« Comment on the NAS Report on Women in Science and Engineering-BECKER | Main | Women in Science; DDT and Overpopulation--Posner's Response to Comments »



Feed You can follow this conversation by subscribing to the comment feed for this post.

U of C guy

This is very typical of "must I believe this" versus "can I believe this"

Becker says sex discrimination causes a lack of participation on the sciences by women. A study comes out and supports it, he asks himself "Can I believe this?" Because it supports his thesis.

Then Posner ("That pesky consitution just gets in the way of proper law enforcement") sees the same study, disagrees with it and asks himself, "Must I believe this?"

See, it doesn't agree with his view, so he requires it to be held to a higher standard than Becker.

Lawrence Indyk, University of Kansas School of Law

The study does discuss at length the detrimental impact of family caregiving on female careers (as is true in most professions) but still rests mostly on its conclusion of pervasive structural gender discrimination and subtle personal bias being the predominant factors in differences in numerical representation and career outcome between men and women in the sciences.

Depending on one's standard of evidence, however, this claim could require more or less evidence to be persuasive. There are several ways to evaluate the claim that have different burdens of proof. You could require that the study show significant evidence excluding the likelihood of contribution from other factors, or you could conclude that differential representation is, in itself, sufficient evidence of discrimination. Suffice is to say that meeting a scientific standard is more difficult than meeting the legal standard of disparate impact which does not even require any discriminatory intent - overt, subtle, or unconscious.

I would think that if the study stands up to scientific scrutiny then it would more than support a Tital VII action against almost all top tier institutions. Conversely, if one believes that such an action were likely to fail in court for lack of convincing evidence, it would most certainly fall far short of being scientifically convincing.

So, if the authors of the study are truly convinced that their work accurately describes the reality of hiring and promotion practices, provides sufficient evidence to be persuasive, and that action is desperately needed, then they should be encouraged by the fact that the legal system already provides an opportunity for redress should their recommendations not be enacted.


Before the discussion begins, I'd like to ask about something that wasn't mentioned here. I wonder if anyone has hypothesized yet about path dependency and/or the Schelling model? I bring it up because it's plausible that at least part of the lack of women at the top science schools is caused by the fact that there have been few women at such schools in the past. Thus, as a result of historical accident [past discrimination against women in education] the relevant fields came to be dominated by men. Even if all discrimination were lifted [which it probably is not, though I'm sure it has declined], it is very possible that most women don't want to enter a field where they would be a minority. The Schelling model explains self-segregation by positing that no group wants to be a minority in its immediate neighborhood. It's plausible that women simply don't want to enter a field where this would be the case. [The same probably applies to men who consider top fields dominated by women, like the mentioned primatology.]


I can paraphrase the beginning of Judge Posner's essay for you:

"The report claiming discrimination agaist women in academia doesn't have to be taken seriously, after all, it was written by WOMEN."


I have not read the NAS report, so I cannot comment on it. But the biggest question, to me, is why does Judge Posner seem to be invested in the theory that men are "innately" superior to women in science?

I have a few theories:

(1) Judge Posner believes that Larry Summers was shafted and wants to avenge this perceived wrong. Perhaps Posner is correct that Summers suffered disproportionately for his inartful musings. But, as was discussed at length in previous blog entries, I believe that Summers was forced out because, first, he handled a sensitive issue indelicately, and, second, he had already drawn the faculty's ire with his combative attitude toward reform. He was a victim of his own series of blunders, not just for suggesting that "innate" science ability ought to be studied.

(2) Perhaps Judge Posner objects in general to the idea of making professions more "women friendly" (I am just speculating here; I don't know this for a fact). If that is the reason, I think Judge Posner is wrong, once again. Perhaps this is a problem of terminology. In my opinion, all professions -- science and law included -- should adopt "family friendly" policies that allow men and women professionals to excel at their jobs and also to have a personal and family life. Unfortunately, too many of the best and brightest feel that they must choose between a family and a career. This is unnecessary and oppressive. While most professionals are well paid, they still should not be treated as nothing more than cogs in an ever-turning wheel. We can trade a bit of productivity in return for a bit more *life*. Moreover, it is undeniably true that the demands of professional life cause women, disproportionately, to forego career opportunities.

