I'm heading up to Cambridge, MA for my fifth reunion and what do I see via Jessica Valenti but an ars technica item about how " new study suggests that, when it comes to math, we can forget biology, as social equality seems to play a dominant role in test scores."
« Take the Mac Train | Main | Flattery Will Get You Somewhere »
Don't Tell Larry Summers
06 Jun 2008 01:41 pm
Comments (48)
But Matt, if you endorse any conclusion whatsoever that seems compatible with political correctness, you're just refusing to see the profound truths that are too socially inconvenient to you. Clearly, this data was compiled by politically correct scientists dead set on justifying liberal dogma. Can't you see that the brave truth-tellers who unhappily report these cosmic realities have just been the victims of "big PC"?
5 years?
I hate you.
You callow blog writing, book writing, annoyer.
Don't forget your blanket.
Exactly. I can't believe Larry Summers would suggest that biological differences between the sexes are a possible explanation for differences in their representation in math-related professions. We all know that, a priori, biological explanations like these are invalid.
The only thing I'm left wondering is, why bother doing these studies at all when we already know what the correct conclusion is?
Correlation = causation. Howabout that!
In any case, I wonder why we worry so much that girls score 2 points lower on math scores, when boys score 7 points lower on reading. Must be the blatant sexism in favor of girls.
Matt the study looks interesting and is compelling as an answer the question of differences in averages, but doesn't address the question that Summers was asking whether the flatter bell curve for men when it came to math ability and thus a higher number of men at the highest levels compared to women had any basis in biology. (To my mind the jury is still out on that question--though I do not believe we should be careful with assuming it has to do with biology.)
Let's be fair to Summers here, who was, somewhat ham-handedly, trying to ask an honest question.
Jim,
Because treating Summers' remarks as the man made them instead of a more provocative version that he failed to say would generate both fewer blog hits and fewer dates.
Correction: Tis should read (To my mind the jury is still out on that question--though I DO believe we should be careful with assuming it has to do with biology.)
Let's be fair to Summers here, who was, somewhat ham-handedly, trying to ask an honest question.
Come on. He was the president of Harvard. He should have known that was a stupid thing to do.
...but doesn't address the question that Summers was asking whether the flatter bell curve for men when it came to math ability and thus a higher number of men at the highest levels compared to women had any basis in biology.
Obviously, its social conditioning or something. Maybe girls are taught from a young age that they have a tighter bell curve. Or maybe the people responsible for academic hiring are taught that. I mean, if this kind of difference was really biological, then you would have weird results, such as the vast majority of chess grandmasters being men.
Here's a related question. How do the anti-biology zealots out there explain the fact that a much higher proportion of left-handed people are present in math-related professions, as well as in chess?
Is this also social conditioning? I've got it: it must be because they were forced to use right-handed scissors as kids. Their resulting alienation led them to choose asocial professions at a higher rate.
So why do men score lower on verbal tests - because of biological inferiority or societal discrimination?
Either explanation would be politically incorrect, so there must be a 3rd reason out there somewhere.
For fsck's sake. How many times do we have to go through this?
Summer's point was not about the comparison of average men to average women when it comes to math; it was that at the very very high end men dominated.
You may claim that this is social, not biological, but using a study that tests something completely different does prompt the question "why is it that the people who so desperately want this to be true are such complete morons about what they are actually arguing about"?
"anti-biology zealots"
Something must explains the piss-ignorant machismo of ev. psych people, but I don't think it's biology.
Math, my boy, is just the lesbian sister of biology.
Yeah I never understand why Summer's name gets invoked when these sorts of studies come out. I'm sure people have thought that there's a big gender gulf in terms of *average* IQ or math ability or something, but that's neither here nor there regards to what Summers said. And yet almost every response to Summer's I saw elided this crucial point, as Matt does here. I would also like to know if anyone denies that men dominate the *lower* ranks?
Addendum: One begins to suspect willful blindness, really.
What, Matt, you're going up to Cambridge and you're not going to use it as yet another chance to plug your book? Shame on you!
Once again, Larry Summers, who used the power of the presidency of Harvard to zealously censor criticism of Israel during his tenure, deserves no sympathy on freedom of expression grounds. Reap what you sow.
Freddie,
Its not about him as a person. He said something extremely stupid about Israel and something perfectly reasonable about women in math-related professions. Which one did he get widely criticized for? How will that affect how other people behave in the future?
Good point, Jim.
