Jonah pointed me to this artice, The '60s Begin to Fade as Liberal Professors Retire, which chronicles the shift toward political moderation among the professoriate. That moderation seems to be less about changes in views toward the center of the American political distribution than it is a greater focus on career as opposed to striking a pose as a social revolutionary. As it happens, you can find the working paper which has most of the data which the article is based on, THE SOCIAL AND POLITICAL VIEWS OF AMERICAN PROFESSORS. Below the fold I've placed the most interesting graphs, where you can see one which shows that the majority of professors ascribe the underrepresentation of females in science & engineering not to discrimination (about 40% of professors are female).
Well I am glad to see that most people do not believe that the lower percentage of women in science is due to differences in ability (despite the views of some famous ex-college presidents, there is no data to support this idea).
Instead 75% of those surveyed believe it is because of a difference in interest. That could mean of course that those surveyed still believe that a career in science means that women who choose a science career may not be able to have a family (unlike men who are assumed to be able to do both).
Based on my conversations with students at UC Davis, I suspect that if the question was worded differently (e.g. "if you could be guaranteed to have a successful career in science and a family would you be interested in a science career or would you prefer a different career?), the results would be quite difference.
despite the views of some famous ex-college presidents, there is no data to support this idea
actually, i think there is (whether that data convinces or is an artifact or is ambiguous as to the question at hand is a different matter). second, differences of interest may not be easily malleable either.
Well I am glad to see that most people do not believe that the lower percentage of women in science is due to differences in ability (despite the views of some famous ex-college presidents, there is no data to support this idea).
It wouldn't hurt to keep in mind that the big report (NSF, NAS, whichever it was, it was the one with Robert Birgeneau as the sole male author) that was widely (if somewhat puzzlingly) seen as a rebuttal to Lawrence Summer's comments, agreed with the statement that there is substantial underrepresentation of women at the high end of the aptitude distribution. That is, it endorsed the properties of the aptitude distribution he discussed, agreeing that the high-aptitude people are predominantly male. Where it disagreed with his stated hunch was that it attributed this to social and environmental factors rather than innate factors.
Of course, we could just reject the report and say that it's full of it, but since it seems to be a favorite of people who disagree with Lawrence Summers, one can envision further problems with that strategy.
1) I want to see a breakdown of the data. Who thinks discrimination is a bigger factor, liberals or conservatives? Men or women? Those in science or those in business or those in humanities?
2) People at "elite" (defined how?) PhD granting universities are more likely to describe themselves as moderate. Does this mean they are more moderate, or the say they are? Are these people more suceptible to societal image pressure, thus leading them to pursue a status-oriented life at an elite university and avoiding views that might be offensive so they will be better liked?
3) Woohoo, I can be the first Marxist biology professor!
4) 52% of professors in "health sciences" are idiots.
5) @razib- show us yur data!
6) @ El Christador- never forget no one has yet devised a robust system for determing who has the "aptitude" to contribute to science. SAT scores need not apply. All we have is measures that correlate somewhat with who actually succeeds in science- which is obviously tricky to attribute 100% to true aptitude.
I want to see a breakdown of the data.
did you read the whole working paper? i read a lot of it and didn't see that info, but if you really are curious and it isn't in the working paper i would suggest emailing the authors (they said stuff in the text which wasn't in a table, so i could have missed it).
Does this mean they are more moderate, or the say they are?
you should really read the paper before you asking this question; they query them pretty deeply in regards to their attitudes toward various issues. i didn't reproduce them cuz they weren't interesting, i.e., professors are very socially liberal compared to the average american.
5) @razib- show us yur data!
no, i don't have the time or energy to really get in on this (IOW, i have things i need to do more than rehash why i think what i think, and why don't think what you might think). i just wanted to make sure that people are aware i'm not going to "amen!" to an unqualified summers bashing.
i'm sure someone will step into the void :-)
I can be the first Marxist biology professor!
richard levins and dick lewontin are marxists. jbs haldane and john maynard smith were marxists at one point, though smith disavowed marxism explicitly and the biographical information suggested that haldane drifted away from communist sympathy after the lysenko affair.
did you read the whole working paper?
I'll play the bad cop here -- Becca, don't expect other people to read the entire article to you as a bedtime story, free of charge no less.
"Oh yeah, well whaddabout this?! Ah hah, well whaddabout that?!" If you have so many questions, settle down and put in some damn effort.
My favorite three:
Larry V. Hedges; Amy Nowell (1995). "Sex Differences in Mental Test Scores, Variability, and Numbers of High-Scoring Individuals". Science 269: 41-45.
IJ Deary, G Thorpe, V Wilson, JM Starr, LJ Whalley (2003). "Population sex differences in IQ at age 11: the Scottish mental survey 1932". Intelligence 31: 533-542.
Ian J. Deary, Paul Irwing, Geoff Der, Timothy C. Bates, Brother-sister differences in the g factor in intelligence: Analysis of full, opposite-sex siblings from the NLSY1979, Intelligence Volume 35, Issue 5, September-October 2007, Pages 451-456.
