
The other day on Twitter, Nick B. Steves challenged me to find data supporting or refuting his assertion that Nerds vs. Jocks is a false stereotype, invented around 1975. Of course, we HBDers have a saying–“all stereotypes are true,” even the ones about us–but let’s investigate Nick’s claim and see where it leads us.
(NOTE: If you have relevant data, I’d love to see it.)
Unfortunately, terms like “nerd,” “jock,” and “chad” are not all that well defined. Certainly if we define “jock” as “athletic but not smart” and nerd as “smart but not athletic,” then these are clearly separate categories. But what if there’s a much bigger group of people who are smart and athletic?
Or what if we are defining “nerd” and “jock” too narrowly? Wikipedia defines nerd as, “a person seen as overly intellectual, obsessive, or lacking social skills.” I recall a study–which I cannot find right now–which found that nerds had, overall, lower-than-average IQs, but that study included people who were obsessive about things like comic books, not just people who majored in STEM. Similarly, should we define “jock” only as people who are good at sports, or do passionate sports fans count?
For the sake of this post, I will define “nerd” as “people with high math/science abilities” and “jock” as “people with high athletic abilities,” leaving the matter of social skills undefined. (People who merely like video games or watch sports, therefore, do not count.)
Nick is correct on one count: according to Wikipedia, although the word “nerd” has been around since 1951, it was popularized during the 70s by the sitcom Happy Days. However, Wikipedia also notes that:
An alternate spelling,[10] as nurd or gnurd, also began to appear in the mid-1960s or early 1970s.[11] Author Philip K. Dick claimed to have coined the nurd spelling in 1973, but its first recorded use appeared in a 1965 student publication at Rensselaer Polytechnic Institute.[12][13] Oral tradition there holds that the word is derived from knurd (drunk spelled backward), which was used to describe people who studied rather than partied. The term gnurd (spelled with the “g”) was in use at the Massachusetts Institute of Technology by 1965.[14] The term nurd was also in use at the Massachusetts Institute of Technology as early as 1971 but was used in the context for the proper name of a fictional character in a satirical “news” article.[15]
suggesting that the word was already common among nerds themselves before it was picked up by TV.
But we can trace the nerd-jock dichotomy back before the terms were coined: back in 1921, Lewis Terman, a researcher at Stanford University, began a long-term study of exceptionally high-IQ children, the Genetic Studies of Genius aka the Terman Study of the Gifted:
Terman’s goal was to disprove the then-current belief that gifted children were sickly, socially inept, and not well-rounded.
This belief was especially popular in a little nation known as Germany, where it inspired people to take schoolchildren on long hikes in the woods to keep them fit and the mass-extermination of Jews, who were believed to be muddying the German genepool with their weak, sickly, high-IQ genes (and nefariously trying to marry strong, healthy German in order to replenish their own defective stock.) It didn’t help that German Jews were both high-IQ and beset by a number of illnesses (probably related to high rates of consanguinity,) but then again, the Gypsies are beset by even more debilitating illnesses, but no one blames this on all of the fresh air and exercise afforded by their highly mobile lifestyles.
(Just to be thorough, though, the Nazis also exterminated the Gypsies and Hans Asperger’s subjects, despite Asperger’s insistence that they were very clever children who could probably be of great use to the German war effort via code breaking and the like.)
The results of Terman’s study are strongly in Nick’s favor. According to Psychology Today’s account:
His final group of “Termites” averaged a whopping IQ of 151. Following-up his group 35-years later, his gifted group at mid-life definitely seemed to conform to his expectations. They were taller, healthier, physically better developed, and socially adept (dispelling the myth at the time of high-IQ awkward nerds).
According to Wikipedia:
…the first volume of the study reported data on the children’s family,[17] educational progress,[18] special abilities,[19] interests,[20] play,[21] and personality.[22] He also examined the children’s racial and ethnic heritage.[23] Terman was a proponent of eugenics, although not as radical as many of his contemporary social Darwinists, and believed that intelligence testing could be used as a positive tool to shape society.[3]
Based on data collected in 1921–22, Terman concluded that gifted children suffered no more health problems than normal for their age, save a little more myopia than average. He also found that the children were usually social, were well-adjusted, did better in school, and were even taller than average.[24] A follow-up performed in 1923–1924 found that the children had maintained their high IQs and were still above average overall as a group.
Of course, we can go back even further than Terman–in the early 1800s, allergies like hay fever were associated with the nobility, who of course did not do much vigorous work in the fields.