You know, even when I was in first grade I always got "Post High School" on reading comprehension but only "High School" at math. I freely admit Math is my weakest subject though with some work I became excellent at Algebra and Trig.
Guess I am in touch with my feminine side.
I just want to second the people who are pointing out that Summers was not talking about a difference between the means but the standard deviation. The Ars Technica artical is interesting because it does very clearly show that the scores on these tests have are affected by societal differences. I really want to see a study comparing the variances of men and women in different countries.
So why do men score lower on verbal tests - because of biological inferiority or societal discrimination?
We have never fully developed our verbal capabilities because we have the option, when words fail us, to substitute a punch in the nose. It's the Billy Budd effect, with both biological and social aspects.
Matt,
Here's a little secret for you. Girls don't find guys who parrot the feminist party line to be sexy. They just sound like dweebs to them.
Steve
How do the anti-biology zealots out there...
I believe that differences between males and females in academic performance are not due to differences in innate ability. That doesn't make me an anti-biology zealot.
a much higher proportion of left-handed people are present in math-related professions, as well as in chess
Just for the sake of argument, I could make the case that there is a social cause:
Becuase left-handed people tend to be treated as slightly abnormal, they are on average less comfortable socially, and thus gravitate towards activities in which success is not dependent on social skills.
I love how Matt always waves around any half-assed, marginally relevant study that seems to confirm his prior prejudices as proof positive of said prejudices. "See, I TOLD you!!! I KNEW it!!" You sound like an idiot Matt. As Sailer said, he also just repeats what the feminists tell him to say. Double-whammy in this case.
cs,
You are a zealot only if you treat people who believe that the differences in outcomes may be due to differences in innate ability as beyond the pale.
In fact, there are huge differences between the sexes that are overwhelmingly likely to be due to innate ability, such as in spatial reasoning (eg, mentally rotating objects). What's difficult to determine is how important those differences are in math-related professions, as different people use different ways of thinking to solve problems.
I'm heading up to Cambridge, MA for my fifth reunion
===============================================
I've never even heard of a 5 year reunion. How lame. What would you talk about? What could have happened? Most of the eligible people are probably still in graduate school or just out
Matthew seems determined to continue to pretend that Larry Summers said something completely different to what he actually said.
Summers' great sin, of course, was to suggest that any of the sex differences in academic achievement at all might be the result of biological sex differences. All right-thinking people must adopt an a priori commitment to the position that all academic sex differences are the result of discrimination and have nothing to do with biology.
cs,
I believe that differences between males and females in academic performance are not due to differences in innate ability. That doesn't make me an anti-biology zealot.
Perhaps not, but I think it makes you irrational. There is no definitive answer to the question yet, but there is plenty of evidence.
Ten varieties of evidence for innate psychological sex differences:
1. Many known biological differences that are plausible causes of psychological differences. Large and persistent differences in sex hormones, especially pre-natally and during childhood. Large number of hormonal receptors in the brain, including the cerebral cortex. Many small physiological differences in men's and women's brains, including overall brain size (even correcting for body size), density of cortical neurons, degree of cortical asymmetry, size of hypothalamic nuclei, and others.
2. Many, possibly all, major anthropological sex differences are found universally across all human cultures, including the view that men and women have different natures, greater involvement of women in direct child care, more competitiveness in various measures for men than for women, greater spatial range travelled by men than women, and male domination of the political/public realm.
3. Stability of sex differences over time. Surveys of life interests and personality have shown little or no change over at least two generations.
4. Many large sex differences found in other mammals: aggression, investment in offspring, play aggression vs. play parenting, range size (predicts a species' difference in spatial ability), interest in objects vs. conspecifics.
5. Many observed sex differences emerge early in childhood, some within the first week of life. Boys far more often than girls engage in rough play, involving aggression, physical activity and competition. Girls engage much more often in play parenting. Also sex differences in intuitive psychology. Most or all of these differences have been found across cultures and nations.
6. Evidence from genetic boys raised as girls. In 25 documented cases of cloacal extrophy where genetic boys were born without male normal genitalia, castrated and raised as girls, all 25 reported feeling they were boys trapped in girls' bodies, and they exhibited male-specific patterns of behavior such as rough-and-tumble play.
7. Lack of sex differential treatment by parents and teachers. Large scale meta-analysis of gender-specific socialization studies have shown that there are little or no differences in the way parents and teachers treat boys and girls.
