Are American schools better than the American people?

Eduarto Porter has an excellent column on that topic, here is one bit:

In a report released last week, Martin Carnoy from the Graduate School of Education at Stanford, Emma García from the Economic Policy Institute in Washington and Tatiana Khavenson from the Institute of Education at the Higher School of Economics in Moscow, suggest that socioeconomic deficits impose a particularly heavy burden on American schools.

“Once we adjust for social status, we are doing much better than we think,” Professor Carnoy told me. “We underrate our progress.”

The researchers started by comparing test scores in the United States with those in France, Germany, Britain, Canada, Finland, South Korea, Poland and Ireland. On average, students in all those countries do better than American children.

Then the researchers divided students into groups depending on the number of books in their homes, a measure of the academic resources at families’ disposal. This adjustment significantly reduced the American deficit, especially among students on the bottom rungs of the resource ladder.

American students from families with the least educational resources, as it turned out, scored better on the PISA math test than similar children in France and about the same as Britons, Germans and Irish.

Read the whole thing.


Academic performance is a proxy for IQ (g). IQ is 50+% heritable. Books in the home of young people reflects the preferences of their parents, another pretty good proxy for IQ. This sort of study simply recapitulates the well-known, though politically-incorrect, fact that smarter parents have smarter kids, generally-speaking.

It's largely an elaborate effort to contrive a synthetic proxy for race.

American schools do pretty well on the PISA when the race of the students is taken into account, but you aren't supposed to be aware of the fact that different races average differently on tests.

I'm one of those that acknowledge race makes a difference & so also parental intelligence.

But now what? How does acknowledging that position lead to any different a school policy. What exactly are you advocating?

Sure, if we magically made all the black kids disappear overnight US PISA scores would suddenly rise. Shall we?

African-Americans are my fellow American citizens. They are part of American history. We owe them educations suited to their abilities.

On the other hand, there 7 billion people in the world who are not American citizens. We American citizens owe ourselves, including our African-American citizens, immigration policies that are based on statistical realities rather than on schmaltzy fantasies.

What does "education suited to African American abilities" mean in your book? Can you elaborate?

KIPP, for example, does well with black students (volunteers, of course) by emphasizing discipline, order, basics, and hard work. Most progressive education ideas, in contrast, are invented by smart white people and, not surprisingly, work better for white students than for black or Latino students.

There's a new history of the education reform movement, "The Long Crusade" by Raymond Wolters, the Keith Professor of History Emeritus at the U. of Delaware, that goes into detail on this. Here's my review of Professor Wolters' book:

KIPP, for example, does well with black students (volunteers, of course) by emphasizing discipline, order, basics, and hard work

So if discipline, order, basics (whatever that means), and hard work result in better educational outcomes for black kids, it couldn't have been their inferior intellectual abilities that accounted for prior disparities in, say, test scores, could it? Or are you saying that KIPP students were self-selected for those traits? In which case, Rahul's question remains unanswered.

"Or are you saying that KIPP students were self-selected for those traits?"

Yes, KIPP students are self-selected for impressive moral characters. The KIPP schools then drill like crazy, rather than use progress styles of education better suited for affluent white students.

I have no problem with multiple paths for education, but surely those paths should be chosen with individual testing and counseling.

I think it is bad reasoning, when you are trying to help an individual, to invoke "race." Not to mention, racist.

An acknowledgement that our school system, after disaggregating the stats by race, is actually pretty good would be a start. Then instead of requesting more money, we could focus on efficient schooling and getting our spending in line with other developed nations.

Americans spend a lot more on K-12 education per student than other countries. But at least our test scores aren't bad.

By the way, the place that really wastes money is Puerto Rico, which spends more per student than Utah, but gets disastrous Third World test scores. Most strikingly, Puerto Rico spends more per student on a couple of shady-sounding administrative categories than any of the 50 states.

while i have your attention could you please confirm or deny that this is you?

thank you.

Don't certain states already have race specific goals? I recall some Southern states implemented them and there was some push back from progressive circles.