My impression, based on studies I’ve seen previously, is that athleticism and IQ are positively correlated. That is, smarter people tend to be more athletic, and more athletic people tend to be smarter. There’s a very obvious reason for this: our brains are part of our bodies, people with healthier bodies therefore also have healthier brains, and healthier brains tend to work better.
At the very bottom of the IQ distribution, mentally retarded people tend to also be clumsy, flacid, or lacking good muscle tone. The same genes (or environmental conditions) that make children have terrible health/developmental problems often also affect their brain growth, and conditions that affect their brains also affect their bodies. As we progress from low to average to above-average IQ, we encounter increasingly healthy people.
In most smart people, high-IQ doesn’t seem to be a random fluke, a genetic error, nor fitness reducing: in a genetic study of children with exceptionally high IQs, researchers failed to find many genes that specifically endowed the children with genius, but found instead a fortuitous absence of deleterious genes that knock a few points off the rest of us. The same genes that have a negative effect on the nerves and proteins in your brain probably also have a deleterious effect on the nerves and proteins throughout the rest of your body.
And indeed, there are many studies which show a correlation between intelligence and strength (eg, Longitudinal and Cross-Sectional Assessments of Age Changes in Physical Strength as Related to Sex, Social Class, and Mental Ability) or intelligence and overall health/not dying (eg, Intelligence in young adulthood and cause-specific mortality in the Danish Conscription Database (pdf) and The effects of occupation-based social position on mortality in a large American cohort.)
On the other hand, the evolutionary standard for “fitness” isn’t strength or longevity, but reproduction, and on this scale the high-IQ don’t seem to do as well:
Smart teens don’t have sex (or kiss much either): (h/t Gene Expresion)
Controlling for age, physical maturity, and mother’s education, a significant curvilinear relationship between intelligence and coital status was demonstrated; adolescents at the upper and lower ends of the intelligence distribution were less likely to have sex. Higher intelligence was also associated with postponement of the initiation of the full range of partnered sexual activities. … Higher intelligence operates as a protective factor against early sexual activity during adolescence, and lower intelligence, to a point, is a risk factor.

Here we see the issue plainly: males at 120 and 130 IQ are less likely to get laid than clinically retarded men in 70s and 60s. The right side of the graph are “nerds”, the left side, “jocks.” Of course, the high-IQ females are even less likely to get laid than the high-IQ males, but males tend to judge themselves against other men, not women, when it comes to dating success. Since the low-IQ females are much less likely to get laid than the low-IQ males, this implies that most of these “popular” guys are dating girls who are smarter than themselves–a fact not lost on the nerds, who would also like to date those girls.
In 2001, the MIT/Wellesley magazine Counterpart (Wellesley is MIT’s “sister school” and the two campuses allow cross-enrollment in each other’s courses) published a sex survey that provides a more detailed picture of nerd virginity:
I’m guessing that computer scientists invented polyamory, and neuroscientists are the chads of STEM. The results are otherwise pretty predictable.
Unfortunately, Counterpoint appears to be defunct due to lack of funding/interest and I can no longer find the original survey, but here is Jason Malloy’s summary from Gene Expression:
By the age of 19, 80% of US males and 75% of women have lost their virginity, and 87% of college students have had sex. But this number appears to be much lower at elite (i.e. more intelligent) colleges. According to the article, only 56% of Princeton undergraduates have had intercourse. At Harvard 59% of the undergraduates are non-virgins, and at MIT, only a slight majority, 51%, have had intercourse. Further, only 65% of MIT graduate students have had sex.
The student surveys at MIT and Wellesley also compared virginity by academic major. The chart for Wellesley displayed below shows that 0% of studio art majors were virgins, but 72% of biology majors were virgins, and 83% of biochem and math majors were virgins! Similarly, at MIT 20% of ‘humanities’ majors were virgins, but 73% of biology majors. (Apparently those most likely to read Darwin are also the least Darwinian!)
College Confidential has one paragraph from the study:
How Rolling Stone-ish are the few lucky souls who are doing the horizontal mambo? Well, not very. Considering all the non-virgins on campus, 41% of Wellesley and 32% of MIT students have only had one partner (figure 5). It seems that many Wellesley and MIT students are comfortingly monogamous. Only 9% of those who have gotten it on at MIT have been with more than 10 people and the number is 7% at Wellesley.
