The New York Times presents us with a chart today showing that rich kids are far more likely to get high SAT scores than poor kids:
What's the cause of this huge inequality?
Children from rich and poor families receive vastly different educations, in and out of school, driven by differences in the amount of money and time their parents are able to invest.... private schools, summers traveling the world.... test prep.... tutors.... neighborhoods of concentrated poverty or affluence.... time and connections.... volunteering in classrooms, lobbying on behalf of the school and raising money through school foundations.... friendships.... segregated neighborhoods.... what children do in the evenings and on summer breaks, their parents’ vocabularies, and the level of stress in their home lives.... private extracurriculars, counseling, tutoring, coaching, therapy, health management.... high-quality preschools.... intensive parenting.... bedtime reading, museum visits and science summer camps.
There's truth to all this. But out of 2,000 words, there is only one passing suggestion that there's any other cause of this disparity in SAT scores:
Although the heritability of cognitive ability appears to play some role on an individual level....
"Some" role. Research suggests that by first grade there's an IQ difference of 11 points between children of affluent and poor families. That's a lot! This might very well be partially explained by home life and neighborhoods, but it's long before private schools, world travel, and test prep classes have any impact.
There's nothing controversial about how this happens, either. Smart people tend to make lots of money; marry other smart people ("associative mating"); and then produce smart babies who go on to get high SAT scores. Some of this is indeed due to environment, but most of it is up to heritability and genes. We're all just afraid to say so for fear of accidentally brushing up against forbidden race-IQ topics.
In any case, the most interesting aspect of the chart isn't the SAT differences between rich and poor. It's the difference between the top 1% and the top 0.1%. No one thinks there's any cognitive difference between these two groups, but then again, there's probably not much difference in preschools and test prep either. These are both very wealthy cohorts. So what causes the difference? Further empirical research on this score might be illuminating.
If one believes this nonsense, then all of Secretariat foals would have won all the Kentucky Derbies. The reality is that the horse breeding industry has become an unmitigated disaster with frail horses and poor practices (Ferdinand became dog food). But we digress. This appears to be a classic case of people confusing causation with correlation. The truth of the matter is that poorer folks live in environments less conducive to learning, parents spend less time with kids and schooling tends to be disruptive at a younger age. The list of factors goes on and on. In countries that do not have the incredible wealth disparities that this country has don’t have the type of results reflected here. Or can look at mixed income couples.
No Einstein.
Secretariat's foals had a better chance than a random foal.
A couple of years later, Affirmed and Alidar (SOrry don't remember the spelling) were very close. It turned out the foals of the horse that came in 2nd were great horses. That horse's stud fees skyrocketed.
Genes do count. We don't know how much, but you can't ignore facts. (OK, you shouldn't ignore facts.)
What if the NBA had a requirement that only the descendants of former pro basketball players are allowed to play in the league? That would be equivalent to modern horse racing, and under that system the NBA would have had Steph Curry and Kobe Bryant.
But not LeBron James, Michael Jordan, or Larry Bird, all of whom came from families of no particular athletic distinction. If you only look at the descendants of talented people then you will indeed find some gems. But you will miss a lot more.
The question of how this fact applies to Kevin's topic is left as an exercise for the reader.
The idea that breeding, which is based on gene selection, makes no difference to horse capabilities is completely absurd. Race horses are very different from work horses, for example, and this was achieved by selective breeding. Dog breeders have produced sub-species so different that, as Darwin said, paleontologists would likely classify them in different genera.
Race-horse breeding is something of a disaster because it has gone in the direction of producing ever-frailer legs and stronger muscles to drive the legs. This no longer gives much improvement in speed but has increased fragility.
The role of genetics in human intelligence is an extremely complex subject, probably more complex than either Kevin or the authors of the NYT piece realize, but denial that it could play any role at all on specious evidence is not constructive.
"but it's long before private schools, world travel, and test prep classes have any impact."
This might be the craziest thing Kevin has said since his stance on driverless cars.
His stance on automated cars is based on reality.
His view of kids, tho, well... Rich kids by the age of six would had been showered in books, media, computers, special daycare, preschool, etc.
So are you saying genetics have no impact on intelligence? Or are you just arguing the minutia of Keven's post?
The argument is that one must be terribly naive, not smart, or biased, to assume that the impact of being born in a rich vs poor environment doesn't start having an effect long before first grade as Kevin does.
Things like the mother's overall health and nutrition before pregnancy, and then during it, on their own have big effects on the child's development, and this is long before you the child is even born, after which economic differences keep stacking the deck in each case.