8. Studies of prenatal sex hormones. Girls subjected to an increased dose of androgens in-utero because of the condition congenital adrenal hyperplasia exhibit male-typical play patterns, greater competitiveness, and male-typical occupational preferences. Variations in fetal testosterone show a relationship to various cognitive abilities.
9. Circulating sex hormones. Many studies show that testosterone levels in the low-normal male range are associated with better spatial manipulation abilities.
10. Imprinted X chromosomes. In genetic imprinting, a chromosome can be altered depending on whether it was passed from the mother or the father. When a girl with Turner Syndrome inherits an X that is specific to girls, on average she has a better vocabulary, better social skills, and is better at reading emotions, body language and faces.
Let's just take a look at 2.
"2. Many, possibly all, major anthropological sex differences are found universally across all human cultures, including the view that men and women have different natures, greater involvement of women in direct child care, more competitiveness in various measures for men than for women, greater spatial range travelled by men than women, and male domination of the political/public realm."
The "major" differences have changed over time. Back in the day, women were only good as calculators, etc.
The "different natures" that men and women have are differnet in different cultures -- in some places women are too horny, in others they don't like sex, etc.
"Direct child care" differs quite widely cross-culturally.
Measures for "competitiveness" -- they're a hoot!
"Greater spatial range" I guess this means distance. Don't know myself, but sure has the bright true ring of the bogus
"Public/political realm" -- definitions please. It's not at all clear that public/political realm means much applied cross-culturally, nor, in any case, that it has anything to do with the nature of women as developed on whatever mysterious African plain you'd like to dream up.
Mush, mush, and more mush, always delivered with the authoritative voice of a drunk boy in a bar. Same voice that the same boys tend to talk about trade, and the right to patents cause people need to be rewarded for their hard work in the same breath.
Mixner, 7 sounds tendentious, and 8, 9, and 10 sound like variations on 1. Otherwise, fantastic list, the kind of thing you should cut and paste into all these discussions. Sex-difference denialists, respond to these points please.
Mixner's points 1-10 have nothing to do with math. So, duly responded.
Summers should have been shunned because (1) from context, he floated the genetic differences point to suggest that gender imbalances in grad programs were not a problem he felt like doing anything about, in response to a question from a female chemistry professor, and it's absurd to think he was "floating out new ideas" at an administration meeting; (2) >99% of people are genetically inferior to do ivy league grad level math, and there's no articulable reason to think that of this extremely small percentage of people more are men than women; and (3) Summers should know better than to speak when he doesn't know what he's talking about -- if he was indifferent to or didn't anticipate the reaction to his comments, then he has no business running either a university or a hedge fund, or a university/hedge fund.
Mixner's points 1-10 have nothing to do with math. So, duly responded.
They're evidence of significant innate psychological sex differences. Psychology includes cognitive abilities, and cognitive abilities include math skills. Spatial manipulation abilities, specifically mentioned in point 9, are strongly associated with math skills. Psychology also encompasses traits that may influence interest or enthusiasm for math, which may obviously also influence test scores--not to mention all sorts of other life choices affecting career, income, social status and political power.
The rest of your post is just nonsense. I especially like "he floated the genetic differences point to suggest that gender imbalances in grad programs were not a problem he felt like doing anything about."
Your points, at best, are overdetermined. If anything, they'd speak to the mean aptitude. But the question was about people who are already exceptional, and both men and women are going to have people falling at the right-hand edge of any distribution. Even among that small subset of the population, there are still more women who could do higher level math and science professionally than are doing it now.
I see how you must be good at math, yourself, because your points are numbered. On the other hand, point 9 doesn't tell anyone anything about methodology and might not adequately control for environmental factors. For all anyone knows, cultural signals saying math is for boys might induce higher production of testosterone. Saying people think with hormones is also just a metaphor.
If, as Summers suggested, the gender differences are the be all and end all, there's nothing to be done about gender imbalances in Harvard's grad programs. Saying it's genetic differences at issue, in context, was a way to give the brush-off.
Certainly there are gender differences. But I don't think they would apply to something that is BY DEFINITION abstract, and I don't think it's accurate for Summers' defenders to say he was being disinterested, when he was really expressing his disdain for the FAS.
David B,
But the question was about people who are already exceptional, and both men and women are going to have people falling at the right-hand edge of any distribution. Even among that small subset of the population, there are still more women who could do higher level math and science professionally than are doing it now.
