The Gender Gap in STEM is NOT What You Think

In a new NBER working paper David Card and Abigail Payne have a stunning new explanation of the gender gap in STEM at universities. The conventional wisdom is that the gender gap is about women and the forces–discrimination, sexism, parenting, aptitudes, choices; take your pick–that make women less likely to study in STEM fields. Card and Payne are saying that the great bulk of the gap is actually about men and their problems. At least that is my interpretation of their results, the authors, to my mind, don’t clearly state just how much their results run against the conventional wisdom. (Have I misunderstood their paper? We shall see.)

The authors are using a large data set on Canadian high school students that includes data on grade 12 (level 4) high school classes and grades and initial university program. Using this data, the authors find that females are STEM ready:

…At the end of high school, females have nearly the same overall rate of STEM readiness as males, and
slightly higher average grades in the prerequisite math and science courses.  The mix of STEM related courses taken by men and women is different, however, with a higher concentration of women in biology and chemistry and a lower concentration in physics and calculus.

Since females are STEM-ready when leaving high school you are probably thinking that the gender gap must be a result either of different entry choices conditional on STEM-readiness or different attrition rates. No. Card and Payne say that entry rates and attrition rates are similar for males and females. So what explains why males are more likely to take a STEM degree than females?

The main driver of the gender gap is the fact that many more females (44%) than males (32%) enter university.  Simply assuming that non‐STEM ready females had the same university entry rate as non‐STEM ready males would
narrow the gender gap in the fraction of university entrants who are STEM ready from 14
percentage points to less than 2 percentage points.


On average, females have about the same average grades in UP (“University Preparation”, AT) math and sciences courses as males, but higher grades in English/French and other qualifying courses that count toward the top 6 scores that determine their university rankings. This comparative advantage explains a substantial share of the gender difference in the probability of pursing a STEM major, conditional on being STEM ready at the end of high school.

Put (too) simply the only men who are good enough to get into university are men who are good at STEM. Women are good enough to get into non-STEM and STEM fields. Thus, among university students, women dominate in the non-STEM fields and men survive in the STEM fields. (The former is mathematically certain while the latter is true only given current absolute numbers of male students. If fewer men went to college, women would dominate both fields). I don’t know whether this story will hold up but one attractive feature, as a theory, is that it is consistent with the worrying exit from the labor market of men at the bottom.

If we accept these results, the gender gap industry is focused on the wrong thing. The real gender gap is that men are having trouble competing everywhere except in STEM.

Hat tip: Scott Cunningham.


Of course, the non-STEM fields have easier grading on average.

Schools don't have any interest in solving that issue either.

Because they don't need to compete for student, because, as usual, on barriers to business

Because they don't need to compete for student, because, as usual, of barriers to business

Then why don't men go for those easy subjects?

Preferences: Things vs. People.

If you work for Google prepare to be fired :-\

Perhaps because (1) high school is more soul-sucking for men and (2) men are ahead of the curve is avoiding the bullshit jobs [that may be driving the increasing unhappiness levels of women]. I could call these speculative, but (1) is damned well true and I think (2) could be studied and found to be true also.

Because they leave you with debt and nothing to show for it

Men do go for the easier subjects and end up as laborers in fixing cars, fixing plumbing, and other low wage jobs that women do not want to do like garbage collection and sewer maintenance.

SJWS/Feminists would rather attack the males in STEM that do take the harder classes, and do find better paying jobs because they have experience in STEM areas. SJWS/Feminists want women that take gender studies to have the same job and salary as a man with a computer science degree and more STEM classes and more experience in STEM, and the woman even if she has no STEM experience should be paid the same.

Excuse me, I was forced out of my STEM job, I have to go dig some ditches now.

"Low wage jobs"? A skilled plumber can make more money than most humanities Ph.Ds.

Because men feel they have to make a living and support a family.
A degree in Anthropology is less likely to provide that.

Not necessarily easier grading, just more subjective.

Is that really true of the things we are talking about here?

Foreign language study is not easy (although particular curricula vary hugely in what they expect of the student). Also English class at high school is not a particularly soft course from my memory.

Like biology and chemistry?

I think I would find it easier to get a B or better in a STEM course than, say, French.

'ave I misunderstood their paper? We shall see.'

I'm sure we will, but part of the charm of reading this web site is how being right or wrong is unimportant to what is posted. See, you don't even have to read to the end to find out just how true that observation is - ' I don’t know whether this story will hold up but one attractive feature....'

'The real gender gap is that men are having trouble competing everywhere except in STEM.'

Well, that and the number of faculty in many programs are often notably male, not that one would expect a member of the GMU econ dept. to notice that in their daily routines.

Oddly, every blog post is not peer reviewed.

Hint: Tyler's just starting a conversation.

Hint: it's not Tyler.

You can usually tell who it is from topic alone.

Sorry Alex! But this is more of a Tyler post.

Too Much Cuck

From topic, nor so much. But fro the style, obviously.

You can usually tell who it is by the grammar. Tyler does not produce run-on sentences like this:

"At least that is my interpretation of their results, the authors, to my mind, don’t clearly state just how much their results run against the conventional wisdom."

Tyler also wouldn't write "NOT" for emphasis

"... but part of the charm of reading this web site is how being right or wrong is unimportant to what is posted. "

Well now that's an ironic comment.