@razib- I accept that academics are liberal- you should really read my comment more carefully before replying. I am wondering what is going on in a very specific subset of people- those at elite PhD granting institutions. Why should they be more moderate than people than people at other PhD granting institutions? It's an odd finding. Possibly it's just one of those odd things that come out if you measure enough variables...like 0.011 P values in sex VR ;-).
@agnostic- why can't I expect a free bedtime story? They do exist, if not here (Nature podcasts come to mind). If razib doesn't have a quick-and-easy answer to my questions, I fully respect him saying that. I will go look it up myself. But it shouldn't hurt to ask.
@rik
The first paper is a bit odd. I'm not impressed with Table 2.
Did you notice that, of all the cognitive tests, the one administered by the Army (NLYSA data set) displays the most variability in males compared to females? Strange.
(as an aside- that was a pain to look up without the issue #- don't cite many Science papers for people, do you?)
The second paper- 1932, Scotland, wtf mate? Am I missing something that makes that arcane paper particularly relevant?
The third paper is interesting. Women only do better marginally better in word recognition, but significantly better in numerical operations and radically better in coding speed? That is news!
Also, the biggest sex difference-men doing well in "auto and shop information". Yeah, no possible influence of socialization there.
For the record, I'm not disputing that differences can exist in tested performance. However, I do think some people mean something closer to the immeasurable "innate aptitude" when they say "ability" than whatever "auto and shop information" those tests examine.
Fact: Men and women differ in cognitive abilities - and this is bound to have an effect on their representation in academic activites, especially at the highest levels (because relative differences get much bigger a few standard deviations away from the mean).
Think of it biologically. The hypothesis that male and female humans (unlike any other sexually differentiated animals on the planet) have identical psychology (instincts, behaviours, aptitudes) has near-zero credibility.
The fact that men and women look different, are of different sizes and have anatomical structures should give us a clue that they have different psychology.
There are dozens of reasons why. But think of the brain. Steroid hormones such as sex hormones affect (essentially) every cell in the body via DNA transcription. Men and women have different steroid hormone profiles. The male and female brain have therefore been incubated and developed ocer many years in very diferent 'culture media'. How could men's and women's brains possibly end-up identical? They couldn't and they aren't. (Leaving aside that natural selection would not 'want' them identical.)
But this 'is there a difference' discussion is childish, I mean really...
*Of course* men and women have different psychology: the interesting questions are how they are different and by how much. And *of course* differences in men and women's psychology leads to differences in the workplace - the interesting questions are the nature of these differences.
The work of Kingsley R Browne is excellent on this stuff:
http://www.law.wayne.edu/Faculty/Fac_web/browne/index.htm
Obviously our societies are still far from full equality (even if there's been much progress in the last century). Therefore it's just impossible to conclude that if less women are active in sciences this is because of intrinsecal gender-difference reasons. It might be but certainly this is not the kind of data that could support such conclussion.
Example: my sister, a clever and disciplined person, an excellent student, wanted to study engineering. My father, who is an engineer himself, persuaded her out of that idea on his belief that it's a field dominated by males where he would have to strive a lot to suceed. She studied informatics and is now a succesful professional in that field and has two children. She's sufficiently valued at work as not to have any problems at all with maternity leaves and the like.
Instead my brother was a much more mediocre student and not as smart at all (not dumb, just average). He choose engineering and my father supported him blindly. He had serious problems to finish his career but well, he finally managed to do it.
At least my experience suggests, not just in the mentoned cases but in many others as well, that our culture has still a long way to go to reach full equality. We are still thinking in terms that discriminate women somewhat and until we do correct that fully (what may take several generations more) we just cannot conclude anything on this kind of data.
There was a great little bit of investigation that said that societies that were more equal had women perform just as well as men on maths tests. Frankly, that says more about how the dynamics of the fields work that a simple 'difference of interests'.
If there was such a huge disparity between the sexes, then humanity as a speices would be a bit screwed.
@Charlie - was this from the OECD 2003 student assessment? If so, then according to a recent issue of the Economist what that data actually shows is huge country-to-country variation in a gap in overall male-female achievement and relative stability of girls comparatively outperforming boys on tests of verbal ability. At the extreme, girls score better than boys on math and verbal tests, but still outscore them on verbal tests by about the expected margin.
Their interpretation of this is that at the same level of overall ability, women will have a comparative advantage over men fields that emphasize verbal ability over math, which could contribute to them favoring non STEM fields. All a nice way of saying that ability differences are part of the issue.
@Becca -- For the record, I'm not disputing that differences can exist in tested performance.
Then those difference may explain none, some or all of the differences in male-female STEM achievement and occupation. You'd need work showing, for example, that SAT-math scores predict the number of patents awarded at the population level, and so on.
However, I do think some people mean something closer to the immeasurable "innate aptitude" when they say "ability" than whatever "auto and shop information" those tests examine.
If by "innate aptitude" you mean general cognitive ability -- then that's what the IQ tests are attempting to get at when they measure the variable g. Overall the data are pretty clear that men outnumber women at the extremes of the distribution by 2 to 1 or more.
