#13589: DarkSyde — 01/18  at  07:08 AM
I used to think there was something to it also. The thing is though, when I taught math, I noticed that foreign kids came and just blew the Americans kids away. (It was really kind of ugly how far ahead they were. The US kids were almost like remedial in comparison.) But I never saw any difference among the foriegn kids in regard to male vs female. So, I'm no loner so sure. I think it might all be cultural/developmental. I don't know much about developmental physiology, (Not that that will keep from opining on it). Nevertheless it sounds plausible to me that perhaps clinically measured performance differentials on spatial visualization could be explained by youg males simply being exposed to building things more and at yunger ages. Model building of all kind for example, building tree-forts, etc.

#13593: — 01/18  at  08:52 AM
So do the same people who argue for inherent differences of ability between men and women in math and science also argue that americans are inherently worse at math and science than europeans or asians? I mean, it can't have any societal component right?

Actually, here's an interesting tidbit I learned from my wife. Apparently, in some african countries, science is considered "women's" work because of its obvious association to cooking. The african woman who told her this was thinking mostly of chemistry, I believe. Unfortunately, I can't remember what countries she was referring to right now, but I'll try to find out. If true, it would seem to be relevant for this discussion.

#13595: rabbit — 01/18  at  09:19 AM
It shouldn't be any mystery how it happens that girls lose interest in science and math. Personally, I was pilloried, teased, and completely ostracized for my math ability in junior high. My teachers did the best they could to slow me down so I'd fit in. So, after finishing calculus in eighth grade, the last thing I was going to do was make a big stink so as to get to take more advanced classes at a local college.

It also shouldn't be any surprise to anyone who's spent a minute in academia that hiring and tenuring decisions aren't purely meritocratic -- even when all those making the decisions mean to be completely evenhanded. With many more qualified candidates than slots, it often comes down to elusive qualities: "buzz" and "fit" and "comfort level." Not to mention the old boy network, not coming to the interview pregnant, etc.

#13596: Barb — 01/18  at  09:54 AM
Last week I participated in a panel discussion at the American Astronomical Society meeting about recommendations put forth by the Committee for the Status of Women in Astronomy. One of the most interesting facts is that currently over 50% of the membership of the AAS aged 18-23 are women. Yup. Most of the undergrads and graduates that are members of the AAS are women. No one is sure if this is a bubble in the population that will propagate or if it will burst at some point. The problem is that women are not advancing proportionately to men.

Last summer I attended a talk at the AAS meeting by Elaine Seymour, a sociologist at the University of Colorado. What they are finding is that it takes a "straight trajectory" for women to advance from undergrad to grad to postdoc to tenure-track. Those that fall off this straight path are very unlikely to get back on again. The reasons women fall off? Family planning, family illness, and lack of support are just a few. If we think of who in a relationship is the one most likely to take off time for, say, a new family or a sick parent, it's most likely going to be the woman.

I'm an unusual case in that I fell off the track and found my way back again. That's due in large part to an understanding thesis advisor who is supportive of my decision to work only half-time as a graduate student and half-time as a contractor in order to be in a financially comfortable place.

#13599: — 01/18  at  10:26 AM
1. It's difficult for anyone who falls off the track to get back on again. I'd like to see a comparison of men who fell off. I took off a decade to work in politics, and my law career never recovered. But that's just an interesting little nit. We need to be open, in hiring, to people from differing backgrounds.

2. Even were there some differences inherent due to gender, legally we may choose to ignore or ameliorate them. As one of my profs used to say, the Declaration of Independence says all "are created equal," but it doesn't say all are created "equally." Equal in the eyes of the law, or in the eyes of a faculty hiring process, is a different thing from equal in everything. Feynman and Hawking both achieved things in physics -- in some pantheon they might be side by side. Only a fool would argue that they are equivalent. "Equal" is a legal right, and it should give much room for differences in many things.

Women tend to make superior leaders in private enterprise, too -- they demonstrate more of the qualities that surveys and studies show we need leaders to demonstrate. Look at the average corporate board, however, or executive suite, and just try to find the women. We can't blame that all on the family track.

#13600: Mis-nagid — 01/18  at  10:31 AM
The error in that quote is that the second sentence is not a consequence of the first -- it's a non-sequitur.