This looks fascinating. From the abstract, it sounds like a straight-up example of Simpson's Paradox. (Read up on that, it makes for very interesting reading.)

MMMM . . . Donuts.

I can't explain it. Why can't more women do math and science? Is it emotional, physiological, psychological?

Not to worry! Young boys are circling the aggregate, academic drain at rapidly accelerating rates.

Judea Pearl has been making your point, and in this same context, for a very long time. Here's a short write up:

Averages/means are dimensionless and yet many bright people continue to think they can be used to measure things in the same way that a yardstick measures length.

"Averages/means are dimensionless": not in general. What did you actually mean?

Average/means have the same dimension as their population pool. IE the average height of an American male is 176 centimeters.

Shouldn't 560 million meters divided by 322 million people equal 176 centimeters-per-person?

Why is "people" in the denominator? The fact that they're people is irrelevant, you're trying to find the mean of a series of heights measured in centimeters, that's the only unit in the calculation, not people.

I took it to mean you can't say whether the numbers being averaged represent like things sufficiently to describe with perfect accuracy the dimension of the average.

I can think of three possible explanations:

- radical feminists discriminate against men in gender studies departments. Humanities departments are not a welcoming environment for men

- unqualified jobs (construction, factory work, farming) favor men over women. Hence unskilled women prefer to go to college for a soft degree than men (also, women who want to get married instead of working can go to college and study a soft major in order to meet men)

- some men still associate a negative sigma with some non-stem professions like nursing

Some of these hypothesis could be discriminated looking at the data with more detail

You left out some obvious hypothesis:

Males prefer Challenges, Females prefer Security. So men tend to gravitate to professions that are mentally and physically harder and have high stress and women tend to gravitate to professions that are easier and have low stress.

The Bell curve for Men is shallower and wider than the Bell Curve. So, any job that selects for either end of the Bell curve will tend to be heavily populated by Men and jobs toward the center will tend to lean towards Women.

That's not obvious at all. Isn't studying STEM much more secure? Even a mediocre engineer can find stable employment, but only the very best Medieval History PhDs get tenure.

"That’s not obvious at all. Isn’t studying STEM much more secure? Even a mediocre engineer can find stable employment"

Nah that's the big lie about engineering. In reality it's as much a passion profession as being an artist.

Maybe it's about passion for the people who insist on working as code artisans at hip startups in expensive locations. There is plenty of stable boring work in logistics, advertising, corporate infrastructure, networking, security, administration, support and so on.

Lots of Indians willing to do that kind of work for pennies an hour too.

Lots of Indians are also very good nurses and teachers.

I don't know if it's what JWatts meant, but the security could be security in actually finishing your degree. STEM majors have a much higher washout rate than the humanities. So, when choosing your major, going with an easier humanities degree is the safer choice (for college).

JayT is onto something.

"security" for 18-22 years old means passing all college classes. no one is thinking on income 5, 10 or 30 years later.

More secure? Not compared to 'Eds and Meds' jobs. Most male-dominated STEM jobs are private sector, while most female-dominated 'Eds and Meds' jobs are public sector with much greater job security.

"...but only the very best Medieval History PhDs get tenure."

Well that assumes that females getting degrees in Medieval History are planning a career in Medieval History. Generally speaking most students going that route aren't planning on being a professor. They are pursuing an interesting degree and delaying entry to the work force. And frankly who thinks that Medieval Historian is a stressful job?

Academic tenure in the humanities is about many factors other than academic performance. It's passing the ritual of induction into the tribe. Ideological purity is demanded, taboos must be honored.


I was replying to this: "Even a mediocre engineer can find stable employment, but only the very best Medieval History PhDs get tenure."

There are simply very few Medieval History positions. Probably 20 a year max. (The Medieval society lists approx. 100 schools that specialize in Medieval history.)

#2 seems quite correct to me. One question for discussion might be: in the market for skilled-but-not-quite-professional labor, why have primarily female jobs (HR, nursing, elementary school teaching) trended towards requiring more & more college degrees, even master's degrees, while primarily male jobs (electrician, auto mechanic) have not?

Is nursing considered non-STEM? I guess you could draw a distinction between CNAs, on one end, and Nurse Anestheseologists, on the other. Or maybe STEM is not really that useful of a distinction.

#2 seems quite correct to me. One question for discussion might be: in the market for skilled-but-not-quite-professional labor, why have primarily female jobs (HR, nursing, elementary school teaching) trended towards requiring more & more college degrees, even master’s degrees,

Because the HR departments in most large companies have utterly fucked everything up.

Yeah that's why the S&P500 is at all time highs, because of how fucked up those companies are.

It's at an all time high because we finally have adults in Washington.

A hurricane hits Florida, but ten years later the GDP is higher. That doesn't imply the hurricane was a great thing, just that there was something that happened to cause GDP to go up. Probably not the hurricane.

You really think the credentialist rat-race is rational?

I think credentialism is extremely damaging, no analysis of labor markets is complete without looking into the harm done by modern HR ideology.

Granted I haven't done a study, but I inherently doubt that adding more masters degrees per HR drone per capita is a meaningful driver of GDP growth.

The HR department of the hospital that hired my sister-in-law as an RN sent her a letter months after hiring her saying that unfortunately they could find no position for her. It's astonishing the hospital has any staff at all. People figure out how to go around the HR department.

