Asymmetrical Information

Loved this line....

"Of course, that doesn't explain why men dominate the op-ed page of the New York Times, where great competence in writing is obviously not a required skill."

Mark seems to ignore the differences in the attitudes and relations of men and women to child-rearing. These certainly affect the paths that women take in the world, not all of which lead to fame and fortune as well as happiness.

In working for a Pharma company that had one of the best selling drugs on the market, I was "accidentally" included in a conversation in which an important and brilliant female scientist was being discussed. She was the natural one in line for a promotion to our research team....but these two senior executives were worried because she was a woman. However, her two immediate supervisors came form the "show me" crowd, and were only concerned with her abilities. She got the job.

I completely except that cognitive and physical differences will indeed influence how the stats play out. But so does prejudice. And we have not become so "clean" a society that prejudice has ceased being a consideration. Certainly we have come far in how this plays out, but to suggest there is only one aspect effecting this, is something more suited to Bizzaro world.

As much as I like Kleiman generally, his comment isn't smart -- in fact, it's rather dumb. Where it falls apart is the following:

For excellent evolutionary reasons, human males display higher variance than human females on many important traits, including measures of mental capacity.

The 'men have higher measured variance on many measures of mental capacity than women' bit? Uncontroversially true (to the best of my non-expert knowledge). The 'For excellent evolutionary reasons' bit, implying that the reasons are known to be genetic and are well-understood, rather than environmental? Not so much.

To the best of my knowledge (again non-expert, but interested -- please link to the research if you know of something to the contrary) there is no well supported biological mechanism or genetic explanation for the difference in variance in mental traits between populations of men and populations of women. The argument for the difference being biological/genetic/innate is essentially: "I can't see how environmental factors would have that effect. While I don't know how a biological mechanism would cause the effect either, I find a genetic mechanism I don't understand more plausible than an environmental mechanism I don't understand."

None of this says that an environmental explanation for the greater male variance in intelligence is proven, or that a genetic mechanism for it is impossible, or that it's wrong to raise such a mechanism as a possibility, or that research into the area should not continue. What it does say is that comments like Kleiman's (who again, I think highly of in general) that assume that that research has already been done and that we understand on a biological level why men out-achieve women at the highest levels of intellectual endeavor are thoughtless and badly supported.

Thanks for linking to this -- I wanted to comment on it when I first saw it, but he doesn't have comments.

LizardBreath (interesting sobriquet):

I agree that Kleinman would have been well-served to include a citation, and I don't know of specific research (yet another topic to occupy my insomniac nights!), but I can posit an evolutionary mechanism/reason for greater variance among men than among women - two, possibly, and is there a biologist in the house to slap me down if I'm completely off-base? Here I go, preparing to talk exclusively out of my [fill in body part here].

First: the Y chromosome is sort of "broken" - shorter, more prone to mutation - than the X. More mutation tendencies should equal more variance.

Second, and related to the first (this is pretty much pure conjecture on my part - I have a vague memory of reading something along these lines at some time in the past, but darned if I remember where or when): women "hold" the species; men introduce variance into the species. (Of course, mutation can occur in either parent - this is a generalization.) The mom's genotype, including her X, is passed (usually and generally) unchanged from mother to offspring; the other half-set of chromosomes, including X or Y, coming from the man, is the variant that keeps human babies from being clones of their moms (OK, keeps critters that reproduce sexually from being clones of their moms). From a natural selection perspective, the mom chooses, consciously or un-, from the pool of potential dads, trying for the best genetic pick to carry on her genes, with his tagging along. The species benefits from a large pool of applicants, and the male can "apply" multiple times with multiple partners (though he'd better NOT!) during the time that the female is reproducing, so there's no compelling evolutionary reason to limit variance in the male genotype beyond "disallowing" fatal mutations (or highly damaging ones, which will by and large be selected out by moms anyway): successful variants will increase in the next generation; less successful ones will proportionally decrease. So moms' choices will tend to converge on a set of characteristics that best responds to current conditions, while potential dads' change-happy Ys will keep on producing wide variance so that changed conditions can be met effectively.