What the chart is doing is showing a correlation between wealth and high SAT scores of the children of the wealthy. But how many of those people in the top 1% had high SAT scores?
So you hate waffles?
All I said was I like pancakes.
Atticus with the dishonesty once again.
“His stance on automated cars is based on reality.”
As exists in sunny and prosperous Orange County, maybe. Still don’t see poor municipalities maintaining the streets well enough for full automated driving to work everywhere. And I don’t see much progress on resolving the whole “you might need to take over at a moments notice” problem if it starts raining, snowing or nightfall happens, even in places where streets are well maintained. (Expecting distracted humans to take over the wheel with just seconds’ notice seems like a big pipe dream, yet every automated driving scenario I’ve ever seen seems to default to this for whatever they can’t program successfully enough.)
Adoption and implementation was never going to be instantaneous.
I'm not sure why that's an argument against it existing. It clearly exists.
Why is that crazy? You think those things have an impact on the IQ level of kindergartners?
Yes.
World travel and test prep classes? How many 4 and 5 year olds do you know that take test prep classes? How many are world travelers? (Undoubtedly some children and infants go on family trips overseas but probably not in significant numbers. And it's not likely they absorb enough during their travels that it would impact their intelligence level.)
Are you that desperate to grasp at these straws instead of admitting that genetics may play a role?
I taught for a year in private school in Oakland. First grade. 6 and 7 years old. Many of them had already visited places like London and Paris.
What is the mechanism by which a trip to London or Paris increases IQ?
It’s almost like learning new places and things forces one to make new mental connections where previously there were none.
The IQ level of a kindergartner very much depends on the family they are born into, rich families can afford high quality pre-schools or nannies and also read to their kids a lot more.
Does genetics contribute?
But I do completely agree with you re high-quality pre schools.
You make some very big assumptions that being smart means you're going to be rich and the rich generally only marry other rich people, in which case, there is limited upward mobility.
Is this somehow part of your sensibilities on determinism?
Obviously there are all kinds of expectations. (Especially here in America. Part of what makes our country so great.) But at a macro level, yes, of course the smarter you are the better chance you have of making more money.
**that should be "exceptions", not "expectations"
Maybe but it's hardly a linear scale though. Plenty of smart people (especially SAT smart people) never get rich, and plenty of people who get rich are not SAT smart. I'd actually be surprised if there was much correlation between the ability to score well on the SAT and getting rich on your own (meaning not born to rich parents).
How could there not be a correlation? If you're smarter you can better conceive of new business ideas, better understand the markets, better express your ideas. If you're smarter you can get into better more prestigious schools where you get a better education and make connections that can help you in the future.
Obviously I'm not saying being smart and doing well on the SAT guarantees any of this. But I don't see how it could be possible there is not some positive correlation.
I am not quite sure what to say about the main body of Kevin's post...but in his closing sentences, I also would be very interesting in researching the differentiation between the top two cohorts, but I would include the 20% group also...that might be actually a worthwhile piece of research. Best Wishes, Traveller
“Smart people tend to make lots of money; marry other smart people ("associative mating"); and then produce smart babies who go on to get high SAT scores. Some of this is indeed due to environment, but most of it is up to heritability and genes. We're all just afraid to say so for fear of accidentally brushing up against forbidden race-IQ topics.”
Yeah, I gotta weigh in with my experience, even though I realize I may be a freak exception.
I’m Hispanic. Mom dropped out in 8th grade. Dad not much better. They divorced when I was 6, and mom and her 2 boys were poor and semi-homeless (living in public housing and then in our car for a while and then back into public housing for a while. Rinse and repeat until I hit college.)
I attended 6 schools in 8 years, often starting a new school in middle of the year (when social groups are already formed, which means I got bullied and beaten up a lot).
Buuuuuuuutttt, I wasn’t just a straight A student, I was a straight 100 student, no matter the school, no matter the subject, until I finally missed 1 word on a 100-word spelling test in 8th grade. Despite my sense of failure, i became high school valedictorian (finally got enough stability to attend same school for 4 years), which won me free tuition at college, where I won the school’s most prestigious scholarships.
See the problem?
All I could come up with for an explanation is during my first 6 years, when parents were together, some slick salesman convinced mom to sign me up for a monthly book club, but they were books way above what would be considered my reading level, which mom didn’t realize. Because of that “mistake,” before I had started kindergarten, I had read… Huck Finn …. Tom Sawyer …. Little Women (and Men) …. Captain Courageous…. East of the Sun, West of the Moon .. Call of the Wild …. Yada yada yada (a classic) and a number of thick anthologies with themes like “holiday stories” and “foreign lands.” No matter what happened after the divorce, I always carried that love of reading with me. That might have something to do with genes, but I don’t see it.