Utah spends far too little per student and barely gets acceptable results only because it's filled with Utahns. Local progressive urban and high achieving suburban districts add substantial supplemental spending from various local tax bases that subsidize public school, some of them off budget, because Utah's effort is so low. The Church provides substantial Institute subsidies to schools in affluent stakes.

Point being, Utah's spending levels leave very high marginal return educational activity unfunded and is far too low to enable Americans to reach their potential. The problem is mitigated because Utahns organize deep and serious supplemental activity because education is venerated and students start with very high potential.

My research, on the other hand, shows that Puerto Rico is full of Puerto Ricans.

My understanding is that there's a lot of traditionally effective educational policies, like student tracking, which are eschewed by the modern establishment because of racial disparities. A blank slatist must assume that all racial disparities stem from discrimination in one form or another. Someone who accepts statistically different genetic norms can view racial disparities as a positive. Particularly in contexts of trying to identify talent at the right end of the bell curve. For example a national program to train elite Olympic sprinters would certainly direct disproportionate amounts of resources towards black. A racially balanced sprinting program that had as many whites as blacks would be a sure sign of mis-management.

The small PISA differences of the developed countries are unimportant, we should do nothing. In fact we should spend less on schooling.
If you want to win the silly contest have an immigration policy that recruit immigrants who have shown that they are very good at school and PISA. Call it the Urban Meyer plan.


This is really old news. Tino Sanandaji has been writing about this for years. Of course, since the truth isn't PC, his work isn't that well know. See "The amazing truth about PISA scores: USA beats Western Europe, ties with Asia." ( Quote

"What I have learned recently and want to share with you is that once we correct (even crudely) for demography in the 2009 PISA scores, American students outperform Western Europe by significant margins and tie with Asian students. Jump to the graphs if you don’t want to read my boring set-up and methodology.

The main theme in my blog is that we shouldn’t confuse policy with culture, and with demographic factors."

"In almost all European countries, immigrants from third world countries score lower than native born kids.

Why? No one know exactly why. Language, culture, home environment, income of parents, the education level of the parents and social problems in the neighborhood and peer groups norms are among likely explanations. But it is generally not true that the schools themselves are worse for immigrants than natives. In welfare states, immigrants often (thought not always) go to the same or similar schools and have as much or likely more resources per student."

"So similar to my comparison of GDP levels, let us compare Americans with European ancestry (about 65% of the U.S population, and not some sort of elite) with Europeans in Europe. We remove Asians, Mexicans, African-Americans and other countries that are best compared to their home nations. In Europe, we remove immigrants.

The results are astonishing at least to me. Rather than being at the bottom of the class, United States students are 7th best out of 28, and far better than the average of Western European nations where they largely originate from."

"Libertarians in the United States have often claimed that the public school system (which has more than 90% of the students) is a disaster. They blame this on government control and on teachers unions. However, it is completely unfair to demand that a public school in southern California where most of the students are recent immigrants from Mexico whose parents have no experience in higher education (only 4% of all Mexican immigrates have a college degree, compared to over 50% of Indian immigrants) should perform as well as a private school in Silicon Valley.

The libertarians have no answer why European and Asian countries that also have public school systems score higher than the United States (unadjusted for demography). Top scoring Finland has strong teacher unions, just as California.

Similarly, the left claims that the American education system is horrible, because Americans don’t invest enough in education. The left has no answer when you point out that the United States spends insanely more than Europe and East Asia on education. According to the OECD, the United States spends about 50% more per pupil than the average for Western Europe, and 40% more than Japan"

Read it all.

By your logic, race is correlated with IQ, and IQ has a very strong impact on student performance. Yet you seem to prefer to use race over IQ when attempting to adjust school performance for quality of student. Why not just use IQ?

For one thing, PISA doesn't have separate IQ test scores for students, it just has PISA test scores.

Surely PISA scores are very, very highly correlated with IQ...

Yes but the question is how good the education system is so the idea is to compare PISA scores to IQ scores (or race) you can't compare them to themselves

@Sailer, if race is that important why West Virginia being 93% white is at the bottom of PISA scores? How can New Jersey with 40% of population being non-white can be 4th in the US in PISA scores? There's a world outside California, just look at it.