Radical feminists? When I went to college as an engineering/premed male, I felt no hostility from the humanities departments. In fact, a male friend was having a ball in a woman-dominated field. The hostility was in the reverse direction: toward the non-majors as we called them (short for non-science majors)

Unqualified jobs is a weird way of putting it.

Nursing isn't STEM? The S in STEM is science. Is medicine not a science?

Think a bit harder.

"The hostility was in the reverse direction: toward the non-majors as we called them (short for non-science majors)"

I would agree with this. I use to call them hacky sack majors. The guys that you'd see playing hacky sack on the quad as you'd trod into your class. And still see playing hacky sack when you trod out and into the library at 4 pm for 3 hours worth of homework. I never saw them leaving the library at 7 pm. They were long gone.

But there wasn't a lot of deep hostility. Just the general muttering about it must be nice to have classes so easy that you can skip them and still pass. And of course, it's never nice to be the person heading into the library to spend 3 hours per night on homework + another 8 hours for the weekend. All the engineering students had enough electives to realize that your average liberal arts major probably had very little real effort outside of a crunch period studying for finals. Indeed, we'd always laugh about how crowded the main library was 2 weeks before finals, when it had been empty for weeks on end before hand.

"In fact, a male friend was having a ball in a woman-dominated field. "

Sure, he had plenty of off time and plenty of women around him. I'm sure he had a ball.

"Nursing isn't STEM?"

No. It, like most of health care, is a trade. The technology is science based, but the training doesn't require much understanding of science per se. Witness the rapid growth of pseudoscience in the health professions, even in training facilities. It is a serious misjudgement to equate training in trades that use technology with understanding of, or practice of, science based thinking.

This story makes a lot of sense. It's also not the first paper to look at the STEM gender gap from a "pre-college" skills perspective. See also Speer (2017 Labour Economics ); Aucejo & James (2016, working paper).

From Speer: "The gender gaps in test scores, particularly in science and mechanical fields, exist by the mid-teenage years and grow with age."

From Aucejo & James: "Finally, we study the gender gaps in college enrollment and STEM field enrollment, showing that verbal skills and comparative advantage in skills are key determinants of these gaps."

The results from all three studies are consistent.

I'm glad to see the new paper, but I was bothered by the lack of references (including me of course, but they missed lots of other papers -- one page of references for such a huge literature?). Thanks TM for pointing this out.

Happy to point out great research on an important topic. There are probably other important papers in this vein that I don't know about since I'm not deep into this literature.

Wang et al (2013) "Not Lack of Ability but More Choice: Individual and Gender Differences in Choice of Careers in Science, Technology, Engineering, and Mathematics":

It sounds like the argument boils down to the claim that STEM-qualified females have other choices and STEM-qualified males don't -- they're 'stuck' in STEM. But given the relatively high pay in STEM fields, why do women leave in droves (or never enter in the first place)? And do we *really* think that STEM-qualified males would really prefer to switch to Humanities but are prevented from doing so because their language skills aren't up to it? Also, isn't it pretty clear that Biology and Chemistry aren't as 'mathy' as Calculus and Physics (and aren't women already a majority of University biology majors)?

Lastly, shouldn't we all be vastly more worried about this:

"many more females (44%) than males (32%) enter university"

Than how many females choose STEM careers?

STEM careers are mostly shit, I think women recognize that more.


Hey leave us cuckolds alone!

I think this may be exactly the point. You get chemistry and biology to get into the health sciences. Doctor, nurse, pharmacology, the medical technicians, vet. These are good paying jobs, usually with pretty good female oriented workplaces and structures for childcare. STEM fields are typically not.

In other words people are making rational choices.

STEM-qualified males are not stuck. They could switch to other majors also. The paper's claim is that it is sufficient that males on average have a comparative advantage (not an absolute advantage) in STEM fields, coupled with the overall difference in the numbers of men and women entering universities, to explain the current distribution of choice in majors. Given that this is a 2-outputs problem (STEM vs. non-STEM), there are plenty of simple comparative advantage toy models that show this kind of result.

It seems self-evident that (most) women don't feel as highly pressured to find highly paying jobs as men do. My STEM wife could easily make $100K+ working for an oil company but chooses to teach instead. It was left to her non-STEM husband to pay the mortgage.

This is probably the real answer, but maybe not for long, if male workforce participation continues to decline. Most women don't go into the job market under the assumption that they will be the primary wage earner in the household. Also, if you are assuming that your career is going to be sacrificed in favor of your husbands, why bother? Go into a field that is in demand everywhere so you can move with him. Nursing, teaching, doesn't matter where you live, you can find a job. But a scientific, technical career, there are certain centers that you have to be in to progress. If you have to move to follow your husbands job, you can't really stay at the top rated research lab or technical hub for whatever you're doing.

Again, that all might change if trends continue and women start becoming primary wage earners and start developing an expectation that they will be the primary wage earner.

Men are mostly the ones stupid enough to waste time studying engineering and such things. When I was in school women heavily went into the actuarial science program, a lot of men went into CS. Guess whose making WAY more money now (not the men).

CS covers everything from small town web pages to big city high frequency trading. Prospective students probably know where they fit in, and how aggressive they plan to be.

It's kind of like saying "how much do salesmen make?"