NOW, on “associative mating", I should say that, without me realizing it until recently, in every serious relationship I was in that could have led to marriage (and the one where I did get married), I was competing above my weight class. It always took a very short time, after me meeting her parents, for said parents to scream, internally, “ Oh hell no!” (Which I contemplated using as my legal middle name) .. I can clean up only so much. But “what would the kids be like” had to be part of the parents’ equation. … Would I have been happier marrying a fellow guttersnipe? Would my kids have been good students? Would my hypothetical kids have been better students if I had married the debutante? I’ll let you guess, mr. Drum. I still don’t see genes being that important, though they probably play a role
Moral of the story, maybe: GIVE YOUR KIDS BOOKS, NO MATTER HOW MANY REPUBLICANS YOU HAVE TO KNOCK DOWN TO DO IT! SCREW “GENES”. Thaaaank you for your time.
"marry other smart people ("associative mating")"
I think Kevin means "assortative mating."
My maternal grandmother had an 8th grade education. My maternal grandfather had a 6th grade education. My maternal grandfather died of black lung when his three kids were teenagers. My maternal grandmother went to work as a domestic. All three kids (my mom was the youngest) finished college and two obtained PhDs.
It's possible that education attainment isn't always a reliable proxy for intelligence.
Certainly, educational attainment may not be a proxy for intelligence, but for Drum's hypothesis (high income equals high scores because of assortative mating) to work out, it most definitely has to be a proxy for intelligence NOW. The top .1% is not full of people with 8th grade educations.
I too have grandparents who did not get to attend high school. I'm also pretty sure they would have struggled to get a 1300 on the SAT. I sailed right past 1300 without even trying very hard. It's pretty easy to get a high score when you've read so many books that the literature passages on the test are from novels you've read before.
I would guess that the difference between the top 1% and the top .1% is due to actions taken after kindergarten. Intensive tutoring can boost scores, and it may be that the top .1% is willing to spend the extra time and money to get that last burst of 50-100 points that gets them over the 1300 mark. It seems extremely unlikely that the children of the very top earners are just born significantly more intelligent than the people just below them.
One final note about assortative mating...as any statistician will tell you, there is still regression to the mean.
The top 1% and .1% are also excellent at cheating. Do watch the college admissions scandal doc on Netflix.
Excellent point. If the wealthy are naturally more intelligent and intelligence is inherited, then why did those wealthy parents have to cheat to get their kids into prestigious schools?
Um, reading the classics does not help you one iota on the Math section. Well, yes, there are "word problems" but they're not written in Faulknerian English. Maybe Hemingway English, though......
That story is true for most American families three generations ago, simply because when our grandparents lived fewer than 10% of Americans had "office" jobs. There were (obviously in your case) very intelligent people working in coal mines.
That's not true today, or, at least, it's very rare.
I think interest in reading still has a strong genetic component. Case in point: my wife and I both have advanced degrees and are avid readers. As in most weekends/evenings are spent in a comfy chair with a book if at all possible. I would conservatively estimate that we probably have 1000 books of all kinds in our home and thousands more in our offices. Our kids grew up immersed in books and conversations about literature and history and mythology. None ever became leisure readers. Ever. I would read to them, including some literary classics, but they never once picked up a book on their own that wasn't a comic book of some kind (not that that's bad, but we'd like to see some variety...). Our youngest *hates* to read. I mean, literally, will refuse to do anything involving words. Now, she reads fine in a technical sense, but only because she's forced to in school. She doesn't even like reading the recipe for chocolate chip cookies if we're baking together. She only texts using emojis and abbreviations (okthx!) Not dyslexic or anything -- just really, really doesn't like reading or writing.
They were all adopted. All very smart, very creative kids, and they'll be fine, but probably will never buy/own/read a book on their own, which we find very sad.
1. How do you know your parents were not smart?
2. Not clear what your issue us with cleaning up. I went to college with kids who had a far more deprived childhood than you. One student I remember was a Cambodian with a massive scar almost from ear to ear. A Khmer Rouge murderer had messed up cutting his throat and he survived, ended up in a refugee camp as a child, and then came to the US. He cleaned up perfectly well.
My understanding is that intelligence research suggests variation breaks down to 50% genetic and 50% environment. My direct experience is that intelligent people do stupid things at a frequent enough rate to suggest intelligence does not translate to a life well-lived.