The first sentence is backed by some real cognitive science; that men and women have controllably testable, repeatable and falsifiable differences cognitively. It even warns about the caveats relevant to that research.

The second sentence spurns that warning and leaps right into an unsupoprted assertion: that these differences account for academic differences. They *might*, but controlling variables is brutally hard, and there's no real reason to think they are. In fact, the diversity of observations in academia casts serious doubt on the hypothesis.

If Mixing Memory is serious about the hypothesis, he will have to do the hard work to turn into a theory, not just commit logical fallacies.

#13602: Mis-nagid — 01/18  at  10:34 AM
Bah, never mind. That's what I get for not RTFA.

#13609: Barb — 01/18  at  11:04 AM
I'm certain that it's just as hard for men to get back on the track. I think my point was not that it's more diffucult for women to get back on the track, but that it's more likely that a woman who will fall off the track in the first place.

If a couple decides to start a family, who is more likely to stay home with the baby? The wife. If there is a family illness, who is more likely to stay home with a sick child or parent? The wife or daughter. I'm not saying that there aren't cases where the husband or son will take over these duties, it's just more expected that the wife or daughter will.

I agree that we need to find places that are more open in their hiring, and we need to be more supportive of letting people back in if they've decided to jump off the straight track for a bit.

#13625: Cleis — 01/18  at  01:04 PM
Ed Darrell nicely glosses over what the Declaration of Independence actually says: that all men are created equal. Not an accident.

#13627: — 01/18  at  01:43 PM
So what if there are cognitive differences? And if they mean that men are, in general, better mathematicians than women?

Joule was a boy and such a bad mathematician that physicists continue to make fun of him after 150 years. But he got his name put on a physical property. How many of the men on Summers' faculty will be able to carve that on their tombstones?

There were about 4,000 boys in my freshman engineering class, and, I believe, 2 girls.

Today, in a small community, I know at least a dozen women graduate engineers, most of them either CEOs of their own businesses or top level execs of large firms. All of them at least 20 years younger than me.

The argument that Summers makes is a stupid one, no matter what the data say.

#13638: DarkSyde — 01/18  at  03:10 PM
It's simply a fact of human physiology that our brains undergo a great deal of modification/developmental changes after birth, contingent on stimulus. If you compare the anatomy and phyioslogy of two brains after this period, one from an individual who was taught to read and write, acquired language, was nurtured, i.e., the norm of modern human interaction, with an individual who was raised in relative isolation, a horribly "neglected child" or feral-child type of environment, even if they're identical twins, the difference between the two in easily observed clinical measures of size, relative difference in key processing areas, and activity using active MRI's, would be enormous. Without even sophisticated imaging, an exerience anatomists can often visually distinguish which is which with a superficial glance. That difference is so enormous, it almost overlaps variation between hominid species. With that kind of plasticity, it becomes problematic to say the least to isolate broad, inherent, differences even among the sexes, let alone between subpopulations. And even then, what value would be gained by treating individuals with a genetic differential in the 0.1 % range when the aforementioned developmental differences exceed 20 % in mass alone, and perhaps an order of magnitude in specific cognitive performance?

#13662: — 01/18  at  06:53 PM
"when the aforementioned developmental differences exceed 20 % in mass alone, and perhaps an order of magnitude in specific cognitive performance?"

Holy $*#&! Do you have documentation for that?

"So, after finishing calculus in eighth grade, the last thing I was going to do was make a big stink so as to get to take more advanced classes at a local college."

Holy $*#&! Euler is in the house! ;)

#13670: rabbit — 01/18  at  08:15 PM
“So, after finishing calculus in eighth grade, the last thing I was going to do was make a big stink so as to get to take more advanced classes at a local college.”

Holy $*#&! Euler is in the house! ;)

No, really not. I met a lot of people with similar experiences -- at Harvard. (Or are you making fun of my "finishing" remark? -- I only meant first-year calculus, i.e. finishing what my school had to offer. Sorry to be imprecise...I'm not a math person, after all!)

#13695: — 01/19  at  06:09 AM
DS - although animal experiments would suggest that you are somewhat correct - you have to remember that abused or feral children also suffer from elevated stress levels (not good for the brain in the long-term) and poor nutrition. You might as well say that if you shoot one identical twin it is much worse at maths than the genetically identical sibling - proving that its all nurture, not nature!

