the final word? Probably not. The both sets of authors have chosen to look at standardized tests. We know these aren't perfect for assessing outcomes (see here, for example). Meta studies, like both of the articles cited, are a whole lot more informative than small studies like the primate work cited above, IMO.
posted by bonehead at 12:39 PM on July 25, 2008


On the other hand there's no good evidence that there are more male than female geniuses.

What about the massive predominance of men among winners of major prizes -- Nobel , Pulitzer, Booker, Kyoto, Turing, etc.? Of course, you could argue that it's all conspiracy, social bias or an old boy's network, but it seems on balance like pretty solid evidence of genius.

Unless you accept Summers' other argument, that men are simply driven to over-focus on career and achieve more. Then, there could be an equal number of genius women who simply lead more balanced lives and don't get famous. But Occam's Razor cuts against this, I think.
posted by msalt at 12:51 PM on July 25, 2008


If we're talking about gender differences in young children (<5 yrs), we need to keep in mind that this is the time when children start exploring their own gender, mostly because this is when they have started to have a sense of themselves as individuals. It's totally natural for young girls to play with dolls mostly because they have JUST REALIZED "Oh hey, I'm a girl." Same with boys and trucks. It's also totally normal for girls to play with trucks and for boys to dress up like princesses at this age because the very *idea* of gender is TOTALLY NEW.

This has nothing, absolutely nothing, to do with intelligence, math ability, reading ability, or even a potential future as a transvestite. NOTHING. This is a totally normal part of development, just like crapping your pants until you learn how to use the potty. You play with dolls and trucks so you learn what *you* think it means to be a boy or a girl. Sure, society plays a role in it and your parents probably play the biggest part by approving/disapproving of play choices, but it's really an individual exploration.

To say that "two year old girls like to play with dolls because girls are more nurturing and they will end up as nurses" is complete bullshit. Two year old girls like to play with dolls because two year old girls have JUST NOW figured out that they ARE girls. It has nothing to do with any sort of "nuturing" instinct. It's also a back-handed way of saying that boys *won't* be nurses because they're *not* nurturing, which also isn't true in the slightest.

Oh hi, I think a lot about kids and about gender studies. I know this is kind of a tangent, but I just wanted to point out that the remarks about young children's gender development in terms of their future math abilities is not only a strawman at best, but totally baseless and idiotic.
posted by grapefruitmoon at 12:57 PM on July 25, 2008


What about the massive predominance of men among winners of major prizes -- Nobel , Pulitzer, Booker, Kyoto, Turing, etc.? Of course, you could argue that it's all conspiracy, social bias or an old boy's network, but it seems on balance like pretty solid evidence of genius.

...which is perfectly explainable, given the data of Guzio et al., as a cultural artifact. In fact, their data set suggests that there should be, on average, a 60/40 split women to men for Booker and Nobel literature winners. If so, there's a whole lot of Virginia Wolfes (Wolves?) out there waiting to be discovered.
posted by bonehead at 1:04 PM on July 25, 2008


Msalt, do you really think that female geniuses of the past weren't recognized because they 'simply lead more balanced lives'? Your argument doesn't work, and if we followed its logic not only would we get statements like:

"What about the massive predominance of men among US Presidents -- Washington, Jefferson, Clinton, Bush, etc.? Of course, you could argue that it's all conspiracy, social bias or an old boy's network, but it seems on balance like pretty solid evidence of political intelligence and ability."

But we would be unable to logically account for changing gender ratios in the present. There are many more female authors/composers/etc. now than there were in the past. My reasoning would be that in the past, women were discouraged from writing and having an education and that some females concealed their gender in order to write. Your line of thinking would suggest that men are perhaps simply better writers, since there are more recognized male writers. But what about the fact that since the female empowerment movement more and more women are public figures, writers, composers, everything? Is there something innate about modern women makes them better at these things than women in the past?

Are you really going to argue that there has been no historical social bias against women (in the workplace) (in the sciences) (having the right to vote) (everywhere)?
posted by Solon and Thanks at 1:08 PM on July 25, 2008 [1 favorite]


Beyond that, it's a distraction from the hard job of discovering the nature of things -- which I take it is the job of science. I imagine it's hard to advance science, when you constantly have to conduct a rear guard war with everyone who gets the brave "new" idea that maybe gender (or race) and intelligence are linked.