@Axa, I don't think you really understand generalized concepts. It's like hearing a short man say: "if men are really taller than women then why am I shorter than most women?" If you're actually asking in good faith then try learning about concepts like "brain drain" to start with, or the fact that NJ is almost 10% Asian, or that New Jersey is mostly a suburban state.

"if men are really taller than women then why am I shorter than most women?"

This deserves to be quoted twice @Axa

Brain drain, that sounds interesting. But, is there any research that relates quantitatively the population trend since 1950 in West Virginia with low NAEP results?

Also, mostly a suburban state is a factor not related to race.


All of the NAEP data is available online. For example, the 8th grade math scores are

NJ - White 303, Black 274, Hispanic 283

NY - White 294, Black 262, Hispanic 265

National public - White 293, Black 263, Hispanic 271

This is more or less the common pattern (all grades, all states, all subjects). Hispanic scores are better than Black, but closer to Black than White. New Jersey probably benefits from the Cuban population. New York has a large Puerto Rican population. Asians typically outperform Whites.

Because it is inconvenient to talk about class and poverty. So much easier to blame it on skin pigment. Additional government spending can't alleviate skin pigment issues, after all. So much easier to just throw up your hands and write it off.


Life for the cognitively challenged. Adoption (Holt) and twin studies show that class has only a small impact on educational outcomes. See "Nature, Nurture and Income" by Alex Tabarrok and "The Inheritance of Education" by Alex Tabarrok. If adoption and twin studies show that affluent circumstances don't substantially raise outcomes, why would you think poverty would have a large negative effect?

In fact, Caplan and Choy ("Indochinese Refugee Families and Academic Achievement") showed that the children of poor Vietnam war refugees did outstandingly well in very same schools were black and Hispanic children were failing. 100+ years ago, Jewish children were already doing very well in spite of the meager incomes of their parents. Today, the children of poor Asian immigrants continue to do quite well.

Conversely, the children of black professional do not thrive in spite of the high incomes of their parents. See ''Black American Students in an Affluent Suburb: A Study of Academic Disengagement'' by John U. Ogbu. Quote

''What amazed me is that these kids who come from homes of doctors and lawyers are not thinking like their parents; they don't know how their parents made it,'' Professor Ogbu said in an interview. ''They are looking at rappers in ghettos as their role models, they are looking at entertainers. The parents work two jobs, three jobs, to give their children everything, but they are not guiding their children.''

Another important point is that urban schools in the U.S. are lavishly funded. The results are dismal anyway. The "additional government spending" theory has been repeatedly tested and has failed every time.

You might be out of a job, soon, if all this noticing continues and the MSM catches up to the HBDsphere.

Race is a crappy proxy for... just about anything. This study is talking about poverty and lack of resources in the home. I can't believe so many people jump to BLACK! as soon as some one mentions poverty. Once you have stacked 3 or 4 "proxies" on top of each other, your conclusions have become completely unjustified. It's like scientific uncertainty - it adds and multiplies with each layer.

Race is not a proxy for SES since the highest SES blacks perform about the same as the lowest SES whites on these tests

"the researchers divided students into groups depending on the number of books in their homes, a measure of the academic resources at families’ disposal."

Why, the solution is obvious: provide every student eligible for a reduced-price lunch with some bookshelves in that student's home, then fill them with Great Books (or something).

What could go wrong?

Maybe black USA citizens have a much more sensible attitude toward school, who is to say?

Race is an illusion, and a tiresome one. The pointless 19th century obsession with "race" lead to more troubles than can be counted, and continues to annoy today.

lulz someone's parents had no books, just a radio tuned to NPR all day

All those scientists are fools! Only NPR knows the truth! All those DNA maps showing genetic population clusters are lies made up by the race lobby!

Genetic "population clusters" exist, but do not generally agree with casual ideas about race. In particular, "black" does not correspond to a single ancestral population. Africa is incredibly heterogeneous. Two people from different African villages are usually as genetically different from each other as Europeans are from Asians.

TonyB breaks out Lewontin's Fallacy.

TonyB, yes popular "race" is a very crude measure though statistically still very meaningful.