#13786: — 01/19  at  04:24 PM
"Or are you making fun of my 'finishing' remark?"

Absolutely not! When I read "calculus in eighth grade," I immediately though of the only other person I knew of who'd learned so much math at such a young age. Ironically, though, I had the right idea but the wrong name: I was thinking of Paul Erdos, who published his first paper when he was 18.

"I only meant first-year calculus, i.e. finishing what my school had to offer."

"Only" first year calc? ;) It's very cool that your school even offered classes that advanced -- where I live, you'd have to be bussed to a local high school for most anything past algebra.

Trackback: Sex and science! Tracked on: Preposterous Universe (67.18.73.162) at 2005 01 20 10:37:35
Hey, have you heard? Apparently Harvard President Lawrence Summers has made some sort of remarks about underrepresentation of women in science being attributable in part to innate cognitive differences, rather than some sort of discrimination. And it...

#13908: — 01/20  at  03:59 PM
I am surprised and disappointed to find that there are so many people who still believe not only that sex differences are virtually 100% environmental (except where it is utterly incontrovertable that they are not, such as in many physical abilities), but that to consider a hypothesis other than the 100% environmental one is downright immoral. The moralizing I see here and at other lefty blogs is both disgusting and disappointing. I would hope those who are so busy fighting the religious fundies pushing creationism could also look to themselves for elements of fundie wackiness in the form of political correctness.

For one thing, it does not take a large gap in math ability between the sexes for there to be a significant gap at the high end. For another, it should be obvious that there are sex differences in ability, and perhaps more importantly, interest. Do you really think that hobbies like DXing(long distance radio reception) and ham radio are dominated by men solely or even largely because of "sexism?" It's not like participating in such hobbies carries much prestige, much less money. I think that the person who said that "wives and antennas do not mix" was on to something.

#13934: PZ Myers — 01/20  at  10:26 PM
Dude, dude, dude. I'm not surprised at all to see the reactionary fans of eugenics failing to understand and continuing to push this nonsense of slight differences in an average yielding large differences at the high end. Think about your assumptions, man: it is soooo obvious that y'all are fixated on the idea of a single linear parameter accounting for the entirety of a behavior, and that it is inherited to boot.

You're talking about a complex function with multiple dimensions. Your beloved bell curve doesn't apply. Imagine that one important ability in the climb up the academic ladder is bootlicking, and men are 2% better at it than women. Another important parameter is schmoozing, and women are 2% better at that. Which sex is more likely to succeed? I'd say you can't tell. What will determine it is which individual better implements the complex array of skills they have.

What's disgusting and immoral is the fondness of righties to unwarrantedly impose simplistic and inaccurate stereotypes on people that diminish them because of irrelevancies. Get over yourselves; being male does not endow you with intrinsic superiority.

I have never had anyone tell me in all seriousness that a fondness for ham radio was a sex-linked inherited property that was not a consequence of social factors. Thanks for the laugh.

#13943: — 01/21  at  06:31 AM
PZ where did dude claim that being male endowed him with "intrinsic superiority"? I thought his claim was that discrimination does not account for all of the difference in numbers between female & male faculty in the sciences, or possibly the stronger claim that a difference in numbers is not prima facie evidence of discrimination, which admittedly seems rather less likely, given the various experiences attested to above.

Still, although women have received many Nobel prizes in literature, & the sciences despite discrimination, & it is easy to think of dozens of great women writers & scientists, there has never been a female Fields Medalist & I am hard-pressed to think of any really significant female
mathematicians or theoretical physicists except Noether. Possibly this is just my ignorance & I would love to be corrected.

Is discrimination worse in mathematics & closely related subjects? Or might there be something else to it? I do not claim to know the answer. Maybe the population of Fields Medalists is just too small to draw any meaningful conclusions.

#13972: — 01/21  at  12:42 PM
You’re talking about a complex function with multiple dimensions. Your beloved bell curve doesn’t apply. Imagine that one important ability in the climb up the academic ladder is bootlicking, and men are 2% better at it than women. Another important parameter is schmoozing, and women are 2% better at that. Which sex is more likely to succeed? I’d say you can’t tell. What will determine it is which individual better implements the complex array of skills they have.