As a geneticist, my observation is that in humans, the traits of high intelligence and good judgement are unlinked.
Case in point: Sam Bankman-Fried.
The list is long, but also includes Henry Kissinger, Rafael "Ted" Cruz and RFK, Jr.
Sam Bankman-Fraud.
Or is there bias in your observations because you're subconsciously holding intelligent people to higher standards? If my boss at my Fortune 500 employer robbed a liquor store that would be pretty surprising. If the high school drop out robs a liquor store that would be much less noteworthy.
SBF robbed a lot more than a liquor store.
So did Elizabeth Holmes.
You seem to be confusing intelligence and morality.
Just spitballing here, but MAYBE IT'S THE TEST BIAS.
If, say, you're parked on the living-room floor in diapers while Dad teaches your older sister Nannerl how to play the piano - "He always grew silent and stared up at us, transfixed" - plus many days as a toddler in or under the first rows of seats in long rehearsals, with actual composers prepping their work with your dad?
After Dad, no mean composer himself, runs out of stuff to teach you at age 8, and you evolve authority-figure sophistication by age 15, becoming positively scary with 270 works in your catalog by age 21, you will be more than ready to write 356 additional works over the following 15 years.
Circumstances, it seems, just push young energetic people one way or another.
Laughed out loud at this comment... I hear HIS son was pretty good, too.
Wow, the comments here seem desperate to deny that intelligence, which is significantly genetically inheritable, is a causative factor in any way with with economic success. Seems like a lot of motivated reasoning to suit ideological bias to me. It's nature AND nurture guys.
+1
But the lucky privileged always, always over estimate the percentage of good genes versus good luck to be born into a comfortable stable family. Once we opened up higher ed to a lot more kids, we discovered that there were a ton of smart kids from poor working class families.
Remember. The bar chart shows a SINGLE data point.
How different would the chart look if you changed the cut off from 1300 to 1250 or to 1350? I bet changing the cutoff would show LESS of an advantage to income.
I generally agree with your comments but I always worry about curve fitting to show the data in the best possible light.
OK - I'll go out on a branch and include a few more probabilities:
- If both parents are slim there is a higher probability that their children will be slim
- If both parents are athletic - there is a higher probability that their children will be athletic.
- If both parents are from affluent backgrounds - there is a higher probability that they will be in the top 10% of income
I will classify this as - DUH. So, I'm amused at the nit-picking in comments above as if there is any absolute in any of this. There are ALWAYS outliers, there are ALWAYS exceptions to trends. My parents were absolutely average students in school but all 6 of their children graduated from college. I have a doctorate and my older brother has a masters.
Kevin makes a lot of unsupported assumptions and derides anyone who doesnt accept them.
I dont really know the research in this area, but i know that this:
"but it's long before private schools, world travel, and test prep classes have any impact." is not just misleading, its an obvious lie.
Me and my wife gave countless hours and a lot of resources to our child before age 6. This matters. My genes are amazing too, but the resources started well before 6 years old.
This is pretty obvious and shouldnt need to be said.
You're kid was doing test prep classes at age 5 and younger? How much world traveling did they do before age 6? (And the travel they did do, how much did they absorb to the point it impacts their intelligence level?)
Private schools I understand. There is a big difference in kindergartners that attended some form of school beforehand and those that did not. The quality of the school (i.e. private vs other) also has an impact. This is most evident in reading levels. Some kindergartners come in being able to read and some don't know any letters. I don't know enough to say how this translates to IQ though.
You seem to think that the SAT is given to five year olds. Do you really think that the environment has no impact on intelligence by age 5?
Let's look at an example. Take Arch Manning, star up-and-coming college quarterback, a case involving the inheritability of quarterbackness since his two uncles and his grandfather were all star QBs.
But what if Arch had decided he liked soccer? Maybe he would have channeled all of his athleticism into soccer and been a terrible QB. So is "quarterbackness" heritable or is it athleticism? Would Arch be likely to be a star soccer player? What advantages would Arch have missed out on if he wanted to play soccer that he got growing up playing QB?
Do you suppose that growing up in a family that runs a prestigious academy for young QBs gave him any advantages, in addition to whatever athleticism he was born with?
Or what if he had been denied the ability to work on his athletic skills as a kid. Suppose his mother had joined a religious cult that considered intense athletic training to be a sin. Would the inherent quarterback genes he was born with have shown through? Would he have picked up a ball at age 18 and been able to compete with the best out there after never having played sports or even trained his body at all?
You see, we are completely fine thinking that things like strength and athleticism are inherited AND that they are enhanced, refined, and improved through environmental factors. Why must we be so either/or with intelligence?