Well, what are we talking about, socks or skyscrapers?

The thing is you don't know what you're talking about.

I will not divulge the scale of my success. It would be a shallow way to win the argument. :-)

You comment regularly here so I'm guessing it's zero.

Not using myself as the only datum:

The closing paragraph describes "selling skyscrapers."

That's a fucking stupid article.

You are one weird dude. You start by trashing STEM and CS and end by stubbornly refusing that they all offer an opportunity to hit the long ball.

Geez, the richest man in the world is off and on "CS."

Don't engage him. He posts on every MR Stem related post about how inferior and awful the STEM fields are. It's really weird, I'm not sure what his problem is.

@Natasha: JAMRC might be an engineering major in a shitty job and he's bitter at all the non-engineers that earn more than he does. But more likely he's just trolling you engineers that post here.

" Guess whose making WAY more money now (not the men)."

So, all that talk of disparately low pay for women is complete bunk?

I just think the smart women know better than to go into these shitty fields like engineering or CS. There's far more money and status on offer in other things.

Following Steve Williams (@SteveStuWill) as I do, I find none of this surprising.

People are different, and do things for different reasons. The danger always comes when we start to group people for political advantage.

When populations become more interesting than people.

"a higher concentration of women in biology and chemistry"

How much of the STEM gap disappears when we include medical studies, especially nursing?

"...especially nursing"

And Veterinary Science (80% female students). And Respiratory (60%), Occupational (90%), Physical (65%) therapy. And Pharmacy (over 60% female students).

If you include all those medical related fields, I would guess the answer is 'more than all of it'.

Sure, fine, then we aren't really talking about "science" but people who use sciency things in their jobs. Nurses don't conduct experiments, design things, build things, or perform statistical analysis.

But I'm not faulting your creeping definition of STEM, but rather the original definition of STEM. People have concocted this acronym of professions that are ostensibly contributing to societal advancement. Of course not everyone who gets such degrees makes such contributions. When I was younger it was doctor, lawyer, professor, etc.

There can be gluts of such workers that make these fields poor choices. There is also a lot of competition from foreigners for these jobs.

We need to forget about social engineering of our kids and just let them choose what they want to do without influence. The simplest explanation is that while girls perform well in math and science classes, they just don't like it. What is wrong with that? My two girls can be whatever they want, and I'll be happy if they enjoy their work.

Since "STEM is good," everybody has been working their way into the list. "Urban Forestry?"

"girls perform well in math and science classes": that probably means the exams are too easy.


What's an old cuck like me to say?

Is the entire debate looking at the wrong thing?

Men and women are enormous groups with large overlaps - for instance between high performers and low performers, or between those with particular educational and social advantages. Many disparities doubtless do relate to career-tracks, family roles, time out of the workplace (I know about much of this, being a male who has voluntarily taken many of these options within my family, and seen the impact first hand). These need fixing or ameliorating. The gender side of it meshes with these factors, but is not the driving force behind them - the debate seems rather to be driven by politics, thereby settling on a non-systemic aspect of any disparities, rather than a driver.

Very good points.

The "gender gap" though is at least partially about status.

At the very top of stem fields, the concentration of men is higher (a small sample, though). So the people who accrue disproportionate amounts of status in their field are often men (this is true even in softer fields like sociology, although we need to control for age). Perhaps this is changing; hard to tell or measure.

Positing "greater male variability" might be tempting (at least one notable econ prof got in trouble for it).

Or greater male obsessiveness, competitiveness, etc.

Is law STEM?

Law isn't STEM, it's a career for people who actually have ambitions and ability.

Do you even remember that Bill Gates was pre-law at Harvard before he dropped out?

Law is a much worse profession than engineering or CS. Huge numbers of people rack up huge amounts of debt to go into it and then can't get a job. Those who do get a job are mostly relatively low-paid grunts. Furthermore, lawyers have notoriously low career satisfaction. Sure, some people do very well in it, but the same is true in engineering and CS, probably to a higher degree.

I've heard all this before but I think it's mostly BS. Plus you can be miserable and still be a lawyer. On the other hand engineering/CS basically demands that you be in love with your career, only talk about stuff related to it, spend your free time working on side projects because you love it so much, and then get the boot when you get to be over 35. Nah engineering/CS is MUCH worse.

The judge who was droning on and on about the procedures for deferring my traffic ticket had to be thinking: This is why I went to law school, left my law practice and campaigned for judge?

Law is a much worse profession than engineering or CS. Huge numbers of people rack up huge amounts of debt to go into it and then can’t get a job. Those who do get a job are mostly relatively low-paid grunts.

Median cash compensation for working lawyers is currently $118,000 per year. They're doing all right. A lawyer whose income approximates that of a typical schoolteacher is earning at the 12th percentile.

About 1/3 of those obtaining a law degree find there is no place for them in the profession. Those who do find a place are doing all right.

"Median cash compensation for working lawyers is currently $118,000 per year."

Operative word: "working".

The preferred acronym is SMELT

I thought it was women who were supposed to be more prone to OCD?

OCD does not equal obsessiveness, obviously, just as 4 does not equal 2.

I always tell people that guys like me got into computers because of our limitations, so this study rings true.