The part I'm having trouble with is this: the above would explain women's choices but not their genotypes, I think. Well, maybe. Maybe it's just that the process repeats itself very effectively in each generation: mom chooses dad who converges on the "right" traits; male offspring, because of their short Ys, naturally have more variant tendencies; female offspring, with their long Xs, tend not to vary from "right" as much.

Paging that biologist...

Larry Summers is a wimp. Are men innately superior to women when it comes to mathematics? How important is the nurturing aspect? There currently is not enough evidence to resolve this matter. It remains open to legitimate debate. The woman who objected to Summer’s remarks made a fool of herself. Harvard University has once again disgraced itself. It is obvious that sucking up to the politically correct crowd is more important than intellectual integrity.

I believe (though I've lost my citations) that many forms of intelligence, like several other traits, are primarily linked to the X chromosome (so if you're a guy, your intelligence is mainly from your mother, not your father. If you're female, it's about equally from both). Since the male's intelligence is determined primarily by one chromosome, and the female's is determined by the convergence of two, the male has higher variability. But I could be wrong.

Jamie and Jadagul-

Both of your posts offer bull-session level suggestions for how genetic differences between the sexes might explain the greater male variance on intellectual measures. There's not a thing wrong with that -- bull-sessions lead to hypotheses, which lead to research, which leads to data. That said, in this case the research simply hasn't been done -- not because people aren't interested, but because our understanding of the relationship between genetics and brain structure and measurable mental abilities does not (again, to the best of my knowledge) allow us to pick apart the interplay between environmental/social and physical/biological effects on those abilities.

That research should be done; all hypotheses should be considered. But Klieman, and others who take the same position he does, whether it is considered or thoughtless, should not talk as if the formulation of a hypothesis that intellectual differences between the sexes might be biological is equivalent to evidence that they are.

For accurate and reliable dog health information, go to http://www.doghealth.mypetdogs.com

“But I could be wrong.”

That is why Larry Summers deserves to be rebuked for his cowardly behavior. He should demand an apology from this immature woman. Scientific research is concerned about truth---and not what is politically correct. Harvard University is a vastly overrated school. This incident show embarrass anyone who “earned” their degree from this disgraced educational institution.

LizardBreath again -

I'm trying all over the web to find more than just an abstract of a study performed by DF Halpern on the subject of greater male variance in intelligence and cognitive abilities. I saw it referred to by Judith Kleinfeld but so far haven't been able to locate the article itself. The upshot (from Kleinfeld, so part of this may be from other research than Halpern's) appears to be that certain kinds of cognitive ability, particularly involving visualization of spatial relationships, is linked to testosterone (elderly men, given testosterone to improve sexual function, also show improved cognitive ability in this area; transsexuals taking high doses of testosterone show an increase in spatial cognition and math ability and a decrease in verbal abilities). This datum doesn't specifically address variance, but seems relevant to the Summers discussion since it wasn't VARIANCE that caused Dr. Hopkins to leave in a huff. Concerning variance, Kleinfeld (poss. citing Halpern and/or others, possibly her own work - sorry, out of time to re-check sources if I'm to finish this comment) notes that males are demonstrably overrepresented at both ends of the cognitive spectrum: "The greater variability of males has been reported repeatedly in the scientific literature. See especially L.V. Hedges and A. Nowell, 'Sex Differences in Mental Test Scores, Variability, and Numbers of High-scoring Individuals,' Science 269 (1995): 41-45. See also Cole, The ETS Gender Study and Lubinski and Benbow, 'Gender Differences in Abilities.'" (Kleinfeld, "MIT Tarnishes Its Reputation With Gender Junk Science," 1999). Some of the measures of variance in which males are disproportionate are "afflictions [that] show up even before birth, before cultural influences have had a chance to kick in. Male fetuses are almost twice as likely to be involved in toxemia of pregnancy, spontaneous abortion, and birth trauma" (Ibid., citing Halpern). The preceding para quotes Halpern directly: "Males are overrepresented at the low-ability end of many distributions, including the following examples: mental retardation (some types), majority of attention deficit disorders, delayed speech, dyslexia (even allowing for possible referral bias), stuttering and learning disabilities and emotional disturbances" (Halpern, "Sex Differences in Intelligence: Implications for Education," American Psychologist 52, no.10 (1997): 1091-1102, p. 1102). So here are a couple of sources to support the "nature" side versus "nurture."