(3) Perhaps Posner sincerely believes that men are better at science than women. If so, he ought to take that up with Marie Curie. And, more seriously, if he styles himself a scientist (to the extent that he is an aficionado of economics, and economics is sometimes called the "dismal science"), he ought to know better than to accept such generalizations absent hard evidence of a genetic link to science ability.

Just some thoughts to consider, anyway..

Mr. Harvard

Larry Summers was forced out of Harvard for being a bore. The battle of the sexes was just the final nail, he was uniformy hated by the faculty due to his management style.

It's easy to say, "Summers got forced out because he refused to be politically correct", but it is more accurate to say, "Summers management style was atrocious, so he got shown the door."


Once my computer broke down. After failing to solve the problem myself, I called the technical support of the computer's manufacturer for help. The moment I heard the voice of the tech support person, my heart sank: the voice was that of a woman speaking with a distinct African American accent. I would hate to admit, but I am biased against women and blacks when it comes to a matter to do with engineering or technology. The bias was baseless because the woman helped me solve the problem without a hitch. I am a male professor in science at a major research university. The bias is wrong, is in existence, and yet very difficult to overcome in the absence of a constant and explicit warning. Shalala's report, however deficient it may be, serves as one of such warnings.

There are more female professionals in law and medicine than in science and engineering. But there is reason for why in medicine and law women are sometimes more preferred than men. When we were looking for a family doctor for our daughter, we insisted that the doctor be a female, perhaps on the subconcious belief that a woman doctor would be more caring. When a female friend of mine was going through divorce, she made it very clear that she wanted to talk to a female lawyer first. No such factors exist in science, math and engineering.


This NAS report kind of reminds me of a George Bernard Shaw quote, "Those who can do, those who can't, politic". Just another broadside hiding behind the NAS from the gender/race front. Instead of spending so much time, effort and resources on something like this, why can't we spend them on doing science/engineering instead?

This also reminds me of another GBS quote, "A fools brain digests philosophy into folly, science into superstitution, and art into pedantry. Hence a University education. As for myself I'm going back to the Lab and get some real work done. ;)

a physicist


Marie Curie is pretty much the only outstanding female physicist. She was also an empiricist. This is why people wonder if perhaps male brains handle abstract stuff (relativity for example) better than females that handle the concrete (radiation for example). Personally I don't know what all the factors are behind the segregation, I do know that when you teach a class of males and females how to do a problem in a left handed coordinate system (rather than right handed) the females tend to have a much harder time performing the rotation in their heads.

Tracy W

Before the discussion begins, I'd like to ask about something that wasn't mentioned here. I wonder if anyone has hypothesized yet about path dependency and/or the Schelling model? I bring it up because it's plausible that at least part of the lack of women at the top science schools is caused by the fact that there have been few women at such schools in the past.

Yet in the past there were few women in law or medicine, and yet far more women have entered those professions when they were all opened up.

And even earlier, all lecturers in English or History or Art History or the like were male (because all lecturers were male, and universities only accepted male students), yet women have entered those subjects very happily.

So any such path dependence model would need to explain why path dependence would matter for the hard sciences, but not for law or medicine, or the humanities and arts degrees.

(I am female and have degree in electrical engineering).


Simple Logic:

If one gender's brain is wired in a way that predisposes it to a certain activity, then a larger number of members of that gender will engage in that activity. Thus, odds are that the "best of" that activity will be of that gender. That does not mean that it's impossible for the "non-predisposed" gender to be the best, but it does mean that it's unlikely.

I have something else that is really off the wall and completely subjective, but may tie in somehow. Many people who are interested in science are also interested in science fiction and, if they are young, video games. For the most part, women abhor both of these things (e.g., Star Trek conventions and video arcades aren't the best places to go looking for women).