Utter BS. To give an example, a point in 3D space can be only be specified with 3 coordinates--for example, {x,y,z}. But that doesn't mean that you can't have a linear measure for say, distance from the origin, with some points being farther away than others. What is called 'mathematical ability' may have more than one component, and there may not be a perfect linear measure of everything encompassing mathematical ability that everyone could fully agree on, but that *doesn't* mean the idea of a linear measure of mathematical ability is completely invalid. "Imperfect" and "useless" or "meaningless" are far from the same thing. The fact that ability in some area--whether it be mathematics or basketball or anything else--is based on a variety of traits does not mean that everyone is magically equal, or that a distribution in said ability does not tend to be fatter towards the middle than towards the ends.

#14561: — 01/27  at  10:48 PM
PZ wrote: "Dude, dude, dude. I’m not surprised at all to see the reactionary fans of eugenics failing to understand and continuing to push this nonsense of slight differences in an average yielding large differences at the high end....
You’re talking about a complex function with multiple dimensions."

Spoken like a true biologist. "Don't bother me with all that math stuff -- it's never applicable to biological systems."

Talking about mathematical capability in terms of a single hypothetical parameter is a reasonable first step. Two facts: (1) given two randomly sampled people, it is almost always possible to _order_ them for mathematical ability, i.e. to say which has greater mathematical ability, and (2) there is an obvious directionality to ability, that is, there are people that are about average, there are people that are significantly worse than average, there are people who are significantly better. Sure, there may be other factors which increase the dimensionality of the full problem. Ever heard of the chi-squared distribution?

For that matter, your dismissal of quantitative reasoning about the distribution in capabilities is easily turned on its head: How dare you suggest that "academic success" is such a one-dimensional phenomenon that it can be stated definitively that women are worse off? Maybe in factor A (percentage of tenured faculty) men are better off than women, but perhaps there are factors B, C... in which women have the advantage. What happens to your argument then?

BBB
Dr. Bernard B. Beard

#14562: — 01/27  at  10:51 PM
Harry Eager wrote: "Joule was a boy and such a bad mathematician that physicists continue to make fun of him after 150 years. But he got his name put on a physical property. How many of the men on Summers’ faculty will be able to carve that on their tombstones?"

Joules are units of energy. They are not a physical property.

BBB

#14567: — 01/27  at  11:29 PM
bbbeard says,

Spoken like a true biologist. “Don’t bother me with all that math stuff — it’s never applicable to biological systems.”

Dr. Beard, I assure you that Dr. Myers believes in the all-encompassing ability of mathematics to describe events within the physical realm - including biology (Godelian issues aside).

But, is this reason for him to automatically apply mathematical formulae to phenomena which they don't yet accurately portray?

Yes, hypothesizing the existence of a single mathematical parameter has proven to be quite fruitful in the physical sciences.

But, we must be careful to realize when current models have reached their limits and more empirical data are necessary, so as to re-formulate our original mathematical conjecture to fit the phenomenon at hand.

#14568: — 01/27  at  11:36 PM
given two randomly sampled people, it is almost always possible to _order_ them for mathematical ability, i.e. to say which has greater mathematical ability,

How about two randomly selected people with a Ph.D. in mathematics from a first-tier university? When you get to the top, ordering becomes very difficult. Given two universities, it's fairly easy to say which is better overall, but when you restrict your choice to Oxbridge and the Ivy League things get more complicated. Given two countries, it's fairly easy to say which has greater quality of life, but once you only consider first-world countries, it's hard to say for certain (UC Davis > Tacoma Community College and France > India, but Harvard ? Princeton and Canada ? Sweden).

#14590: — 01/28  at  10:14 AM
Jeebus wrote: "Yes, hypothesizing the existence of a single mathematical parameter has proven to be quite fruitful in the physical sciences."

Have _you_ heard of a chi-squared distribution? The point is that you can assign a single "p value" or confidence to the extremes of a multi-variate distribution. And very slight differences between males and femaales in the distribution of mathematical ability, say, can result in large variations in the tail of the distribution. And this observation is valid no matter how many variables you would like to add to the problem.

BBB