By the way, the big innovation that I see in middle and high school is classes dedicated to organization and time management--at least where I live. I suspect that girls are naturally better at those tasks (statistically speaking, please don't fire me), so these classes, while helping everyone, may disproportionately help boys (statistically speaking, please don't fire me).

"I always tell people that guys like me got into computers because of our limitations": yeah, that's my experience of computer people.

"I always tell people that guys like me got into computers because of our limitations,"

Sure, and the NBA is full of multi-millionaires that got into basketball because of their limitations.

I think dearieme has the better point here. Most computer people are pretty low IQ or have severe problems in other respects which makes higher status professions off limits for them.

"Most computer people are pretty low IQ"

LOL, one of the silliest things I've read on the web this week. That's not remotely true. Are you signalling or ignorant?

Median IQ for computer programmers is roughly 110:

110 is a pretty low median for professionals - computer programmers are people who went to school but didn't have the iQ points to become anything better.

What you want from a major, and career, is the opportunity to meet or exceed the average compensation for a given ability.

It sounds like you can do that in law, and from what I have seen in CS as well. The billionaires I knew were pretty smart. The millionaires less so, which from a certain standpoint is a good thing.

JWatts, he's trolling. This is an old schtick for him, look at his handle.

I remembered an even better one. I didn't know Steve Jobs, but at least for a week Steve knew me.

This was a long time ago, well before he was sick. I made a criticism of Job's management on USENET and a magazine picked it up and used it as a lead, with my true name.

Friends of friends knew Steve and asked through the grapevine if Steve had seen it. The answer came back "you'll never work in that town."

Law puts a ceiling on your possible success. No lawyer makes more than low seven figures because law firms can only grow to a certain size before conflicts of interest prevent them from taking on new clients. If you're smart enough to be a BigLaw partner you might be better off going into tech if you are capable

"Are you signalling or ignorant?"


This is a great analogy. They had limited educational options (or chose to limit it after seeing the sports potential), but they landed on something that was very lucrative.

My limitation was that I couldn't tolerate the wishy washiness of the liberal arts, so I ended up in STEM..

I see a couple of comments above about nursing, a field that is still disproportionately female.

We consider nursing a non-STEM occupation, but why? There seems to be a fair amount of science in both the courses taken as part of the degrees, and the use of scientific and technical information in the actual practice of nursing.

Sure, there are non-science, shitty aspects of the job (in the case of nursing, we can mean that literally), but that's true of most jobs.

I would think any degreed nursing would be a STEM degree.

Because using science is not doing science.

STEM is a poorly conceived notion of careers that are valuable for economic development, and I don't think that's objectively true. Nursing is obviously a well paying and valuable career field. I'd consider it professional/technical before I call it "science." There's no need to pigeonhole every job into the STEM model.

That would also apply to most engineering fields.

Using science instead of doing science is what the T and E in STEM stand for.

"There seems to be a fair amount of science in both the courses taken as part of the degrees"

No, there isn't. Look up the degree requirements at your local university if you don't believe me.

"and the use of scientific and technical information in the actual practice of nursing. "

Under the direction of the doctor, who actually understands the science and makes the decisions.

Nursing can be anything from an AA to a PhD. It can be a temperature taker, or a Nurse Practitioner bumping an MD off traditional roles. A wide gamut.

yes, when I hear "nurse", I tend to assume at least a PhD.

You are more likely to encounter the high powered nurses at an HMO, less likely at a traditional MD's office.

In both cases:–agent_problem

"Look up the degree requirements at your local university if you don’t believe me"

Here's the curriculum for a local BSN (bachelors of science in nursing) program. Looks reasonably sciency, although of course there are a lot of clinical practice courses.

If STEM were just science it would be S.

If the E is in there for engineering, I would think degreed medicine analogous. Unless it was originally meant to contrast to medicine? Whatever.

Does selling heroin on the street corner count as medicine and therefore science? if so, there are a lot more people in STEM than we think.

I would imagine that the manufacture and refining of the drug would count as STEM, but the distribution would still be retail/sales/marketing, etc.

Hamburger generally doesn't have pharmaceutical type effects on the human brain.

"Men dominate women in chess primarily because smart men disproportionately go into chess whereas a larger number of smart women enjoy playing other games like dress-up and make-believe."



I thought everyone knew that if a university has a math program and an English program, the English program is easier to get into, but apparently not. And note the use of grades to measure "STEM readiness" while ignoring test scores: test scores show a clear male advantage.

Disregard the second sentence, I glanced over the part where it was a study of Canadian schools.

"And note the use of grades to measure “STEM readiness” "

A couple of years ago Aussie universities were complaining that first year physics students couldn't do the requisite mathematics. It turns out that year twelve physics has morphed into more of a history subject than science. I can guarantee this was not done to advantage boys.

This may explain the "gender gap in the fraction of university entrants who are STEM ready", but not the gender gap in number of STEM entrants. I guess the former is only a small part of the later.

Being "good enough" to choose a certain career path is not the same as "this is what I will be best at"

I'm sure Michael Jordan is a good golfer, and Serena Williams is good at basketball.

There are likely more males whose best academic talent, by their own measurement, is math. This is likely due to biological differences between men and women. That does not mean men are better, just different.

The conventional wisdom is that the gender gap is about women and the forces–discrimination, sexism, parenting, aptitudes, choices; take your pick–that make women less likely to study in STEM fields. Card and Payne are saying that the great bulk of the gap is actually about men and their problem.