Video games are actually a perfect example of something that illustrates how the sexes are predisposed to different things. When video games first came out, was it taboo for women to play them? No. Were games like "Pong" and "Pac-man" geared to a specific gender? No. Were arcade owners discriminating against women and driving away their quarters? No.

Software companies have been trying for years to create video games that appeal to women. Not only are they not discriminating against women, they are actively trying to draw women into their market base. And although they have succeeded a few times, their problem remains: girls are not interested in playing video games.

Granted, math and science aren't as "gender polarizing" as video games and science fiction, nor do all scientists like science fiction and video games. But I think this correlation does mean something...


I wonder if anyone has hypothesized yet about path dependency and/or the Schelling model? I bring it up because it's plausible that at least part of the lack of women at the top science schools is caused by the fact that there have been few women at such schools in the past.

While I intuitively agree that a historical lack of women in the science departments at top universities probably plays a substantial role in this continuing lack thereof, I am unsure as to how applicable Schelling's model is in this scenario. As you say, he did his work with racial segregation in spatial modeling, finding that neighborhood seclusion by race will naturally happen as people move away from neighbors who may be prejudiced against them.

But moving to a different neighborhood presents far lower transaction costs than moving to a new career. As has been noted in both academic and popular literature, the problem is not a lack of women obtaining doctorates in the sciences. The problem is that those female PhDs then somehow fail to obtain tenured positions at top universities. If Schelling's model was at work here, wouldn't the expected result be that women would just choose to not enter the science field at all, and instead apply themselves in an area where they were not a minority? Why would they go so far as to earn that PhD, but then drop off somewhere along the line afterwards? It seems that a path dependency model cannot explain the full extent of the phenomenon here.

Marie Curie is pretty much the only outstanding female physicist.

Of course, it depends on what you mean by "outstanding," but Hedy Lamarr, if not an academic physicist, did invent spectrum hopping: www.demimondeonline.com/blog


It is ironic that those of a particular political mindset react with venom against anyone who makes generalisations about aptitudes and anything with which they happen to disagree, yet themselves make endless generalisations about the existence and causes of bias and discrimination.


I don't think institutional factors or some kind of discriminatory measures are still working against women's performance. To a large extent, both men and women are playing in a level field now, especially in these intellectual areas, in which physiological differences donot matter much.
But how can we explain women's underperformance than men in maths and engineering areas etc., as is the plain fact. Just as professor Becker said, Women have a lot of other things to do, they usually share more household responsibilities, thus cannot possibly devote as much time and energy to their work as men.Still, I think another reason may also contribute to this problem.It's relevant to women's preferences. Although women's participation in the labor force has increased a lot these years, but their involvement are still not so extensive as men's, probably in these scientific areas.Maybe some really high-potential women just don't want to do it.
Finally a small defense for Larry Summers. I don't think there's any tone of discrimination in his remark on women's performance in science and engineering.What he said is just a possible explanation to the situation.What if future researches did show the innate differences have led to this?Will you still say:"oh,God, why are you discriminating me?" I think men and women are born to be distinct in this society, even if genetic differences make them perform differently, whether the worse side are men or women, what bad was it?


Are men better basketball players than women? Is this due to discrimination, innate differences between the sexes, or both? It seems that society has no problem in accepting that innate (i.e. physical) differences account for some of this performance gap in sports. Why do we refuse to accept that this could be the case in some fields of study? The NAS should have recommended we study the human brain more closely, that would be the best way to get the bottom of this debate. Maybe a woman would have been the scientist to make the ground breaking discovery!


Men are more prolific murderers as well. Could we extrapolate that the police/jurys are biased against them.

From what I remember of the sumners deal. One of his assertions was the performance distribution was different between men and women in science.... there was more variance among men. Is this an accepted fact? if so do some try to explain it by citing bias?

