The Greater Male Variability Hypothesis
one way that sex differences matter
Comments for paid subs are now open and will be open until I go to bed on Monday, November 11 (through the long weekend). My comments are now closed during the workweek. It may not always be this way, but for now it’s helpful as I acclimate to a challenging new job. Thanks for understanding.
In this post, I’m going to explain why there are more men than women who win Nobel Prizes, Fields Medals (the most prestigious award in mathematics), and other such accolades — and why sexist discrimination is not the deciding factor.
Baseline Assumptions: Sexism
To start, I must ask you to supply your own “not alls”. I am not claiming to describe every situation for every human being on earth perfectly. I am making broad, sweeping points, not commenting on the details of anyone’s life except my own. Please try to read this with charity and in good faith, knowing that I’m fully aware that I’m not describing every situation or person perfectly.
Next, I’m going to lay out my priors and some of my experiences. Controversial topics tend to cause people to read what they want to read and see what they want to see, but I truly believe that understanding my topic here would help enough people that it’s worth the effort.
Sexism exists. It exists explicitly and in both directions. Some explicit examples:
Not being telepathic, I cannot be sure, but I suspect that a sexist assumption was the motivation when a client who had just heard me give a careful, detailed, rigorous, clear explanation of the math behind a recommendation asked a male coworker (who was literally just our PowerPoint guy, with no mathematical expertise, and in fact I doubt he could balance his own checkbook) if he had checked my work and thought I was right.
One of my guy friends, who is a devoted and loving father, has repeatedly expressed his frustration to me that there aren’t diaper changing facilities in most public bathrooms for men—and this is true despite the fact that he, like me, lives in a deeply blue area. The built-in assumption that men cannot or will not care for their children in a hands-on way is an example of sexism against men. So are the looks he sometimes gets when he corrects people who gush at how wonderful he is: “I’m not babysitting my children. I am raising them, and it is not an achievement to be taking care of them. Your abominably low expectations are noted and rejected.”
Some implicit examples:
Men who are of average or below-average attractiveness can generally have a public presence (i.e., a YouTube channel) without constant insults to their appearance filling the comments and chat. The same is not true for women.
Women who are of average or below-average levels of diligence, work ethic, and financial/career success can generally go through life without this being regarded as evidence of their laziness or other character defects. The same is not true for men.
In other words: I am fully and personally aware that sexism exists.
Baseline Assumptions: IQ
IQ measures something real, but every measure we have for it is highly imperfect. And it does not measure many types of intelligence that are very important. A few examples:
I have almost no sense of direction at all, whereas a friend of mine with a below-average IQ (who hasn’t read a whole book since his last assigned book report in fourth grade) has an excellent sense of direction and inborn talent for spatial navigation. In daily life, especially emergencies, you would almost certainly be far better off to have him with you than me.
A friend’s grandmother, long since deceased, was abandoned by her alcoholic husband in the 1950s. She managed to keep five kids in school, fed, and housed through sheer resilience and hard work. She made a flexible, work-from-home income stream for her family when she taught herself how to make gorgeous quilts — quilts that are truly works of art, some of them museum-quality, with intricate geometric designs that she did by eyeball, not even making notes. Her IQ test score would be quite low, as she only got a first-grade education, quickly forgetting most aspects of reading once her education ended.
I have had my IQ formally tested five times now, and there is a 49 point spread between my highest and lowest scores. The more severe my PTSD and other mental health issues are on a given day, the worse I do on such tests. That means that, as a measure, IQ testing is highly dependent on a person testing when they are in a calm, decent-mood, nervous-system-not-too-activated state. As both the circumstances that cause stress and the physical symptoms of stress are not things that anyone has full control over, this means it is a useful-but-limited measure.
In other words: I am fully aware that IQ is imperfect and measures only one type of intelligence, and that it’s arguably not the most important kind for most human endeavors. I am also aware that sociopaths tend to have above-average IQs, which is just one of the many reasons that IQ means nothing about a person’s morality, character, virtue, or capacity for goodness.
People Are Individuals
I have written about how regarding people first and foremost as members of their groups is a collectivist notion, and the fact that even anti-Woke people tend to do it makes me suspect that the Left has probably already won, in the long run.