Men are overrepresented at both the high and low ends of the distribution of many traits. Innate math ability may be one of them. And while it is surely true that "there are still more women who could do higher level math and science professionally than are doing it now" that isn't terribly relevant since the same is surely true of men.
On the other hand, point 9 doesn't tell anyone anything about methodology and might not adequately control for environmental factors.
It wasn't intended to tell you anything about methodology. It was just a short description of one type of evidence for innate psychological sex differences. Your observation that the studies on which this evidence is based might be flawed is also irrelevant. Obviously, any study on any topic might be flawed.
The rest of your post is further ad hominem attacks on Summers. Your comments in general are just silly and defensive, a knee-jerk reaction to the promotion of a possibility you find emotionally threatening rather than any sort of rational, informed critique.
To me, the really interesting question is: does Matt's gratuitous bashing of Summers appeal to Megan McArdle? Anyone know the answer?
As president of Harvard, Summers had no business playing biologist. Still, now that he's gone, Matt doesn't get a pass on scoring brownie points with hot chix by misrepresenting his views.
[i]99% of people are genetically inferior to do ivy league grad level math, and there's no articulable reason to think that of this extremely small percentage of people more are men than women;[/i]
Um, actually there are actually many reasons to think that people in the 99th percentile are far more likely to be men than women. Do you know what a bell curve is? Seriously, do you? Because the italicized sentence above indicates that you do not, or at least, that you are not in any way familiar with the argument that is being advanced by those against whom you are arguing.
DJ says:
"Matt doesn't get a pass on scoring brownie points with hot chix by misrepresenting his views."
I doubt that hot chix get turned on by boys trying to impress them by denouncing Larry Summers
A good example of the kind of female who actually _does_ gets passionately angry at Larry Summers was UC Santa Cruz chancellor Denice Denton, who stood up after Summers speech and, in her own words, "spoke truth to power."
Denton got into a scandal about arranging a $192,000 job for her lesbian lover, one of several financial scandals involving lesbian University of California administrators. Then, on June 24, 2006, Chancellor Denton leaped from the 42nd floor of her girlfriend's luxury apartment building to her death.
http://www.vdare.com/sailer/061001_diversity.htm
Academic feminism is a financial racket.
Sailer: "Here's a little secret for you. Girls don't find guys who parrot the feminist party line to be sexy. They just sound like dweebs to them."
This is true.
Quoth Sharon Corr: "What turns me on is when I see a man being a man and not being... all liberal and understanding and bohemian because of something he's read in a magazine. I don't mean chauvinism or macho. I mean standing up like a man and appreciating me as a woman."
An alternative explanation for why men and women have similar math abilities in Scandinavian countries might be that the Scandinavian environment is an extremely and consistently harsh one that has selected against aggressive and competitive tendencies in the men, and in favor of cooperative and nurturing tendencies (i.e. more stereotypically 'female' tendencies). The cooperative, egalitarian and nurturing tendencies in Scandinavian cultures have been around for centuries and perhaps they developed under the force of environment-driven selection. If Scandinavian men are less 'male' in terms of competitiveness, aggression, etc. (probably a good thing) then perhaps this has also led to less of an advantage in doing math.
As a female (former) Rhodes Scholar in a scientific field I feel I have the experience, evidence and knowledge to say that all of you are full of horse shit. Women fail to succeed in science and math, in extraordinarily large part because from a very very very young age they are inculcated with the idea that it is unfeminine, that their male classmates feel bad when they are surpassed by girls, that they will never find a husband/be social pariahs/be freaks, etc. As we age, we come face to face with direct evidence that our work is taken less seriously than that of male colleagues and that asshats like all of you think this is okay because we are just little ladies who need to be coddled and loved and kept away from all that hard mathy stuff that might hurt our poor fragile ovaries.
Also, clearly no one's ever told you that feminists (of any gender) make the best lovers.