However, the bit about African genetic diversity is mostly a canard. Most of the differences are in genes that are not expressed. They do have higher phenotypic variation as well but certainly not to anywhere near the extent of your claim

The overwhelming majority of America's black population can trace their ancestry to an area of Africa the size of Germany. The slave trade didn't bring a random subset of Africans. It was largely from a strip of West Africa coast from Senegal to Nigeria.

A fairly diverse area to be sure, but it's not like Kenyans, Sudanese or Zimbabweans were being trafficked to the New World.

Moreover, there is very little over the last 8 generations or so that tie African-Americans genealogically to specific tribes in Africa. For example, the African American novelist Zora Neale Hurston was a Ph.D. student under Franz Boas in the 1920s. She was surprised to find in the West Indies blacks who identified with specific African tribes, such as Yoruba, since that was, at the time, extremely uncommon in the U.S., where genealogical mixing had largely led to diversity among blacks' ancestral origins (apart from their white admixture) being irrelevant and thus forgotten. By 1920s, most African Americans were already part Yoruba, part Ibo, and so forth.

The "gap" as well as absolute performance of young American students has improved more on "math" than "verbal" tests in recent decades. I think this is because of "teaching to the test" which is much more effective for basic math where memorization-- partly of "facts" like multiplication tables, and partly of algorithms-- is key and easily inculcated by rote learning. Verbal tests are actually more g-sensitive than tests of basic arithmetic* (witness the old SAT analogies which were infamously g-sensitive) so rote-learning doesn't help much. This explains why verbal performance has not improved so much and the gap hasn't narrowed; the verbal tests are still measuring innate intelligence which teachers cannot much influence.

*Higher math is different, but vanishingly few students tackle higher maths in primary school.

You can get better at things like vocabulary by reading more on your own time for fun. But almost nobody does math problems at home for fun, so schools matter more for math scores than for reading/verbal scores.

Sure, where this really makes a difference is at the ultra-elite level of math competitions like the IMO. The cognitive difference between a national Math Olympiad competitor and the most gifted math student from a random middle-American country is not substantial. But the former most likely attends an elite prep school and spends most of his free time practicing on Putnam like problems.

You can get better at things like vocabulary by reading more on your own time for fun. But almost nobody does math problems at home for fun

I guess this is where culture makes a difference. In India (at least among my school-going compatriots back in the 80s and 90s), I observed that it was far more likely for kids to try and solve math problems for fun than to read for fun. It's evidently the opposite in the US. Among those of us who were trying to get into grad school in the US, we found the quantitative section of the GRE to be laughably easy; an average Indian high school graduate could get a high score. But the verbal section would literally kill us, since few had the vocabulary or were much into reading to handle the questions.

I think Indians are f********* morons. Browns can never compete with yellows - leave aside whites.

You probably work in QA and kiss up to white engineers.

I think white people are just great!! They are the best in the world. Much more intelligent than the rest of the world especially black brown and yellows!! hehe. Also now that middle-aged nincompoop white people are killing themselves, white race is getting stronger IQ wise.

Is the g-loading of analogies actually due to a direct causal relationship (i.e., analogies are hard because it's hard to figure out the relationship), or just to the fact that high-g individuals tend to have large vocabularies (i.e., analogies are hard because they use obscure words).

My recollection from taking the SAT was that all the analogies were easy if I knew the words, whereas a handful of the math problems were legitimately difficult even having memorized all the relevant rules. I got a 5 on the Calculus BC test in tenth grade even though my class only covered the AB material, so this wasn't a matter of being weak in math.


My intuition is that there's no such thing as a hard verbal problem, it's just a question of vocabulary. But the g-loading is well established, so I'd like to understand it better. My intuition seems too simple to be the whole story.

Math is probably the only subject where it really makes sense to teach to the test. Maybe physics and chemistry too ...

I find that very strange. Math and physics were subjects I hardly used to study for; as long as I understood the basic concept and knew the underlying laws (like conservation of energy) and axioms (sum of angles in a triangle is 2 right angles), I could handle any problem throw at me during an exam. For humanities subjects, I really had to study hard, and I found teaching to the test helped me score higher.