Denver Nuggets NBA star Popeye Jones' two kids having grown up in the Denver area played hockey as kids and are NHL stars. It does happen.
I 100% agree with everything you said nd have not made any comments otherwise.
The dumb part was Kevin implying that direct test prep, private schools and world travel are the best or only ways to use resources to prepare a young child for future testing.
This seemed like a very obviously misleading take, so i didn't think i needed to spell it out so directly. Forgot about Atticus though.
Everyone should have an equal opportunity to succeed, but not everyone has an equal capacity to succeed. Genes do matter.
How much do genes, and making selective choices generation after generation, matter?
I have to confess that I'd rather date Margot Robbie than a chimp. Their genes might be mostly similar, but there is an undeniable difference.
Your own mileage may vary.
Margot Robbie, famous mostly for her intelligence. I'm not saying she's dumb, but her smarts are not what you are interested in. I mean I noticed you didn't say Mayim Bialik here.
There are different kinds of intelligence.
Not to derail, but
my first wife was a tall, smart, slender redhead who worked in the UM law school. Very logical, like me. Together, we made tons of money and neither of us were taking care of the emotional side of things, so we divorced.
I'm now with a woman who is much more like Margot Robbie. She is everything that I am not; caring, athletic, wise about people, and certain of her strong beliefs.
With that part of my life taken care of, I can focus on what I'm good at (economic rape and pillage) without constantly being distracted by the things that I'm not good at, and as a result, my income has more than doubled.
While being highly intelligent (IMO), she doesn't do well on written tests. (Maybe that's what she sees in me, because I do that easily.) From a distance, this union might be seen as a "regression to the mean", although I see it as a big step up the evolutionary ladder. Lol.
"Margot Robbie, famous mostly for her intelligence."
I daresay, though, she's significantly smarter than a chimp.
"When someone is shot dead, the children who live on that block perform much worse on cognitive tests in the days following the murder."
- from "Poverty, by America" by Matthew Desmond
That's a very extreme example. There are also big variances in intelligence for the vast majority of kids that don't live on blocks where people have been shot dead.
The advantages that rich kids have on cognitive tests is of secondary importance. Of primary importance is the disadvantages that poor kids have. Poor nutrition, erratic sleep patterns, high stress due to violent neighborhoods and unstable homes, periodic sudden moves due to eviction, etc. Cognitive performance is heavily influenced by one's environment, and the lived environments of many of America's poor is disastrous.
I agree. My wife, several other family members, and a few friends are all teachers. This is very evident.
Very important point. It's easier to degrade a budding intelligence than to improve a weak one. Thank you for cutting through a lot of junk above.
OT: Jenna Ellis has reached a plea deal w/ Fani Willis. Being at the top of the ladder, Trump must be sweating bullets. 😰
Excellent! Keep them coming. Anything that decreases the chance of Trump being the republican nominee
OK, I'll wade in here. Director of an Honors Program for a decade (before resigning in disgust) and I've lived, off and on mostly, in the Soviet Union and post-Soviet space for a couple decades, teaching and working there in higher ed and lower ed. So, in a way, uniquely qualified to comment on this topic.
1. Heritability is a probablistic phenomenon, not something to consider on an individual level. But when the population gets large... it's statisitics, and it's a harsh referee.
2. Heritability of cognitive ability works both sides of the table. Don't forget that it "the bottom quintile" is affected by this as the "top .1%"
3. This discussion needs to at least address perceived risk and reward. Read "Fear of Falling" (Barbara Ehrenreich).
4. On a level of pure statistics, this table has SO MANY PROBLEMS. Someone above already remarked on the "fitting the curve problem." But this table implies a curve that is immediately suspect: First, the ends are both EXTREMELY flat -- there just aren't many kids whose parents are in the bottom quintile who take the SAT in the first place (see #3 above). And, gosh, I can't start... Comparing quintiles and then "top 1%" and then... top .1%?? Are you serious? Sample size explains, Kevin, the difference between the top 1 and top tenth. Full stop there. (The problem of outliers, bad data, aleatority -- those are only SOME of the problems with very small (relative) sample sizes.)
5. And then there's another, separate problem. (Again, remember that I work in a US university AND I directed an elite honors program AND I live in the USSR.) My guess is that virtually all the kids whose parents are in the "top 20%" (highest quintile) take the SAT, so there's a separate problem in the study -- the implied curve is all wavy, with really flat tails. Any statistician would look at it and immediately dismiss it as a poor study (or poor representation of a great study, or a poor representation of a poor study, or... This world was only the first rude essay of some infant deity who afterwards abandoned it, ashamed of his lame performance...)