Because math and science, dealing with abstract concepts, should be the area best equipped to have no gender bias at all. Suggesting math abilities are a function of sex or gender makes a basic category mistake. The type of bullying posts that suggest otherwise are exactly what expat_uk is describing. Such as suggesting I don't know what a bell curve is when describing one or citing unnamed "studies" repeatedly and then defensively claiming that it's dirty pool to question those studies. As for Summers, he didn't put any preliminary research into these issues at all, and was dumb enough to suggest to a room of female scientists that maybe women aren't smart enough to do math. Academic freedom is to protect the unpopular, not the stupid.
expat_uk:
I can't speak for anyone but myself (and I especially can't and won't speak for the odious Steve Sailor), but it can both be true that gender stereotypes (and conditions of employment etc. that affect women more than men) and a slight difference at the top of the Bell Curve in math ability explain differences in the number of men and women in the top of the science field. Surely, as a scientist, you can see that both could have an effect? I personally have no stake in the issue and would be perfectly happy for it all to be discrimination. I am not naive, I know how gender stereotypes can reproduce patterns of habit. But surely it is possible, and data do seem to show, that there may be more men at the top of the bell curve in mathematical ability. Now maybe the tests are biased, or the way testing is done biases them; fair enough, this is plausible, but it isn't ipso facto true and it doesn't do any good for Matt to cite an interesting study in its own right (that to my mind does show that differences in AVERAGES has a cultural context) and claim that it proves Summers wrong. It simply doesn't.
Expat_Uk and David B.,
Um, the two professors in the hard sciences that I've worked for (in undergrad and now as a grad student) are both female. Three of the four professors on my committee are women. I have several female scientists in my family, who excel in their fields including neuroscience and physics. I am the last person to deny that there are women scientists who are every bit the equal of any man, and who may well be the best in the world at their particular field.
(In the biological sciences, as far as I know, there are actually significantly more women than men).
I'm talking about _statistical_ differences, and I'm not even asserting that such differences exist, just that it's possible that they may exist. As far as I know no study claims to have shown that men are _on average_ better at math than women, but some people argue that men prerformance is more variable- more men than women are really good or really poor at math.
The cooperative, egalitarian and nurturing tendencies in Scandinavian cultures have been around for centuries and perhaps they developed under the force of environment-driven selection. If Scandinavian men are less 'male' in terms of competitiveness, aggression, etc. (probably a good thing) then perhaps this has also led to less of an advantage in doing math.
Posted by Hector | June 7?, 2008 9:12 AM
=================================================
Scandinavian men less competitive and aggressive? How do you explain vikings?
expat_uk,
As a female (former) Rhodes Scholar in a scientific field I feel I have the experience, evidence and knowledge to say that all of you are full of horse shit. Women fail to succeed in science and math, in extraordinarily large part because from a very very very young age they are inculcated with the idea that it is unfeminine, that their male classmates feel bad when they are surpassed by girls, that they will never find a husband/be social pariahs/be freaks, etc.
Do you have any, you know, evidence to support this claim? One would have thought a former Rhodes Scholar in a scientific field would be aware of the inadequacy of because-I-say-so posturing.
Lytton and Romney's 1991 meta-analysis of sex-specific socialization (Parents' Differential Socialization of Boys and Girls: A Meta-analysis), involving 172 studies and 28,000 children, found few or no differences in the way contemporary American parents treat their sons and daughters. In particular, there was no difference in the categories "Encouraging Achievement" and "Encouraging Achievement in Mathematics."
For the thousandth time, can someone please explain to the peanut gallery, matt included, the difference between mean and standard deviation. Maybe at the 5 year reunion, they will do some stats revision, cause it is pathetic that anyone claiming to posses a background in political science be so woefully ignorant of such a basic concept.
If you have two subsets within a population, and one has a demonstrably higher variance in physical, emotional, and mental characteristics, then if you combine the subsets into one population, select for one characteristic, and then screen for the top OR bottom 500, you would expect to see a vast overrepresentation of the high variance subgroup.
Now given this explanation, aka the clear interpretation of Summer's remarks, has been promulgated repeatedly throughout the circles Matt clearly reads, I will ask the classic question. Is Matt simply lying or dumb here?
This study shows, rightly, that social environment can have a material impact on the MEAN of the scores of men and women. It says nothing about the deviations. You could have a situation in which the female mean was higher than the male, and with the types of divergence in variance between the two sexes, you could go to the most egalitarian Nordic country, ask who the top 100 people on math are, and still get 90 dude.
It also fails to recognize that in striving to close the math gap, and widening the reading gap, the paper acts as substantial evidence for the idea that men and women potentially think in different manners.
Now this has not been proven, it is simply a HYPOTHESIS, one that deserves to be tested in the manner explained. Not via histronic character assasination and instapundit like quips.
Comments closed June 20, 2008.
| 
Heads in the Sand
Who goes to 5 year college reunions?
Answer: losers.
Posted by drjimcooper | June 6, 2008 1:52 PM