Trevor Butterworth

Dr. Rebecca Goldin and I have published a piece on www.stats.org which may be of interest to the owners and contributors to this blog. It's a tad too long to post the entire piece, so I encourage you to follow the link to


Curveballs: The Fuzzy Math Behind Sex Discrimination in the Sciences[/b]
October 2, 2006
Trevor Butterworth and Rebecca Goldin Ph.D
As the evidence for sex discrimination in the sciences mounts, media pundits continue to cite math test scores for innate differences between women and men. Here’s why the numbers don’t add up

Best -

Trevor Butterworth

Yevgeny Vilensky

Mr. Butterworth's and Dr. Goldin's analysis is quite lacking. Again, it is very easy to dismiss the "pundits" who are not statisticians as being idiots. It's quite another thing to actually prove that it is _impossible_ for innate differences to account for lack of female representation in the higher echelons of the physical (as opposed to life) sciences, which is the attempt of the article he links to.

For example, they cite a survey of mathematicians and physicists as to what they believe. But presumably, since they are not cognitive neuroscientists, they have no more authority on the subject than the economist Larry Summers. I'm not sure why their opinion of what the truth is ought to trump what the actual facts are or be indicative as to why we should not pursue this line of research.

Here is an example. Microbiology (esp. virology and immunology), currently, is dominated by female graduate students, while the faculty is predominantly male (no doubt due to past discrimination against women in addition to previous pressures on women to stay home and raise the kids rather than burn the midnight oil at the lab). Biochemistry, a very closely related female, still has predominantly mostly male graduate students. So, a) it's not really the lack of female mentors or b) the daunting task of entering a male-dominated field that are keeping women out of biochemistry. It is either due to lack of interest, or due to the lack of ability to perform what are often more mathematical (e.g. a substantial portion of biochem involves molecular simulations in addition to benchwork) tasks.

Let's not look at SAT scores. Let's actually look at the research Summers was discussing (which the organizers of the NBER conference actually asked him to look at). This research was by a cognitive neuroscientist at Ohio State, who was using serious experimental investigations of young children's brains in her research. So, it's not like Summers speculated out of thin air by looking at SAT scores. It seems that there is some actual scientific research behind this. When people talk about the SAT, even those stupid pundits, they are using it as an example of where this innate difference appears, not as evidence in and of itself, I think. Maybe some are, but I think what they are trying to say, even if ineffectively, is that this is but another symptom.

Finally, why might SAT scores matter? Well, if a student gets a 650 on the math portion of the SAT (still well above the mean), they are likely not to become research mathematicians or physicists worthy of tenure at Harvard. Those people get probably somewhere in the 770-800 range. So, yes, SAT scores do measure the ability to do simple mathematics in a time constraint. But, I can't imagine there being many distinguished physicists or mathematicians (male or female) who scored below a 750 or so. Just because one scores an 800 doesn't mean he will understand Feynman Path Integrals. But it does mean that a person who got a 650 probably never will.

Yevgeny Vilensky

By "closely related female", I meant "closely related field."

Chris N


There have been some good scientists who also happened to be women. Cecilia Payne and Lise Meitner (See Otto Hahn) in addition to Marie Curie. That aside,

anyone truly interested in science does not have the time nor energy to devote in defining their ideas within the boundaries of current social science, politics or gender politics of the day. At least not their scientific ideas.

Anyone interested in science must keep a fundamental sense of wonder and awe at how the natural world works. This includes a solid base in mathematics and physics. Trust me, lots of people drop out. One must be very persistent, against-the-odds persistent, to make discoveries among a community of people who have grappled with similar problems, and enriched one another through experimentation and competition and discussion.

What disturbs me is that the best social science, and even the subtly and analysis that the best jurists offer, is not the same as the best science. Each field has its great benefit to mankind, but to mix them stridently is foolish.