As I mentioned above, this probably indicative of a cultural difference between India and North America. Kids here, smart or dumb, preferred or did better at science subjects than humanities. Definitely in the case of boys.

I was good a rules-based stuff too, but terrible at biology because of all the rote memorization. Humanities came relatively easy to me but I think that's just because I liked to read a lot as a kid. I imagine it would have been much harder if my home life reflected a different culture.

Your exam questions were very most likely very easy. Claiming that just knowing axioms of euclidean geometry is enough to solve problems in exams is a moronic move. Another high-caste idiot sucking up to white folks!! hahaha

If the test is what you are supposed to know, what subject would teaching to the test not be good for?

In literature or history, for example, it's good to allow teachers to teach to their strengths, and allow for different interpretations.

Anything in the lower grades, teaching to the test is fine.

I dare you to "teach to the test" for reading comprehension and also not actually teach reading comprehension.

I guess it might be possible.

IQ is a proxy for something that is very poorly defined.

Don't define academic performance as a proxy of a poorly defined thing.

You mean g? Not sure that it is that poorly defined

What's g?

Academic performance is a proxy for IQ (g). IQ is 50+% heritable.

Can you define precisely what heritability means in this context? If I "inherit my IQ" from my dad, and let's say my dad has an IQ of 100, will I have an IQ of 100 with 50% probability? What will be the probability that I'll have an IQ of 120?

I understand what heritability means in the context of, say, red hair, or lactose tolerance, or adaptation to high altitude, but I don't understand what it means for an variable that falls in a normal distribution.

Twin studies compare sibling IQ among genetically identical twins, unrelated adoptees in the same house, and normal siblings with some genetic differentiation. Regressions tells us that 50-60% of IQ is determined genes, 0% by shared environment (how your parents raise you), and the rest is noise or unshared environment. Could be you had a better peer group, but probably not if your parenting style and support didn't matter. I think it's just random variation in neural development that can't be controlled by genes.

There are a whole lot of genes, non of which we have identified, that effect IQ/g, but thanks to twin studies we can tell how important they are anyway.

Let's unpack that 50% is genes. Statistically everyone starts at 100, but there is a semi Gaussian distribution from 100. That is, the odds of a small gain, to 103 are high, but the odds of a large gain, to 130 are small.

For most people, at the center of the curve, I think you can throw 50% influence in the street and not miss it. Average intelligence is average, whether it exactly matches a twin or not.

And genuinely high intelligence is rare, whether that person has a twin or not.

This I think 50% influence misses the tree for the forest.

Haha !! brown man is confused by probability distribution over continuous variable as opposed to discrete domain. Trying to appear equal to white men. Deluded moron like his other high-caste brothers !!

"IQ is 50+% heritable"

This is only true in upper middle class American families which share an enormous number of confounding characteristics due to class. As studies in France and other places have shown, In families with fewer resources, heritability is much lower.

I wouldn't say it's ONLY true for upper middle class American families but yes, heritability is lower for kids with FAS

I suppose it might surprise you if this study was not using books as a proxy, but was authentically looking at educational resources. Of course, that might not as conveniently confirm biases.

Now I understand how climate scientists feel during their endless arguments.
Certain ideas in behavioural genetics are rejected by both the right and the left. Consequently marginal papers which suggest the opposite of well established findings often gain widespread circulation. For example generally speaking adoption does not raise IQ however their exists a small number of papers which fail to control for the maliability for IQ in young children, or for paternity issues, or use tiny sample some sizes, or have never been replicated. It is pretty much a certainty that someone will attempt to demonstrate the opposite of the weight of evidence by citing one of these papers. The same is true of heritability estimates. Someone is bound to cite the paper by Turkheimer suggesting income/ses affects heritability not realising that no-one has been able to replicate the size of effect that he found or that in general investigations into this effect are just as likely to find either nothing or a reverse effect. One of the most consistent findings of all (consistent across adoption, Twin and GTCA studies) is that shared environment (what everyone thinks of as environment) is somewhere between zero and <5%.