6. Finally, right, assortative mating. I still think Cowen's Complacent Class and Hirschman's Veto, Exit, Loyalty are the places to think through those problems...
7. And none of this is to deny the overwhelming importance of the social element. Maybe presented with this chart, I'd say "Let's make a lot more people a lot more wealthy."
8. Oh, and, someone above, the geneticist, said that there's no link between cognitive ability and good decision making. As a university professor, I'd say: Actually, there's a negative correlation. Looking around me, I'm always astonished by how many smart people make SUCH BAD DECISIONS. I call it the "calls are coming from INSIDE THE HOUSE" problem.
I've written too much...
Some good stuff in there, so thanks.
I'm curious of the impact of motivation and expectations. Something not really encompassed in Kevin's comments. How many in the lower incomes don't really try, as they assume it's moot.
No, not too much, well done.
The only thing I can add is "what are we measuring?" The SAT is set up, at the 1300 plus level, to figure out which students are good fits for top universities, which is the same as saying that the SAT is set up for the maximum possible chance of success in the information economy.
There is no reason at all that the ability of already successful parents to spend money and time on gettting their kids prepared for this test would not result in the benefit shown by the chart. Kevin doesn't have kids, so he probably has not been through the college prep/applying for college routine like many of us, and believe me, this is exactly zero on the surprise scale.
Compare any other elite performance. The NFL is not weighted towards kids of the wealthy or high socio economic class, or high education, and is partially weighted towards parental genetics.
Any of the skill sports, where it really matters how early you start, and lessons are required, like tennis and golf, are crazy related to parental ability to support that sport.
Soccer, the number one world sport, is number one in the world because the barrier to entry is crazy low and the barrier (or route) to elite status is for the most part funded by the professional teams, down a very young age. The bias in top soccer players on the male side is probably, worldwide, skewed toward low income, because on the high income parents are making kids focus on something more practical.
Anyway, nice chart, no analytical surprise.
"There's nothing controversial about how this happens, either. Smart people tend to make lots of money; marry other smart people ("associative mating"); and then produce smart babies who go on to get high SAT scores."
Brother you are out of your g0ddamned mind.
Any such effect on what we might call the "SAT score gene" would be utterly swamped by the environmental factors of privilege. I mean I know you don't have kids. I don't either but I'm not pig ignorant about children. By first grade environmental factors have had a HUGE impact, from nutrition, to parental attention, language, modeling of peers and parents, exposure to pollutants (hello, lead!), exposure to drugs in the womb and on and on. I mean, wow.
So genetics does not play a role?
As I said, they are swamped by environmental factors. The argument that rich kids score well on the SAT, ergo the richer you are, the smarter you are is just plain dumb. The problems with this conclusion are legion.
* SAT scores are NOT a proxy for intelligence
* We don't even really know what intelligence is or how to measure it or how heritable it even is.
* Environmental factors have a massive effect on intelligence
* Rich families become poor and poorer families become rich far too often for there to be any kind of big genetic effect that is easily tied to income.
* The top 1% is full of athletes and entertainers. Are we saying these are are smartest people?
"SAT scores are NOT a proxy for intelligence"
Sorry, that is not true. SAT tests are very g-loaded and thus correlate very strongly with IQ. This is why Ivy League school students have an average IQ of over 140.
"We don't even really know what intelligence is or how to measure it or how heritable it even is."
IQ measures certain dimensions of intelligence, and this is well understood. How well someone does in school on tests, including for both GPA and standardized tests, correlates very strongly with IQ. Ditto for educational achievement as well as income earned later in life. And it is well understood in terms of statistics how heritable IQ is. All of this has long been studied.
"Environmental factors have a massive effect on intelligence"
Look at Figure 2 in the study that KD linked. From age 7 to 16, for each SES group, IQ varies by only 4 points. I would not call that "massive". 20 or 30 points would be massive.
This misses the point that capability for exhibiting intelligence is swamped by environmental factors. So what if the test is geared to show intelligence. If a family practices an extensive vocabulary with children at a young age, they will be more capable than those without (has been researched). I benefited from a family of readers, read voraciously at a young age and it showed up in the verbal section of the SAT. Does this mean I was more intelligent than poor performers or just provided an environment which was geared to the testing.
"Some of this is indeed due to environment, but most of it is up to heritability and genes."
I'm not up on the science, so maybe I missed something, but "most of it" seems to overstate the "nature" case. It's obviously a factor, but you're making it out to be the whole ballgame.