Yes, but I'm not sure that you can say that "comparative advantage" isn't about aptitude and choices. Part of it is that women are doing better in general, but it still happens that a large percentage of the women who have good STEM numbers have an even larger advantage in non-STEM fields, so aptitude and choices push them in non STEM directions-- or in the subfields of STEM that require, e.g., more people skills. (So fewer female radiologists, who never talk to patients and just interact with machines.)

Put (too) simply the only men who are good enough to get into university are men who are good at STEM. Women are good enough to get into non-STEM and STEM fields.

Another (too) simple way of putting it is that, on average, men are better in STEM than in non-STEM, whereas women do better in non-STEM than STEM, but women in college do better on average, so the average female matriculant has a STEM aptitude similar to the average male matriculant, but better non-STEM aptitude.

This does still leave open the oft-discussed notion that men have a greater standard deviation in many factors, even if due to cultural and environmental influences (a fact that, by itself, would cause the college population to shift from male dominated to equal to female dominated as we shift from a minority to half to more than half of people going to college.) Regardless, the larger problem is on the lower end for men. (Note also that lower tier universities regularly practice pro-male affirmative action.)

According to what is stated, around 67% of college students in Canada are female (I'm assuming the overall population is close to 50-50). According to IPEDs data, around 57% of American college students are female. This is a substantial difference and makes me wonder about the story being told. Caveat: I have not read or even looked at the paper, so perhaps they address this - if they do, please let me know.

One thing to consider is that although there are a couple of well-regarded schools in Canada, the median Canadian university has far lower standards and quality than the median US school.

Comment is not snarky enough for JAMRC. You are obviously an imposter.

The fact that the authors treat physics as equivalent to biology shows their biases. Moreover, math at the high school level is much less rigorous than college so average grades are meaningless if not corrected for rigor. All university professors know that there are math A students and "A" students. The STEM degrees that women are likely to avoid or drop out of are computer science, pure and applied math, physics, engineering, etc. Even in the social sciences, women are less likely to survive in Ph.D. econ (especially once you subtract women from Asia) than sociology or political science. And even within poli sci, those doing rational choice or some form of modeling are rarer than those who do international studies or political philosophy.

The problem that Larry Summers mentioned about six sigma math/physics types needed to end up at Harvard, Caltech, Princeton as faculty is only mildly weaker for full time careers in those subjects at less rigorous jobs such as "good" tech companies. So now we're only talking two or three sigma not six. But at that level, men outnumber women. Once you factor in the higher earnings potential of equivalent females in law, business, and marketing -- especially if they have some tech skills -- you understand both their low willingness to enter STEM and, once having entered, low willingness to stick it out.

"Moreover, math at the high school level is much less rigorous than college so average grades are meaningless if not corrected for rigor."

What? You might be right about lower-level coursework like "College Algebra" and "Business Calculus," but you're not anywhere near correct about the courses that actually make you a STEM major, such as Differential Equations or Real Analysis.

Oh, god. Cancel that. I misread you. I am sorry.

Maybe my stereotypes are out of date, but isn't pure and applied math a popular field for women in university?

Where do you think all the math teachers come from?

No you're wrong. Most math teachers don't have degrees in math although as I said earlier actuarial science was very popular with women at my school - and THAT is actually a career with low risk / decently high reward - NOT engineering like the morons around here think.

Median Actuarial pay: $83K

Median Software Engineering pay: $81K

That's actually BS.

All facts that disagree with my world view are complete BS.

When I got my math degree 15 years ago my university had three options for math majors. Theoretical, applied, and math for teachers. I was in applied math, and my courses that were unique to that option were heavily male. My roommate at the time was in the theoretical option, and he had a similar experience. In my classes that all three options had to take there were quite a few women, so I always assumed the large female population was due to the women that were learning to be teachers.


Physics and biology require different skill sets, but I don't think it's fair to say that biology is "easier".

Physics requires much deeper math skills, but it's also simpler, conceptually. It's fundamentally reductive.
Biology is complex. You can't use as much mathematics, because it's just too complicated to apply math straightforwardly. That makes it harder in many ways. There are more facts that simply have to be memorized and more independent variables to keep track of. There are emergent orders that are not easy to reduce to simpler physics and chemistry.

I think there is truth to what you say at the higher end of the profession, but in the average bachelor's program biology is much easier than physics.

Definitely. We (pre-med/engineering double majors) used the pre-med bio classes as breaks from the engineering classes. As a grown-up, biology and medicine may be more complex, but the methods used to study them, are pretty simple cook-book statistics.

I don't think the issue is simplicity vs. complexity, though. If you dichotomize by gender, you'll see trends in the averages and then try to separate nature from nurture. It's fun, but let's not get too tribal about it.

Anecdote: In my own lifetime, I've seen the surgical subspecialties go from a gross male locker room field dominated by macho men to being more gender-balanced. This is in the US. A visiting French medical student was shocked by the number of female surgeons. Now, if I were looking back at this 20 years ago, I could come up with all kinds of theories on why women don't like surgery because of knives, violence, schedule/work-life balance, etc.

Anyway, who cares? If you have a kid who's going to study art or get a physics PhD, tell them they need to mary rich or get very lucky with that choice. And, do everything in your power to dissuade them. They may be the next Picasso or Einstein, but most likely, they'll be your barrista.