If by innate, you mean in terms of causal pathways -- sex differences being "innate" doesn't have obvious analogues to individual differences being innate (i.e. heritable). IMO, it's hardly a useful concept when applied to sex. The sexes differ in one very systematic way with respect to DNA (X and Y copy number) and everything else that follows is a matter of different developmental trajectories. The causal pathways that lead to those outcomes may or may not be environmentally malleable (at what cost?).
But it shouldn't be obvious, as you seem to assume, that sex differences in some specific traits (e.g. "auto and shop information") are more easily modified than others. Maybe the relative male obsession with objects is harder to overcome than whatever kicks off the male advantages in spatial rotation. Based on the international data I mentioned above, I think it's more likely that the specific ability differences are rather resistant to change relative to overall performance levels, which appear to be somewhat malleable.
@BGC- well obviously men and women are ( on average, an important qualifer you seem to have ignored) physiologically different!
Although (as an aside), I wouldn't be at all suprised if the differences turn out to be based more at the epigenetic level than at the level of transcriptional factors (anyone else play around with the idea of X-inactivation chimerism as a partial explaination for fewer women in the tails of a given distribution?).
More on topic- of course sex hormones are incredibly important in shaping psychological traits (any self-aware person who has ever taken them exogenously [e.g. birth control pills] can probably attest to that somewhat).
Though some of the most interesting stuff on hormones may be in the fetal testosterone our developing brians 'marinate' in- I'm not so sure the adult levels matter as much for direct cognitive effects.
Natalie Angier's Women: an intimate geography is great for fun reading on that sort of thing.
I'm not sure Browne is wrong- but I do think that, if he is correct, there are still incredibly important questions. For example- if career sucess in science is so dependent on ultra-high (rare) mathematical genius, why is that so? Could not humanity be better served by encouraging the non-spatially rotationally gifted among to also contribute to alleviating suffering* (or any other noble aim you might use to justify the pursuit of science)?
(*mind, I'm sure that those spatial rotators have something very special to offer big-pharma when it comes to designing new drugs ultra-quick... why, just purify out the enantiomer and re-patent!)
@rik
Naturally, a priori, I would say differences in tested performance could explain none, some or all of the difference in women's success in STEM fields. However, given what is known about discrimination , I find the hypothesis that all of the differences are due to differences in ability (as measured by the tests) to be implausible.
That said, you are right, the impact of ability is a very interesting question!!! Therefore, it seems to me there is only one reasonable thing to do- we should all immediately set to work eliminating the discriminatory BS so we can see how much of the difference is due to differences in ability and psychology. Now, distinguishing out those two factors will be tricky, but at least getting rid of the discrimination is a start.
Aside from outright discriminatory BS, these is a fair amount of socialization in play for some of these factors. For example, with respect to "auto and shop information"- I have no way to know if ability in this area is easily modified in women in general. I do know, however, that it should be highly malleable in me personally. I am the apparently exotic female residing in the upper-ability tail in sptial rotation tests. Furthermore, I have a decent amount of interest in "auto and shop information". I don't know how I would test in the Army's exam, but I'd be willing to put good money on the bet that if I took some time to learn a 4-h project worth of material, I could do much better than I can now.
I leave it as an exercise to the reader to determine why I picked art and computer 4-h projects, but no auto or shop ones.
rik, once again, where are the data on g and gender differences that show this 2:1 (or greater) ratio in the tails? I'm not saying I don't believe you, but I've already mentioned what troubles I had with two studies based (partially or fully) on Army career tests and a study of 1930s Scotland. They're cool studies and all, but when it comes to "g"....I get the distinct feeling "You keep using that word. I do not think it means what you think it means".
Hmm, maybe I'm reading this wrong but doesn't table 3 show that infact people at "elite" phd-granting institutions are the least "moderate"?
Yet at the end, table 14 seems to be saying the opposite.. I guess when being given the choice between "center-left" and "liberal" vs, liberal and moderate, people choose differently.
As for the whole discussion on the ability (or lack thereof) of women, we don't seem to be in any danger of anyone actually doing a good enough study to somehow separate out societal vs. "inherent" factors, or to even get a good standard of what "inherent ability" is.
As such I think the better question is, practically, do we want to go around telling women that they are not capable or interested in doing science? I think the answer there is quite clearly, no.
Let's not forget the culture OUTSIDE academia as well. Just a few months ago I was at a family gathering, where one of the girls -- around age 13-14 -- was being asked what she wanted to do. She's a bright kid, knows how to do proper blueprints and is very good at math -- I hate to say, but better than I am, she picks up the principles very quickly -- and she expressed an interest in building things, maybe being either an engineer or an architect. The immediate response from several relatives, including her mother, was "oh, honey, you won't like it. And you'll have trouble getting boys to go out with you! You should look at something like graphic design, instead."
You can yawp about "innate ability" as much as you like, but until you eliminate that kind of feedback from the environment, you don't know.
Business professors and Nader? What an interesting mix.