Every human being deserves the dignity of being regarded as an individual and judged on the content of his/her character, not his/her identity metrics.
Nothing I say here is meant to imply otherwise.
The Greater Male Variability Hypothesis
The greater male variability hypothesis suggests that there tends to be more variability in male traits, abilities, or behaviors than in female ones, and that this shows up most frequently at the extreme ends of a distribution. This hypothesis has been studied across various fields, including cognitive abilities and physical traits. It posits that while the average abilities of males and females might be similar, males are more likely to be represented at both the high and low extremes.
As with everything related to sex differences, this hypothesis is controversial. But I believe it’s accurate, and that it explains why men tend to dominate the far end — the “win awards and get famous” end — of human pursuits that require unusually high intelligence, like the STEM fields.
To start, we will have a brief lesson in the statistics around IQ.
How the Math of IQ Works
IQ tests are consistently re-normed so that the average stays around 100. As populations change, what 100 means might also change—what an average IQ person’s cognitive abilities are in 2024 might be different from what they were in 1924 or 1974 — but the tests are consistently re-normed to keep the statistics working the way I’m about to describe.
The mean stays at about 100. The standard deviation stays at about 15. And the overall population fits the Bell Curve (aka Normal Distribution) rule. That rule tells us that 68% of the population is within 1 standard deviation of the mean (from one below to one above), 95% of the population is within 2 standard deviations of the mean (from two below to two above) and 99.7% of the population is within 3 standard deviations of the mean (from three below to three above). The other 0.3% are split between the very, very low (more than three standard deviations below the mean) and the very, very, very high (more than three standard deviations above the mean).
This graphic shows the way it works. The character next to the -3, -2, -1, +1, +2, +3 is the Greek letter sigma, which is here just a shorthand for “standard deviation,” and the character in the middle is the Greek letter mu, which is here just a shorthand for “average”.
This graphic shows IQ scores and how they work for this rule:
68% of the population has an IQ between 85 and 115, about evenly split (so about 34% between 85 and 100 and about 34% between 100 and 115).
95% of the population has an IQ between 70 and 130, about evenly split, so you take the 95 and subtract 68, leaving 27, and divide that in half. About 13.5% of people have an IQ between 70 and 85 and about 13.5% of people have an IQ between 115 and 130.
99.7% of the population has an IQ between 55 and 145, and again you subtract what we’ve already covered and split the rest between the two sides. About 2.35% of the population has an IQ between 55 and 70 and about 2.35% of the population has an IQ between 130 and 145.
The farthest tails of the distribution — the unfortunate people who are profoundly retarded, with an IQ below 55 and the fortunate people who are profoundly gifted, with an IQ above 145 — are again about evenly split. About 0.15% of the population fits into each.
What The Greater Male Variability Hypothesis Means for Male IQ
Women and men have an equal mean IQ — both sexes have an average of around 100. But there are more men than women in both tails of the distribution.
If you could gather all humans with an IQ below 55, most of them would be men.
If you could gather all humans with an IQ above 145, most of them would be men.
Men win most of the Nobel Prizes and Fields medals.
But if we gave awards for being unusually stupid, men would win most of those, too.
No, this doesn’t go all the way to explaining disparities in things like Nobel Prizes and Fields Medals. The biological reality of procreation means that women must devote nine months plus recovery time to each offspring, and that year (or longer) is on a largely-unpredictable spectrum from pretty easy (an uncomplicated pregnancy during which the mother continues normal activities as long as possible) to devastatingly difficult (bedrest requiring full cessation of her normal life followed by the major surgery of a Ceasarean delivery). These realities have consequences for careers. And yes, sexism is still a thing in some areas.
But it does most of the work to explain this.
And this is wonderful news!
Evolution made us different, and we can understand what some of those consequences look like on the level of groups and still treat each other as individuals.
What This Looks Like In Normal Life
Though the means are about equal, the greater male populations in the tails means that the populations might look roughly like this:
In a population with exactly equal numbers of men and women, the average intelligence of both sexes would be about the same, but there would be slightly fewer men in the average, above average, and below average categories, offset by the greater number of men in the profoundly retarded and profoundly gifted categories.