Discovering the nature of things requires that you are open to different possible hypotheses. If you remove the possibility that gender and cognitive abilities might be linked in some way for idealogical reasons, then you aren't really doing science anymore. I certainly don't think we have much evidence that eliminates the possibility males and females have different cognitive abilities on average, so why should we eliminate that possibility from the discourse?

Yes, there certainly are people who push this sort of hypothesis more than is warranted and those who take advantage of it, but we can't eliminate the hypothesis from the scientific of public discourse just because it might be distracting.
posted by ssg at 1:15 PM on July 25, 2008


scientific or public, argh.
posted by ssg at 1:17 PM on July 25, 2008


Coming a bit late to this thread, but I want to point out that the Larry Summers angle is a bit of a diversion. Everyone's reading this in terms of what happens to women at the highest levels of achievement in mathematics, but neither study actually has anything to say about that.

On the other hand, what the study DOES show is that the long held belief about girls "falling behind" in high school maths classes is no longer sustainable. Back in the 80s and 90s, many co-ed schools experimented with single-sex math classes for girls, and many parents opted to send their daughters to single-sex high schools because of a widely proclaimed effect of systematic preferential treatment of boys over girls in high school. It was widely claimed in the popular and scientific press that girls performed equally well with boys through primary school, but that in high school girls would, for whatever reason, begin to slip back (a common claim was that girls were averse to doing well in maths because it was regarded as nerdy and unfeminine; what we might call the Barbie "math class is hard" analysis).

Now, it's actually pretty exciting to learn that this "high school lag" has disappeared, but it's a little disquieting in its own way. For everyone who wants to claim that social pressures are determinative at the university level (in terms of the numbers of men vs. women entering engineering majors etc.), there's a problem with accounting for this enormous improvement in girls' performance at highschool. Has gender prejudice been eliminated from the lives of teenagers? Surely not. But then, why are highschool girls immune to these societal pressures, but find themselves succumbing to them as soon as they go on to university? Wouldn't you think that high school age kids would be more, not less, susceptible to that sort of thing?

In any event--as someone who has always thought single-sex schools creepy and retrograde, I'm glad to see that the "but my daughters will do badly at maths in order not to offend the boys" argument has lost its teeth.
posted by yoink at 1:46 PM on July 25, 2008 [1 favorite]


Discovering the nature of things requires that you are open to different possible hypotheses.

SSG: I couldn't agree with you more. My point is that we aren't generating enough hypotheses about intelligence. Instead, we keep looking under the same rock for answers, 'cuz it happens to be close to where we are standing. That's what salience does.

Here's a hypothesis for "testing": Was the Obamas' bump a "terrorist fist jab?" If I answer "no" and suggest that the question is silly, is it because of my ideology?
posted by ferdydurke at 3:09 PM on July 25, 2008


If I answer "no" and suggest that the question is silly, is it because of my ideology?

It is if you simply state it as fact. Believing something is true does not make it true. It is not enough to argue from "principles" if your principles are not falsifiable.

It is not ideology if you present some basis for your statement, e.g., you show that it evolved out of his community and that Pashtun and Arabic tribes use alternate forms of greeting.
posted by bonehead at 3:17 PM on July 25, 2008


Bonehead and SSG. I agree with you!

Will you agree that it would take you more time and energy to document that the fist bump "evolved out of his community and that Pashtun and Arabic tribes use alternate forms of greeting" than the question deserves? Maybe the efficient answer to FoxNews-level hypothesizing is simply for you to say "no, that's silly?" It certainly leaves you more time to develop and test your own non-silly hypotheses, rather than constantly refute the hypotheses of others.

Really, I don't think we're in conflict here. If you think I represent the forces of close-mindedness, you've fingered the wrong person. I'd just prefer we ask a lot of different questions, rather than rehash the same ones. It seems likely that we'll discover more that way. Based on your comments above, I can't imagine that you disagree, but I'm sure you'll let me know if you do.
posted by ferdydurke at 3:40 PM on July 25, 2008


ferdyduke: I'm not disagreeing with you. I think, and I'm a bit embarrassed that this seems to be the case, that I've completely missed the cultural perspective that you have. From where I sit, as a child of the 80s, educated liberally in a fairly liberal country, I think the nature argument is quite underrepresented and even, in some circles, taboo. Unless I'm mistaken, you have the opposite perspective and find that the nurture argument is underrepresented.
posted by ssg at 5:19 PM on July 25, 2008


SSG: Well said. I think we do have different perspectives about who is dominating the debate (and thus drowning out other voices). Oh what a difference a border makes, I suppose.