Supposing that this is true (I don't buy the heritability argument - not to the degree shown here) and to answer the question:
The 0.1% kids are more likely to have been 0.1% kids their whole lives, as well as their parents.
The 1% kids have some nouveau rich mixed in.
Now, a real critique:
Does this control for anything, or is it just a chart sorted by income position?
Standardized tests have a very well known bias towards white (male) test takers doing well for essentially being the in-group in the culture at large, with the system set up for them. The SAT is the biggest standardized test around. How much of this gap is explained simply by the fact that wealthier kids are more likely to be white?
Meritocracies inevitably devolve into aristocracies once the parents in the meritocratic class discover that they can game the system to ensure the success of their offspring.
The old term for Meritocracy was Social Darwinism.
The vast gap between the upper 20%. which is where most of the technocrats fall, and .1% should be enough to strongly suggest that it has jack all to do with inherited ability. This is possibly the dumbest thing Kevin has ever posted.
To be fair, Kevin wrote something at least this dumb in this very post:
Saying that 5-year-olds can't benefit from "world travel" is 1.) false and 2.) an apparently deliberate effort to ignore all the things that Kevin lists in this very post that more obviously benefit 5-year-olds.
Yeah, I've read that the biggest determining factor for the size of a child's vocabulary is the size of their parents' vocabularies. Obviously, genetics is some factor in that, but class, level of educational attainment, vocation are all going to be pretty big factors in that as well.
Nature vs nurture on intelligence is a monumentally difficult thing to study, and to handwave away so much of the argument with "I doubt 6 year olds can be helped by environmental factors!" is just silly.
"This is possibly the dumbest thing Kevin has ever posted."
Yes.
I am honestly a little dismayed that he thought this kind of blithe ignorance and shoddy thinking was worth posting.
Kevin: "In any case, the most interesting aspect of the chart isn't the SAT differences between rich and poor. It's the difference between the top 1% and the top 0.1%. No one thinks there's any cognitive difference between these two groups"
Not sure about the reason, but note that reporting the "differences in percentage of people above a threshold" tends to magnify fairly small differences.
To see how this magnifies small differences. imagine shifting a Gaussian slightly to the right. If you pick the right cutoff on the upper tail, you can show impressive gains in the number of items above the threshold. I guess that is why journalists like to use it.
Kevin, you always call out people who do not adjust for inflation. And you are right about that. We should also call out anyone who uses "percentage of items above threshold".
But even worse is the apples-to-oranges threshold comparison you sometimes see: "While the incomes of the bottom half increased by only 10%, the number of people making more than $500,000 increased by 20%".
Folks, please stop using specific examples counter to a generalization to claim that the generalization is invalid. When one makes a claim based on statistics (assuming that the behind-the-scenes statistical work is valid, which isn’t always a good assumption), one is simply saying, “In general, this seems to be true.” A single, or even many, instances that don’t fit says NOTHING about whether the generalization is valid or not. “I fell off a 100 ft cliff and didn’t die” is not a good argument against the statement, “In general, people who fall off of 100ft cliffs die.”
A few points:
1. Lost in this discussion is the other side of the bell curve. Assuming IQ is equally distributed, people with low IQ really struggle. One indicator, the US Army will not even take anyone with an IQ below 83. Think about that, the Army does not want a janitor with an IQ of 82.
2. The data on things like SAT prep is that they work, a little bit. No way the differences shown in the chart are related to SAT/ACT etc prep.
3. When people are angry at elite colleges for accepting the vast majority of their students from wealthy families, look at the chart. The lower 50% of the income distribution is not creating many qualified, assuming you think the SAT is a valid measure, candidates.
https://skeptics.stackexchange.com/questions/45013/does-the-us-armed-forces-refuse-to-recruit-anyone-with-an-iq-less-than-83#:~:text=The%20US%20military%20uses%20IQ,an%20asset%20to%20the%20military.
https://slate.com/technology/2019/04/sat-prep-courses-do-they-work-bias.html
Trauma negatively impacts cognitive ability. Living in poverty is traumatic… in a multitude of ways. There is really no surprise that a child raised in a disadvantaged environment is more likely to be disadvantaged.
As to heritability, Sapolsky gives a good example of how heritability can happen even when there is no genetic connection. I don’t have the patience to look for the link, but one of his YouTube lectures discussed a researcher that transplanted fetuses between mice that were bred to be anxious and mice that were not. The high anxiety trait followed the the high anxiety mom indicating that environmental effects start in the womb.