Here's my graph of 2012 PISA test results breaking Americans out by race:

"So, Asian Americans outscored all large Asian countries (with the exception of three rich cities); white Americans outperformed most, but not all, traditionally white countries; and Latino Americans did better than all Latin American countries. African Americans almost certainly scored higher than any black majority country would have performed."

Just a question. According to 2010 Census, Massachusetts has the following compositions: 76.1% non-Hispanic White, 9.6% Hispanic. 6.6% Black, 5.3% and 2.4% other categories. Massachusetts is far from the states with higher non-Hispanic white population such as Maine with 94.4%. Indeed, 26% of the US population lives in states with higher fraction of white population. Why Idaho and Montana are not at the top of PISA scores?

Massachusetts has been the center of American academia since 1636.

Thus, since Massachusetts has been the center of American academia intelligent hispnics tend to move there?

Not particularly. Massachusetts Hispanics score lower on the SAT than, say, Texas Hispanics:

In general, Hispanics average score per state on the NAEP fall in a fairly narrow range: Florida, for example, has the highest score, but even Florida Hispanics don't do terribly well.

Why West Virginia is at the bottom of US PISA scores?

West Virginia is definitely not the center of American academia. West Virginia has suffered massive brain drain, making the remaining West Virginia white population a striking outlier in NAEP scores:

Because all the smart West Virginians left West Virginia and raised their high IQ children in other states.

"African Americans almost certainly scored higher than any black majority country would have performed"

Barbados? I think the success of Barbados might show that IQ is not so important.

The alt-right line on Barbados is that its affluence in entirely due to banks and hotels managed by whites. Not that they can give you a reference to any statistical source or even a set of case studies which would demonstrate that.

Once you adjust for enough things, almost any performance can be made to look good.

+100 :)


Once you adjust for enough things, you can have a more accurate understanding of performance.

If you partition the sample in 1000 different ways, you would expect to find some in where US students do better and some where US students do worse. Nothing to see here.

But surely it is logical to partition by race, SES or IQ since these are things that will affect test scores independently of schools. It is not as if the authors have partitioned the population 1000 random ways and then picked out the ones where the U.S. did better due to random chance. When you partition as logic suggests you should and then find the U.S. does way better in every segment that is quite meaningful.

It seems immigrant blacks perform pretty decently on tests, and generally do well in life compared to African-Americans. And according to a number of commenters here (and on other IQ-obsessed websites) African-Americans, even the wealthy ones, don't seem to do well in IQ tests on average.

Has there ever been a low-status sub-population in any society that has been at at par intellectually with the average person in that society? Especially a sub-population that has had its self-respect beaten out of it throughout the course of its existence? I wonder how much of this is psychological. If someone is permanently obsessed about their low status and victimhood, and if broader society has no expectations from them, that person is unlikely to become an intellectual (e.g., the way schizophrenia depresses IQ.)

A book by Amy Chua and her husband that came out last year seems to indicate that self-respect, even self-esteem, coupled with a sense of mission (plus of course a strong work ethic) is THE determining factor for success in American society. Other societies might be different, but they are all likely to be more ethnically-obsessed than Americans are.

-Jews in pre-war Europe, Chinese in Southeast Asia

Jews are a good example, but until Jewish emancipation in the 18th/19th centuries, they had no achievements I am aware of. And didn't they have a competitive advantage in the banking (formerly referred to as moneylending) business after emancipation, given their trans-national links and Christian disapproval of usury. The intellectual achievements seemed to come later, when Jews tried their best to be loyal nationalists and integrate into their host societies. I'm not sure African-Americans have a competitive edge in anything, plus there seems to be resistance to full integration within American society.

Were Chinese ever low status in SE Asia? I thought they settled as merchants with links to the rulers. That would make them high-status in my book, and would seem to be the cause of the resentment the native population expresses towards them.

Isn’t being disliked by most of the population the definition of “low-status”?

Not in my book. If you have privileges most of the population lack, like influence with the rulers and a de facto monopoly on international trade, then you are both high status and the target of resentment. "Dislike" is not synonymous with "resentment". The Chinese could dominate commerce in a way that most natives could not, just like the Jews.