My physicist friends are either analysts for a big tech firm, or work in banking. The ones in banking do double my salary, and I have an MBA. I expected that since the beginning of our careers but, man, physics is hard!


Note that mathematics becomes more important in all STEM areas as you progress up to grad school. That also creates a shift.

Note that mathematics becomes less and less important in the real world of jobs. You hardly use any of that shit from those shitty programs.

At least, math majors don't shit twice in a single sentence :)

> the gender gap industry

Tyler stated the real reason for the existence of the concept of a "gender gap" in the OP.

IF gender disparities in STEM employment can be explained away with "Well, women could do this stuff but (because they tend to be more all-around capable then men) they're more likely to have other, better options available to them, AND if this then makes it possible for employers to avoid Title IX- style proportionality quotas (as a Safe Harbor against discrimination claims) THEN what's not to like?

I think it's partly because women have more options and so can take riskier majors. Riskier like the risk of not getting a good job when you get out.

My son has a friend, a girl, who is brilliant, who chose to major in art. He would say she should be studying engineering so as assure a high paying job, but that was not important to her and I think that it's more likely for a girl to make that choice, she likes art.

We cast a wide net for women, and the result in my observation after years in a community centered around its hyper-competitive high school is that - though it certainly also happens with boys who are told STEM and STEM alone is where it's at - there is more frequently a disconnect between their qualifications, and actual sustaining interest in their subject, in the girls admitted to college on a STEM track.

I was going to make this comment in previous discussions of this subject...

I'm not sure why people think women are less "systematizing" then men. Ever seen a medieval tapestry?
This is the sort of work mostly done by upper class women at the time, and it basically involves keeping track of hundreds of individuals threads in multiple colors, to produce, one line at a time, an image several feet in length and width. Hundreds of threads to keep track of, and hundreds, maybe thousands of lines which have to be held in place with the desired color exposed on top. And you have to know exactly where in the image to start exposing thread to start "drawing" someone's head, or eyes, or other features. It's almost the same thing as manually programming a printer using raw hex to generate a 16 bit color image. Probably harder.

Same thing with a lot of other skills like lace-making - basically tying hundreds of tiny knots in different patterns so when you pull on the thread it makes a desired shape - like protein folding almost. This is the sort of thing that really intelligent people whose skills are going to waste come up with to pass the time.

One theory for why childhood education in the US has gotten worse is that in the past the only jobs smart women were 'allowed' to do were nursing and teaching. So all the smart women today that would have been teaching kids are now doctors, lawyers, etc. Not sure how valid that is, people like to bitch about how much better everything used to be.

Smart women have always been responsible for much or most of what is good and gracious in the world.

If anything, I would say their overall contribution peaked in the past.

Education has gotten worse?

As I said at the end, not sure that's true, but people are always bitching about how "our schools are failing our children!" as if they were better before...

Maybe better in some ways, not in others - but they were almost certainly relatively less important as the source of what children learn.

"It’s almost the same thing as manually programming a printer using raw hex to generate a 16 bit color image. Probably harder.

It's a lot harder to correct an error you made three days ago. Can't just type over a tapestry.

This aligns with my own priors so I hope it's true. I'm finding myself arguing constantly with my friends (in a very liberal peer group) about the status of women in society - their relative problems with school and work. I keep harping about the crisis for boys, and young men. I think this will be more obvious in about 10-20 years, when there's a full lost generation to contend with. And of course it's going to be a more difficult problem to solve at that time.

>men are having trouble competing everywhere except in STEM.

Yeah, good luck selling that narrative. I don't see how it appeals to the people you need to appeal to.


Also, not a true narrative. I had a choice of majors, including art (which my parents were wise enough to advise me against), but picked one that made money.

Definitely true for this engineering graduate, I couldn't do anything except maths and science so I became an engineer. So can everyone please start feeling sorry for me and stop trying to take my money?

If there is only a "small" difference in the ratio of "STEM-ready" to total H.S. graduates between men and women, then given that far more women enroll in college than men, it is preposterous to claim that the cause of the large excess in the ABSOLUTE number of male STEM graduates to female STEM graduates is due to fewer non-STEM ready males entering college. Or did I misunderstand the paper's abstract?

Me Chinese with small penis

Let's map out an extreme example.

Say you have 100 males and 100 females who graduate HS. 75% of females go to college, 50% of males. That will be 75 females and 50 males in college.

Say All males who go to college go to STEM while females split 50-50.

College Non-Stem: 34 Females
College STEM: 50 males 35 females

Males seem to dominate STEM and know-it-all male Google engineers write fifty page emails pondering if males are just naturally too good when the reality is we have more of a male ghetto going on with high dysfunction than anything else.

Careful Alex. You're getting close to breaking a taboo and calling down the vengeful hordes of internet "activists". You're really not supposed to say that men are having having trouble in our society.

STEM is an easy (and stupid) way of telling kids: "Hey, these are the mooney-making jobs that will help people of your race/gender/socioeconomic status succeed."

STEM: Physics PhD doing her 7th postdoc
STEM: Bio major working as a lab tech
STEM: CS major designing web pages

Non-STEM: Lawyer making dough chasing ambulances.
Non-STEM (apparently): nurse, community doc, dentist, veteranarian, making a comfortable living doing fulfilling work.
Non-STEM: Econ major (humanities, right?) running complex algorithms.