I'm going to bow out now.
posted by ferdydurke at 7:38 PM on July 25, 2008


This doesn't follow. It assumes that the system producing the outcome maximizes genetic potential, which it almost certainly does not.

No, not at all. It just assumes that, statistically speaking, genetic cognitive potential has a substantial impact on cognitive outcome. Of course people have differing cognitive outcomes for a variety of reasons other than genetic cognitive potential.

By way of analogy, one's genetic basketball potential has a substantial impact on one's outcome as an NBA player (or, in the case of most of us, non-NBA player), but the NBA isn't composed solely of those people with maximum genetic basketball potential--lots of other life events effect whether one ends up playing in the NBA.

If it turned out that statistically men and women had identical genetic basketball potential, the gender disparity in the NBA would make us very suspicious that sexism was at play (either within the NBA, or culturally). In contrast, if it turned out that statistically men and women had significantly different genetic basketball potential, the gender disparity in the NBA would not be such clear evidence of sexism.
posted by Mr. President Dr. Steve Elvis America at 8:08 PM on July 25, 2008


Msalt, do you really think that female geniuses of the past weren't recognized because they 'simply lead more balanced lives'?

No, and I didn't say that. Ferdyduke claimed that there is "no evidence that more men are geniuses" (as well as more men being subaverage intelligence.)

I said the fact that the vast majority of genius awards go to men -- including today, after massive strides toward gender equality -- is SOME evidence. Not definitive, and I'm not saying there isn't counter-evidence, or arguments to be made the other way. I just don't think it's intellectually honest to say there is NO evidence. Do you really disagree?

Are you really going to argue that there has been no historical social bias against women (in the workplace) (in the sciences) (having the right to vote) (everywhere)?

That's a ridiculous straw man, not worthy of reply.
posted by msalt at 10:07 PM on July 26, 2008




Wait, does anyone here have access to the article? Do they admit that the variance between the two groups is different? Because that's one of the main issues with discrepancies at the top of the ranges. Marginal Revolution just made a post on this point. La Griffe du Lion explored the math behind it when the comments were first made. And though he tends to extrapolate more than I'm comfortable with, the math IS terribly interesting. If you want to dive into the math yourself, you can see that it's not the mean difference which matters, its mostly the variability. That MR link gives a VR ratio of 1.11 to 1.21. This alone will mean that at 2 sigmas above mean, you'll only get 2-3% females selected.
posted by FuManchu at 7:20 AM on July 28, 2008


Uh, wasn't it noted above that the variance within males isn't normally distributed—that there are a great many more retarded males than females, which would increase the VR but not have predictive value at the upper bounds?
posted by klangklangston at 9:48 AM on July 28, 2008


FuManchu, that's very interesting but I have no idea what you're saying. Can you or someone else give translate? Tx
posted by msalt at 11:54 AM on July 28, 2008


I'm not trying to be lazy here -- I read the Marginal Revolution post, which in a nutshell agrees that boys are more variable, which could explain having more math professors (as well as many very poor students.) I just couldn't get to your 2-3% females selected, however I crunched the numbers. On a less brainiacal level, there is this quote from the Wall Street Journal story about this study, which seems to answer klangklangston's question:

In Minnesota, for example, 1.85% of white boys in the 11th grade hit the 99th percentile, compared with 0.9% of girls -- meaning there were more than twice as many boys among the top scorers than girls.

...on the other hand...
posted by msalt at 12:11 PM on July 28, 2008


Oh fer... here I am helping hold down the mean in mathematical ability Apologies, my last sentence above is pure rubbish. The original MR link shows how they derived the 3:1 ratio at 4 female SDs.

Klangklangston, you're right, I totally missed that discussion above, and am clearly not adding anything new.

That said, whatever the reason in the disparities. I'm completely enthralled with how well the results do and do not line up. The supplemental paper in bonehead's comment shows a pretty consistent 2:1 ratio in both reading for girls and math for boys, save for four exceptional countries. The statistical tools we have DO roughly line up with these results (which gives me math-goosebumps). But we're wrong on some fundamental level.
posted by FuManchu at 2:01 PM on July 28, 2008


« Older Do you really know who people are?   |   Breaking Tarantino's Code Newer »


This thread has been archived and is closed to new comments



Post