Another Example of How Sex Differences Play Out
This blog post, by a non-Woke PhD in psychology, discusses another way that sex differences tend to play out, stemming from these tendencies.
About 70% of people who have strong math skills and strong verbal skills are female.
Males are more than twice as likely as females to have the combination of weak verbal skills and strong math skills.
What this means is that there are many more women than men among the population subset who have the skills to choose either a math-heavy career or a verbal-skills-heavy career. Thus, there are many more women than men who have the capacity to choose between a fairly wide range of occupations.
Women are more broadly skilled than men, on average (ha ha, broad).
Men who have the breadth of skill to both get a PhD in math and write good books and also work with people, traveling to teach and make presentations—James Lindsay is an example of this kind of man—are much rarer than women who have this breadth of skill.
I am a data scientist, mostly working with math and coding.
But I write pretty well, I draw pretty well, I’m a pretty good teacher, I’m very good with kids, and I’m pretty good at things like logistics and organizing.
I suspect that, if I got the proper licenses, I could succeed as a classroom teacher in math or studio art. That’s a pretty broad range of skills. Many other people have my same breadth of skills, of course — but if you could isolate that group and take a census, you would find that most of them are female.
What these individual differences mean on the group level is that more women than men have choices, meaning they can look at many possible options and choose which one they think they will enjoy the most, and those choices tend to be reflected in the career fields.
This graphic, from the linked blog post, is a good demonstration:
Women profoundly dominate graduate school in public administration and services, health sciences, education, and the social and behavioral sciences.
That doesn’t mean those fields discriminate against men. Looking at “health sciences” as one example — the dominance of women in this field means that women who have the IQ to succeed in advanced sciences tend to choose clinical fields (where they will work directly with people) rather than more things-focused fields like engineering, computer science, and mathematics.
Another example of how this dynamic plays out is veterinary medicine. Women are almost 80% of veterinary students in Canada and the United States, but there is a crisis-level shortage of veterinarians who will treat large animals on farms.
Veterinary school is very difficult. It’s basically med school for several different species at once. But the women who are smart enough to go into that field tend to not want to be large-animal vets. Why?
Because nearly all newly-minted veterinarians are women, and it’s more appealing to women on average to live in areas with more people. There, they practice veterinary medicine in a manner that develops community relationships with families through treating their companion animals. This is preferable to more women on average than living in a rural area and spending time gloved up to their elbows, sticking their hands up cow asses.
It isn’t because programs that focus on large animal treatment discriminate against women.
Conclusion
One of the worst consequences of identity politics is the tendency it’s created in everyone — including those of us on the center-right who are fully anti-Woke and aware of the dangers of collectivist thinking — to analyze everything on the group level and to see people, including ourselves, as avatars of our group identity metrics.
This is silly, stupid, counter-productive, harmful, and ultimately only makes Leftist efforts more likely to succeed.
I hope that by explaining this in some detail, it can help people who believe in meritocracy but also wonder why some of the group-level statistics shake out the way they do to just appreciate the beautiful variety of humans.
And let go of at least some of those fears.
At least, that’s how it worked for me. I’m very grateful for that, and wanted to pass it on.
Male/Female differences is an extremely touchy subject and its difficult to have a good-faith discussion about it. I've been there - once discussed this hypothesis with my therapist (a self id's feminist), we get along really well but it was like discovering you'd accidentally strolled into a minefield.
Its a subject that needs really good intellectual discipline and reminds me of the mis-understandings about evolution that many fall into. I'm referring to the non-teleology of that process - the outcome is the "result of whatever survived" rather being directed to some gain. In other words these differences between IQ distros may be the outcome of some other selection process, consider how a prehistoric caveman-clan would be better off with one genius and 99 blokes to do the heavy lifting than a 50/50 split.
IQ tests were designed for the French Education to double screen those who were not educable for skilled roles and those suitable for advance study. They are not reliable at either extreme. For the population, this does not matter.
For the bright individual (say >2.5 SD) g intelligence does not work. Most men at the extreme end of intellectual giftedness are extremely incompetent in other areas. One of the original uses of monasteries and universities was to provide such with a sheltered workshop and keep them outodf the gene pool.
The more noxious, midwitted argument is that math is wrong because feelings. Incorrect. The math is a modelthant, as you note, pragmatically works.