Someone needs to find the original study and PUT IT BACK ON THE INTERNET.
But this lack of early sexual success seems to translate into long-term marital happiness, once nerds find “the one.”Lex Fridman’s Divorce Rates by Profession offers a thorough list. The average divorce rate was 16.35%, with a high of 43% (Dancers) and a low of 0% (“Media and communication equipment workers.”)
I’m not sure exactly what all of these jobs are nor exactly which ones should count as STEM (veterinarian? anthropologists?) nor do I know how many people are employed in each field, but I count 49 STEM professions that have lower than average divorce rates (including computer scientists, economists, mathematical science, statisticians, engineers, biologists, chemists, aerospace engineers, astronomers and physicists, physicians, and nuclear engineers,) and only 23 with higher than average divorce rates (including electricians, water treatment plant operators, radio and telecommunication installers, broadcast engineers, and similar professions.) The purer sciences obviously had lower rates than the more practical applied tech fields.
The big outliers were mathematicians (19.15%), psychologists (19.26%), and sociologists (23.53%), though I’m not sure they count (if so, there were only 22 professions with higher than average divorce rates.)
I’m not sure which professions count as “jock” or “chad,” but athletes had lower than average rates of divorce (14.05%) as did firefighters, soldiers, and farmers. Financial examiners, hunters, and dancers, (presumably an athletic female occupation) however, had very high rates of divorce.
Medical Daily has an article on Who is Most Likely to Cheat? The Top 9 Jobs Unfaithful People Have (according to survey):
According to the survey recently taken by the “infidelity dating website,” Victoria Milan, individuals working in the finance field, such as brokers, bankers, and analysts, are more likely to cheat than those in any other profession. However, following those in finance comes those in the aviation field, healthcare, business, and sports.
With the exception of healthcare and maybe aviation, these are pretty typical Chad occupations, not STEM.
The Mirror has a similar list of jobs where people are most and least likely to be married. Most likely: Dentist, Chief Executive, Sales Engineer, Physician, Podiatrist, Optometrist, Farm product buyer, Precision grinder, Religious worker, Tool and die maker.
Least likely: Paper-hanger, Drilling machine operator, Knitter textile operator, Forge operator, Mail handler, Science technician, Practical nurse, Social welfare clerk, Winding machine operative, Postal clerk.
I struggled to find data on male fertility by profession/education/IQ, but there’s plenty on female fertility, eg the deceptively titled High-Fliers have more Babies:
…American women without any form of high-school diploma have a fertility rate of 2.24 children. Among women with a high-school diploma the fertility rate falls to 2.09 and for women with some form of college education it drops to 1.78.
However, among women with college degrees, the economists found the fertility rate rises to 1.88 and among women with advanced degrees to 1.96. In 1980 women who had studied for 16 years or more had a fertility rate of just 1.2.
As the economists prosaically explain: “The relationship between fertility and women’s education in the US has recently become U-shaped.”
Here is another article about the difference in fertility rates between high and low-IQ women.
But female fertility and male fertility may not be the same–I recall data elsewhere indicating that high-IQ men have more children than low IQ men, which implies those men are having their children with low-IQ women. (For example, while Bill and Hillary seem about matched on IQ, and have only one child, Melania Trump does not seem as intelligent as Trump, who has five children.)
Amusingly, I did find data on fertility rate by father’s profession for 1920, in the Birth Statistics for the Birth Registration Area of the US:
Of the 1,508,874 children born in 1920 in the birth registration area of the United states, occupations of fathers are stated for … 96.9%… The average number of children ever born to the present wives of these occupied fathers is 3.3 and the average number of children living 2.9.
The average number of children ever born ranges from 4.6 for foremen, overseers, and inspectors engaged in the extraction of minerals to 1.8 for soldiers, sailors, and marines. Both of these extreme averages are easily explained, for soldier, sailors and marines are usually young, while such foremen, overseers, and inspectors are usually in middle life. For many occupations, however, the ages of the fathers are presumably about the same and differences shown indicate real differences in the size of families. For example, the low figure for dentists, (2), architects, (2.1), and artists, sculptors, and teachers of art (2.2) are in striking contrast with the figure for mine operatives (4.3), quarry operatives (4.1) bootblacks, and brick and stone masons (each 3.9). …
As a rule the occupations credited with the highest number of children born are also credited with the highest number of children living, the highest number of children living appearing for foremen, overseers, and inspectors engaged in the extraction of minerals (3.9) and for steam and street railroad foremen and overseer (3.8), while if we exclude groups plainly affected by the age of fathers, the highest number of children living appear for mine and quarry operatives (each 3.6).