I think Kevin's post is glib, to say the least. But that said, it seems wrong headed to suggest that there is no connection between SAT scores and intelligence, nor any connection between intelligence and financial success. Of course there are exceptions galore. Some of the smartest people I know are not rich and some of the richest people I know are not at all that smart. There are lots of personal factors beyond just intelligence that affect financial success, not the least of which is just the desire to have lots of money.
How big a role does genetics play in intelligence ? Hard to say exactly. How good a measure of intelligence is IQ ? What exactly is being measured ?
But for all of you who want to deny any role for genetics, the evidence provided by identical twin studies suggests pretty strongly that IQ has a significant heritable component.
But the next question is so what then ? Do we really want to live in winner take all society where the strongest smartest people get all the spoils so that they can disproportionately advantage their offspring ? I don't . I think the Rawlsian notion (oversimplified) that imagining a just society should require imagining not knowing the circumstances of your birth (that's both environment and native abilities) is a useful way to start.
I think Kevin is on the right track with heritability. That affects not just IQ, but other factors like being driven. That drive may explain the difference in the 0.1% versus 1%
This doesn't mean that one has to be rich to be driven--immigrants for instance, are a self-selected group of driven people, who overcome large hurdles in language barrier and settling into a foreign, unknown factor. This explains why 2nd generation Asian-Americans do so well even if their immigrant parents aren't able to climb up high on the SES ladder. Asians in Asia are not (significantly) smarter or more driven than Americans, but those who make the leap to immigrate to America are a self-selected group, and they pass on some traits to their kids, whether through genes or through home factors in their kids' early age.
One thing he discounts is that by 1st grade richer/more educated parents are home factors--the vocabulary of the parents, bedtime reading, etc. Immigrants also tend to lecture their kids about all the sacrifices they made for their kids, and that tends to work, even if it could be called a guilt trip.
Kevin is making and a bit conflating two points:
The article is totally downplaying the role of genetics
The chart shows a hard to explain difference between the rich and extremely rich.
Genetics vs wealth is impossible to speculate on from the article because correlation is not causation and the genetic angle was not even measured.
But just noodling, the correlation between wealth and enriching environment seem more probable to me than the that between wealth and intelligence (aside from a like attracts like for the parents).
The seeming difference of extreme wealth vs wealth seems likely to be from sampling bias or test cheating. I mean, how many extremely wealthy kids even bother taking the SAT anymore unless they are strangely motivated and likely to succeed in an elite program. For them “a successful career” means something completely different than the merely wealthy
"There's nothing controversial about how this happens, either. Smart people tend to make lots of money; marry other smart people ("associative mating"); and then produce smart babies who go on to get high SAT scores. Some of this is indeed due to environment, but most of it is up to heritability and genes. We're all just afraid to say so for fear of accidentally brushing up against forbidden race-IQ topics."
Funny, even a few minutes of Internet research suggests that the linkage between IQ and income is questionable, with some studies claiming that, after controlling for other factors, there's almost no effect on income, while others do argue for an effect. Taking something as factual and obvious that experts conducting studies do NOT take as factual and obvious seems sloppy and the sign of a bad argument to me.
Unsurprisingly, there's overwhelming agreement that parental income is closely tied to the income of their children, especially in the US, where the studies I found suggest a very outsized effect. I suspect, if I bothered to search, that in the US there'd also be evidence that parental income has a larger determining factor on marriage than IQ, meaning that you've decided to relate the two data points in this study to a third thing by ignoring income and focusing on IQ, when in fact IQ matters less according to existing studies. But I invite you to do the work to prove that intuition wrong.
What you're ignoring here, besides the effects of nurture vs nature, is that low-income children are more likely to be under biological stress, to be overweight, to be poorly nourished or even malnourished, and to be exposed to factors that harm their natural development. (Lead, for example?) These things can all be expected to depress the biological component of their intelligence (the "nature" portion) in comparison with those at mid-to-high income brackets.
What's really interesting is the jump at the 20% bracket and higher. To me, that jump implies that, once you've eliminated all the factors which can impair child development which are associated with lower income, you're likely seeing the effects of "nurture" more than "nature."
The study doesn't seem to control for who takes the SAT--it is fairly costly, so while I'd wager all children past a certain income bracket take the test, I doubt every child at the low end is taking it--and it doesn't consider the possibility that the SAT is designed in a way which grants tests something rich kids would have an advantage in.
Kevin also elides IQ (tested in a different way, and less accurately as children age) with the SAT, which is pretty definitively NOT a test of innate intelligence and probably a better indicator of education and educational support. That's obvious as test prep can substantially improve scores.