We need to get away from this whole STEM thing. It's stupid and counterproductive. Just like pushing low SE kids to college and its financial burdens vs. going to trde school for HVAC (for example).

Women tend to underestimate their aptitude for STEM and men tend to overestimate. Learning science in a room full of competitive men is very intimidating. The testosterone can be a bit overwhelming. I remember asking a question about NMR and the professor scoffed. Then a meek guy in the corner asked a follow up along the same lines and I knew I was in the right place. Genius is not necessary. The only prerequisites are hard work and diligence.

Ugh. Your prof was a dick. I felt the same way in all my engineering classes (male here), and any overestimations I may have had (so long ago, don't recall) disappeared after the first set of tests :)

"Learning science in a room full of competitive men is very intimidating. The testosterone can be a bit overwhelming."

What kind of classes were these? All my classes were old Indian men droning on and writing equations on a whiteboard while a sea of pasty nerds patiently scribbled in their notebooks.

"Learning science in a room full of competitive men is very intimidating. "

Strange attitude.

If you really want to improve yourself seek out the competition. Hide from it and you'll remain mediocre.

"The testosterone can be a bit overwhelming."

In other words you are already programmed to dislike men.

Um, maybe everyone here should redial a decade or so and ask why Larry Summers (yes, he's a scumbag) was fired from Harvard for merely speaking the truth about gender.

Men exhibit both the extremes of lower and higher IQ. In other words, to be blunt, more men are retards than women, but also, more men are geniuses than women.

So... when you get into the difficult STEMS... the kind that get you hired at like, say, NASA, where genius is an asset, men will ALWAYS outnumber women. But, when you go to an institution for the mentally disabled, you will also notice that most residents are men!

Women cluster around the mean, while men display a far larger range - both high and low.

Yes, men hold almost all positions of power in the world... but men also make up 85% of the least powerful, the homeless. Clustered around the average, are women.

There is enormous scientific evidence for this which I can't be bothered to link to (I already did above).

Maybe everyone should stop screaching about equality so much until they've read Harrison Bergeron - look it up!

Standardized test scores have restricted range, especially at the top. So wider distributions of skills show up more on the bottom, lowering male averages. i.e. males have the same actual skill, but greater variance. After capping that the top, and uncapping at the bottom, men have lower averages.

This is why super high level positions are more male dominated even when average skill levels are similar.

Alex, before making too strong conclusions you should keep in mind that they include nursing in STEM and according to bottom of page 6 nursing accounts for half of all women with a STEM degree in Canada. So if you take out nursing, you will have a a massive gender gap in choosing STEM. This not to say that your argument that men are struggling more and more in the current labor market is wrong. But to make the story complete we would also need to know how much the men who do not study humanities and social science are earning. Some of them, who end up as electricians or plumbers, might earn more than many of the humanities graduates.

Phil Greenspun has been saying this (talented women have better options than being code monkeys) forever. This may be new data but by no means is it a new theory. Plenty of people have been saying this for a long time.

I don't see it. The theory would explain why women dominate non-STEM fields (there's a lot of women in college), but not their near-total absence in some STEM fields.

Well the comments section of this blog seems to attract more of the men who can't compete in STEM either and come here to make lame feminist conspiracy theory excuses.

I can tell you what it was like for me in university in the late '80s and early '90s. I being a major geek who had taught himself assembly code on a vic-20 was naturally in the pre-engineering track. Which was all calculus and physics and chemistry in big auditoriums as that was the only way to accommodate the crush of demand at the beginning of the software boom. And this was genuinely 99% men or more - my classes would be in the hundreds typically with one or two women.

In other words there was an enormous male vs female gap in the tech-industry, vastly bigger than you would find today. So while I would agree that in high school girls on average did as well in math as boys, and overall in high school girls were on average better students than boys, at that time once it came down to choosing a career track, women shied away from tech almost universally. Their ability to do it was not really tested: they just opted out.

I understand there's been a big change since then, but I'd be surprised if opting out of tech and other math and hard science fields isn't still by far the biggest explanation why women are underrepresented. The Damore types who like to think there is some innate lesser ability of women in such areas are lapping up pseudo-science because it gives them a perfect excuse for being jerk-offs.

As for the other side of the coin, just as girls weren't encouraged to get into tech and science, boys were. Boys like me who were naturally good at it were made to feel like it was our calling. Boys from traditional families especially Asian were under heavy parental pressure. There was a brutal filter, less than a third in pre-engineering track got into the engineering school - most of the guys I knew were struggling and one killed himself. I discovered journalism and literature and classes with women in them and never looked back.

As for which are harder, humanities or math, in my experience it wasn't a fair competition because of the massive crush of demand I mentioned. There were far more men wanting a software career than the university was prepared to teach, so they had to run them through a grinder.

One issue to consider is that tests are reportedly routinely adjusted to make boys' advantage in STEM-related fields disappear. I don't know how that would affect the conclusions, though I am sure that without a thorough look under the hood of their "STEM-ready" &c instruments the whole discussion veers dangerously close to being meaningless. Also, don't forget the principle of selection effects in education: if something education-related can possibly be explained by selection effects, it is almost certainly caused by selection effects.

Comments for this post are closed