Though I agree with you about the Jews. Even during the periods when they were despised and restricted to ghettos, they seemed to not fall prey to victimhood and low standards. Perhaps because of their self-image as a "chosen people"?

They adjusted for one thing. Sometimes thats is all it takes

No, they published one adjustment. They for sure, tried many more.

In any case the "one thing" they innocuously call "social status" seems to include many variables including number of books in the home, mother’s education level, family characteristics and lord knows what else.

So you prefer the adjustment by race?

All I see is 'number of books in homes'. If you can point to something else I stand corrected.

Perhaps all the other states should try to mimic Utah and North Dakotas school systems?

Our Fair Advocates of Open Borders should consider this. It won't change their mind, of course, but let's not pretend its easy.

What would be you prior estimate for differential test performance for an elementary school class of 25 students where teacher and students all speak the same language, compared to one where 5 speak minimal English, and 15 speak negligible Spanish, and the teacher is learning basic Spanish? This is a common situation in my home town.

I especially like evaluations of test results which account for things like academic resources available in the household.

If it weren't for a great local library and loads of academic resources at home, I don't think I would have done half as well in school.

Before this is all over, people like you will have teachers in black and latino homes 24/7, constantly lecturing, reading to them, sketching out formulae on the paper towels, etc. Black and latino women will be forced to report to special "resource centers" at the first sign of pregnancy, so speakers can be wrapped around their abdomens to pipe in Mozart and readings of The Iliad.

If it weren’t for a great local library and loads of academic resources at home, I don’t think I would have done half as well in school.

IOW, you had smart, low time-preference parents who married each other. Anyway, you will be thrilled to learn that modern technology has delivered a "great local library" into every home with an Internet hookup.

Bingo, AG.

Not forced, just encouraged.

Nudged. Ok, with maybe a little financial incentive. And if they can take a selfie of their ballot, that would be helpful, too.

I grew up in a dull rural white town. No significant library to speak of. Very limited internet. None of my classmates took advantage of "academic resources at home", lol. Yet our high school performed much better than the state average and had a large portion of graduates going to college. Nathan no offense... you regularly speak with great confidence in politically correct ideas but I'm not sure about the amount of time you've spent interacting with people of different upbringings. Meeting people of different races in college or in an upper income neighborhood doesn't count.

I bow down to you and other white men!!!! You are indeed the master race. You get born and then you conquer the world with your superior IQ!! Rest of the world and especially high-caste Indians on this board can only genuflect!! hehe!!

What do we make of 2003's Natl. Assessment of Adult Literacy, which suggested then that functional illiteracy afflicted millions of American adults?
What tells us today that those perceived or recorded deficits have not persisted or that they were ever overcome or that they have not been repeated in the interim for yet another generation of American adults?
(Was the NAAL measurement methodology really so shoddy its results could not be trusted?)

I think this study is very much saying that the lack of "family educational resources" have an impact.

those guys defined literacy up so that more people would fail. Pretty much everybody can read.

which suggested then that functional illiteracy afflicted millions of American adults?

But did not bother to tell you that 'functional illiteracy' is an arbitrary construct used by people jonesing for grant money or attention? Johnathan Kozol claimed in 1982 that 60 million Americans were illiterate but hey baby just follow my prescription for an all-volunteer participatory literacy program and...

I think cultural belief systems play a big part in the success of students and immigrants, and am marginally inclined to believe there are racial differences.

But I find the arguments so intractable (and nauseating) that I wish the robots would just take over already.

Yes, there are hindering cultural beliefs such as honor. At least Latin Americans care about it, honor is a thing you don't have any control over it. People can hold lifetime grudges, instead of cooperating with each other, that affect their income and well-being. For example, Mexicans still have issues with having a woman as a boss, or even worse the wife/partner having a paid job. Several machos still believe that if the wife works it signals he's not strong enough to provide for the family, poor but honorable. Nowadays, you laugh at any people that takes a duel as a serious thing. Young people shooting each other still operate under those beliefs. So, the day Mexicans laugh at honor instead of celebrating it on music, they'll advance.

Duels and lethal gang battles are an extreme example. A better one is: some guy acts obnoxious at a meeting. An Anglo would say: what an asshole (he has a problem). Someone from a Latin culture is more likely to say: that guy really disrespected you (you have a problem). This type of thinking encourages pettiness to spread.