Here's an interesting factoid (please forgive if everybody knows this but me): Nancy Hopkins, the prof who stormed out on Summers, was also the chief complainant and principal researcher in the 1999 MIT study cited by reference in Kleinfeld's title above (the actual study was "MIT Study on the Status of Women Faculty"). A serial offendee, of sorts.

I wish I could get to the original sources cited in these articles! But I'm kind of housebound with little kids... not optimum research conditions.

I don't mean to be dismissive, but none of those studies (all, I am sure, good science, and all related to the issue) has a particularly strong relationship to the question at hand. The fact that a study shows that additional testosterone improves the mental functioning of men who, either through illness (the elderly), or through medical treatment (transsexuals), have unusually low testosterone levels says nothing about whether the underrepresentation of women at the highest level of mathematical achievement is related to their lower testosterone. If it is, it can't be a simple relationship -- the testosterone level curves for the sexes barely overlap, while most measures of mathematical achievement show equal means for the sexes but a higher percentage of men at the top.

Likewise, the other research you point to seems to report the existence of the greater varience of male abilities, not to provide a genetic/biological mechanism for it except in some cases at the low end. You can make the argument that a higher incidence of genetic retardation among men inherently implies that the higher incidence of expressed mathematical genius is also genetic, but I haven't seen any data supporting it. It's an interesting area for research, but the data that would settle it (i.e. whether intellectual differences between the sexes are primarily biological or social) aren't there yet.

LizBre:

Pardon truncation pls! You're correct - these were the first things I came across in web-only "research" w/no subscription to anything. They're by no means an exhaustive lit. search. But the first does bear on a biological reason for men to outperform women as a group in fields requiring this type of trait, by which I mean "Hopkins needs to do SCIENCE, not PC-ness" and respond to the available science, refuting it if she can, rather than just disbelieving it w/o backup. The second, while the snippet I found addresses only the low-end tail of the bell curve, does point to genetic cause (of unstated description) for greater male variance from which we would be reasonable to infer a similar mechanism at the high end - I don't know if the study addressed the high end at all. So far I can't find THE study that would bear on this question, so you may be right that it hasn't been done yet - but this Halpern thing is suggestive, at any rate.

Upshot for me: no grave wrongdoing in Summers' speech (possibly he DOES have the cite and assumed everybody else did too, or else he was just taking the "greater male variance is well established" that I am finding in many places, together w/sex hormone data etc., and anticipating study results - which was your point in the first place), self-damaging overreaction on Hopkins's part.

My point (and I can't address Summers' remarks in detail, because I haven't seen a transcript. I'm reacting to Klieman, and only possibly to Summers) is that no research on biological explanations for the gap in high-level mathematical achievement gets beyond vaguely suggestive. There are social/environmental explanations for the gap as well and the evidence for those is also suggestive (i.e., the gap used to be a hell of a lot bigger. It has closed a great deal due to social reasons.).

Anyone who takes the position that "Science has established that all or some of the gap in high-level mathematical achievement between men and women is genetically determined", as Klieman implicitly did, is mistaken. Science has done no such thing. To the extent that Summers took the same position, he was also mistaken.

(And to the extent that people abbrieviate my nom-de-net, they usually go with Lizard or Liz. Fewer annoying internal caps. But anything is fine.)

Fair enough, Liz! I had a lengthy discussion on similar topics with Cobra, a sometime poster, off-blog. It's devilishly hard to sort out socio-environmental factors in human intelligence... but I still maintain that Hopkins not only overreacted, but overreacted in a way uniquely difficult to reconcile w ith her training and apparent expertise. Proper response: to refute, or, if unable to do so off-the-cuff, to express an alternative hypothesis - not to reject the position presented because it didn't fit her world view. (It's not as if it's terra incognita to her, after all.)

Comments are Closed.