Redmund Sum

It is interesting that many people, mostly from the PC crowd, steadfastly refuse to admit the possibility that there is any innate aptitude differences between male and female when they argue about women in the hard sciences, yet in the next paragraph, they say the when they choose a family doctor for their daughters, they would insist the doctor be a woman doctor because she would be more caring.


Young women make a decision, either consciously or unconsciously, about which profession to pursue. For example, let's say a bright and academically motivated young female graduate has the following career options: medicine, law, business, science/engineering. If she also desires to have children during the next 10-15 years, she knows that she may need to ramp down her working hours during that period. The bright young woman considers her options:

1) Medicine - She can begin medical school right away and will finish at about the time she's ready to have children. Once she's finished her residency, she will have strong earnings potential and she may even have some scheduling flexibility. She could work in a hospital or private practice, and she could work days or nights and weekends. She is also entering an attractive pool of potential marriage candidates...

2) Law - She could begin law school right after graduation and start earning a strong salary at a law firm once completed. Once she has children, she could make a decision to ramp down her hours and go corporate, or leave the workforce altogether. Either way, she's likely earned a significant sum during her pre-baby work years.

3) Business - Business school is a complicated path for the young woman due to timing. Expected returns appear low because she must spend 3-5 years working for a relatively low wage, followed by 2 expensive years in business school. When she leaves business school she really needs to outperform at work, but she is conflicted because she is also at the age where she wants children. A further complication is the commonly-held view that business women have difficulty returning to their business careers on the same trajectory after taking time off or ramping down for child-rearing.

4) Science/Engineering - She could pursue a career in science and engineering, but the sad reality is that expected earning potential is just not as strong as it is for the other professions. If she wants to take some time away from work to care for children, then she knows that she needs to maximize her earning potential during the years that she does work. She knows that a career in science or engineering, while it may be intellectually stimulating, may not produce the level of income she desires during her shorter working years. Further, similar to business professionals she has the complication of difficulty returning to the work force after taking time off. Not to mention that the pool of potential marriage candidates may be less appealing... just my honest opinion!

I happen to be a young woman that faced the above options and did pursue a career in engineering...only to realize later that the value proposition was insufficiently attractive. I went on to business school and now question whether continued work or more time with my child provides greater utility for me and my family. I don't know the answer, but I do believe I would have chosen another path if I could do it again.

I realize that my views are biased by personal experience and observation. I do believe that more women would pursue careers in science or engineering if the potential payoff were greater. A better question might be why so many bright and capable young men work in the sciences if they could earn more in other fields!!!

Larry Summers

A better question might be why so many bright and capable young men work in the sciences if they could earn more in other fields!!!

Perhaps they have an innate desire to pursue scientific inquiry that is more intense than their materialism.


"Perhaps they have an innate desire to pursue scientific inquiry that is more intense than their materialism."

That's absolutely a possibility! A similar rationale follows for women that choose care-giver careers (nursing, early childhood education, etc.) despite lower relative salaries.

That is not to say that individuals do not make value judgements in weighing one career path compared to another. So the "innate desire to pursue scientific inquiry" must have implied value greater than the salary delta.

I find this somewhat surprising, though. The mind is amazingly flexible and can find challenge and intellectual stimaulation across a wide range of problems or work endeavors. So why not pick an equally intellectually challenging job that happens to have a higher market value?

Perhaps economics pays better than physics, but I imagine that both offer opportunities for scientific inquiry...


Shalala was the Chancellor of the University of Wisconsin when I attended its mechanical engineering program. My class consisted of several hundred males and two or three females. The women did not excel.

Could the answer be as simple as noting that females have a 3-4 point lower IQ in general (Prof. Richard Lynn) and a sharper differential in math (per the 2001 SAT, men scored 533 while women scored 498; year-after-year results are consistent)? Why do our progressives have such a hard time with the numbers?

My wife is so good with our daughter, better than I could ever be. Is it possible that God and/or evolutionary pressures have caused the genders to develop specialized skill sets?

The comments to this entry are closed.

Become a Fan

May 2014

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31