Obviously the job market was very different in 1920–no one was majoring in computer science. Perhaps some of those folks who became mine and quarry operatives back then would become engineers today–or perhaps not. Here are the average numbers of surviving children for the most obviously STEM professions (remember average for 1920 was 2.9):
Electricians 2.1, Electrotypers 2.2, telegraph operator 2.2, actors 1.9, chemists 1.8, Inventors 1.8, photographers and physicians 2.1, technical engineers 1.9, veterinarians 2.2.
I don’t know what paper hangers do, but the Mirror said they were among the least likely to be married, and in 1920, they had an average of 3.1 children–above average.
What about athletes? How smart are they?
“Athletes Show Huge Gaps on SAT Scores” is not a promising title for the “nerds are athletic” crew.
The Journal-Constitution studied 54 public universities, “including the members of the six major Bowl Championship Series conferences and other schools whose teams finished the 2007-08 season ranked among the football or men’s basketball top 25.”…
- Football players average 220 points lower on the SAT than their classmates. Men’s basketball was 227 points lower.
- University of Florida won the prize for biggest gap between football players and the student body, with players scoring 346 points lower than their peers.
- Georgia Tech had the nation’s best average SAT score for football players, 1028 of a possible 1600, and best average high school GPA, 3.39 of a possible 4.0. But because its student body is apparently very smart, Tech’s football players still scored 315 SAT points lower than their classmates.
- UCLA, which has won more NCAA championships in all sports than any other school, had the biggest gap between the average SAT scores of athletes in all sports and its overall student body, at 247 points.
From the original article, which no longer seems to be up on the Journal-Constitution website:
All 53 schools for which football SAT scores were available had at least an 88-point gap between team members’ average score and the average for the student body. …
Football players performed 115 points worse on the SAT than male athletes in other sports.
The differences between athletes’ and non-athletes’ SAT scores were less than half as big for women (73 points) as for men (170).
Many schools routinely used a special admissions process to admit athletes who did not meet the normal entrance requirements. … At Georgia, for instance, 73.5 percent of athletes were special admits compared with 6.6 percent of the student body as a whole.
On the other hand, as Discover Magazine discusses in “The Brain: Why Athletes are Geniuses,” athletic tasks–like catching a fly ball or slapping a hockey puck–require exceptionally fast and accurate brain signals to trigger the correct muscle movements.
Ryan Stegal studied the GPAs of highschool student athletes vs. non-athletes and found that the athletes had higher average GPAs than the non-athletes, but he also notes that the athletes were required to meet certain minimum GPA requirements in order to play.
But within athletics, it looks like the smarter athletes perform better than dumber ones, which is why the NFL uses the Wonderlic Intelligence Test:
NFL draft picks have taken the Wonderlic test for years because team owners need to know if their million dollar player has the cognitive skills to be a star on the field.
What does the NFL know about hiring that most companies don’t? They know that regardless of the position, proof of intelligence plays a profound role in the success of every individual on the team. It’s not enough to have physical ability. The coaches understand that players have to be smart and think quickly to succeed on the field, and the closer they are to the ball the smarter they need to be. That’s why, every potential draft pick takes the Wonderlic Personnel Test at the combine to prove he does–or doesn’t—have the brains to win the game. …
The first use of the WPT in the NFL was by Tom Landry of the Dallas Cowboys in the early 70s, who took a scientific approach to finding players. He believed players who could use their minds where it counted had a strategic advantage over the other teams. He was right, and the test has been used at the combine ever since.
For the NFL, years of testing shows that the higher a player scores on the Wonderlic, the more likely he is to be in the starting lineup—for any position. “There is no other reasonable explanation for the difference in test scores between starting players and those that sit on the bench,” Callans says. “Intelligence plays a role in how well they play the game.”
Let’s look at Exercising Intelligence: How Research Shows a Link Between Physical Activity and Smarts:
A large study conducted at the Sahlgrenska Academy and Sahlgrenska University Hospital in Gothenburg, Sweden, reveals that young adults who regularly exercise have higher IQ scores and are more likely to go on to university.
The study was published in the Proceedings of the National Academy of Sciences (PNAS), and involved more than 1.2 million Swedish men. The men were performing military service and were born between the years 1950 and 1976. Both their physical and IQ test scores were reviewed by the research team. …
The researchers also looked at data for twins and determined that primarily environmental factors are responsible for the association between IQ and fitness, and not genetic makeup. “We have also shown that those youngsters who improve their physical fitness between the ages of 15 and 18 increase their cognitive performance.”…
I have seen similar studies before, some involving mice and some, IIRC, the elderly. It appears that exercise is probably good for you.
I have a few more studies I’d like to mention quickly before moving on to discussion.
Here’s Grip Strength and Physical Demand of Previous Occupation in a Well-Functioning Cohort of Chinese Older Adults (h/t prius_1995) found that participants who had previously worked in construction had greater grip strength than former office workers.
Age and Gender-Specific Normative Data of Grip and Pinch Strength in a Healthy Adult Swiss Population (h/t prius_1995).
If the nerds are in the sedentary cohort, then they be just as athletic if not more athletic than all of the other cohorts except the heavy work.
However, in Revised normative values for grip strength with the Jamar dynamometer, the authors found no effect of profession on grip strength.
And Isometric muscle strength and anthropometric characteristics of a Chinese sample (h/t prius_1995).
And Pumpkin Person has an interesting post about brain size vs. body size.
Discussion: Are nerds real?
Overall, it looks like smarter people are more athletic, more athletic people are smarter, smarter athletes are better athletes, and exercise may make you smarter. For most people, the nerd/jock dichotomy is wrong.
However, there is very little overlap at the very highest end of the athletic and intelligence curves–most college (and thus professional) athletes are less intelligent than the average college student, and most college students are less athletic than the average college (and professional) athlete.
Additionally, while people with STEM degrees make excellent spouses (except for mathematicians, apparently,) their reproductive success is below average: they have sex later than their peers and, as far as the data I’ve been able to find shows, have fewer children.

Even if there is a large overlap between smart people and athletes, they are still separate categories selecting for different things: a cripple can still be a genius, but can’t play football; a dumb person can play sports, but not do well at math. Stephen Hawking can barely move, but he’s still one of the smartest people in the world. So the set of all smart people will always include more “stereotypical nerds” than the set of all athletes, and the set of all athletes will always include more “stereotypical jocks” than the set of all smart people.
In my experience, nerds aren’t socially awkward (aside from their shyness around women.) The myth that they are stems from the fact that they have different interests and communicate in a different way than non-nerds. Let nerds talk to other nerds, and they are perfectly normal, communicative, socially functional people. Put them in a room full of non-nerds, and suddenly the nerds are “awkward.”
Unfortunately, the vast majority of people are not nerds, so many nerds have to spend the majority of their time in the company of lots of people who are very different than themselves. By contrast, very few people of normal IQ and interests ever have to spend time surrounded by the very small population of nerds. If you did put them in a room full of nerds, however, you’d find that suddenly they don’t fit in. The perception that nerds are socially awkward is therefore just normie bias.
Why did the nerd/jock dichotomy become so popular in the 70s? Probably in part because science and technology were really taking off as fields normal people could aspire to major in, man had just landed on the moon and the Intel 4004 was released in 1971. Very few people went to college or were employed in sciences back in 1920; by 1970, colleges were everywhere and science was booming.
And at the same time, colleges and highschools were ramping up their athletics programs. I’d wager that the average school in the 1800s had neither PE nor athletics of any sort. To find those, you’d probably have to attend private academies like Andover or Exeter. By the 70s, though, schools were taking their athletics programs–even athletic recruitment–seriously.
How strong you felt the dichotomy probably depends on the nature of your school. I have attended schools where all of the students were fairly smart and there was no anti-nerd sentiment, and I have attended schools where my classmates were fiercely anti-nerd and made sure I knew it.
But the dichotomy predates the terminology. Take Superman, first 1938. His disguise is a pair of glasses, because no one can believe that the bookish, mild-mannered, Clark Kent is actually the super-strong Superman. Batman is based on the character of El Zorro, created in 1919. Zorro is an effete, weak, foppish nobleman by day and a dashing, sword-fighting hero of the poor by night. Of course these characters are both smart and athletic, but their disguises only work because others do not expect them to be. As fantasies, the characters are powerful because they provide a vehicle for our own desires: for our everyday normal failings to be just a cover for how secretly amazing we are.
But for the most part, most smart people are perfectly fit, healthy, and coordinated–even the ones who like math.