Yes, extreme examples. An everyday issue is trying to give job related feedback to a Latino. It's always received as "you're against me".

This post by Cowen is a companion to yesterday's post about IQ differences in developed countries and less developed countries, but not in the way Cowen imagines. In less developed countries there's little cooperation, not so much because of low IQ, but because of dissimilarities, dissimilarities in education, religion, goals, economic and social status; in other words, inequality. I asked what happens to a developed country when people become more dissimilar, more unequal? My answer: cooperation declines, conflict rises. This new study discussed by Porter seeks to grade school systems by applying a handicap, much like a handicap in golf where the less skilled golfer is given strokes so that she can "compete" against the more skilled golfer. Of course, the handicap doesn't improve the skills of the less skilled golfer it only gives the illusion that she can compete with the more skilled golfer. The alternative to a handicap would be lessons from a professional, to raise the skills of the less skilled golfer to those of the more skilled golfer. But lessons take time, and money, so the convention in golf is the handicap. Now with this new study the convention in assessing school systems is the handicap: different school systems aren't different, the students are different, so like golfers they will be graded by using a handicap system. There, problem solved: the two school systems are now the same. Of course, the alternative, like in golf, would be lessons from a professional, but that takes time, and money, so the convention in assessing school systems is the handicap. I have often commented that when I was a child almost all of the children were similar, similar in color, similar in goals, similar in economic and social status; in other words, equal. Not so today. And the dissimilarities are not in color alone. In my time there were two physically separate and unequal school systems: one for white children and one for black children. Today, there are two separate and unequal school systems operating side by side, one for the students with the ability and the desire to go on to college and one for the students who don't. The latter face a lifetime of low achievement, low income, low economic and social status. Unfortunately in life, there is no handicap system. What we are doing is accepting an unequal society, one with profound dissimilarities, very much like those in less developed countries.

"This post by Cowen is a companion to yesterday’s post about IQ differences in developed countries and less developed countries, but not in the way Cowen imagines. In less developed countries there’s little cooperation, not so much because of low IQ, but because of dissimilarities, dissimilarities in education, religion, goals, economic and social status; in other words, inequality."

-Which is why San Francisco must look like crap, while Detroit must be thriving.

At least the rest of your comment is not idiotic.

Of course, SF and Detroit are in the same country, at least nominally. The segregation of America by education and income and the hollowing out of the heartland are a product of excessive inequality, as the well-educated congregate in places like SF while the less educated congregate in the heartland. There's lots of cooperation in places like SF because of the similarities of the people who can afford to reside there, but little cooperation between places like SF and Detroit because of the dissimilarities. Nominally there is one America but in reality there are two. Unfortunately, inequality is likely to increase, which means relatively more people stuck in the heartland and relatively fewer people in SF. What will the residents of SF do? Build a wall around SF?

San Francisco exists because Oakland exists, as well as some strategically located state parks on the peninsula.

As America continues to fill with people from impoverished, disparate cultures, yes I expect SF will eventually build a wall.

Of course, SF is shorthand for the whole Bay area. Cowen's good friend Peter Thiel wants to create his own country where the highly educated can congregate without any responsibility for the riff-raff here in America, not to mention a popular secessionist movement in the Bay area.

Can we at least discuss the idea that in an age of robotics and "smart" devices, one way to combat inequality is stop importing more of it?

"I expect SF will eventually build a wall"

Isn't that what the SF zoning and land-use policies effectively do?

San Francisco exists because Oakland exists,

No, both exist and exist in the states they do because of the intersection of populace and public policy. Oakland has generic crime rates about 90% higher than San Francisco's and homicide rates about 3x San Francisco's. Oakland also has a deficient police census. New York's per capita census is 3x Oakland's.

Oakland also has a deficient police census.

Right. Because that's what keeps San Francisco from being Oakland: lots of cops.

Just think what the murder rates on the Upper West Side would be without the heroic and unflagging efforts of the NYC police and DA's office.

Right. Because that’s what keeps San Francisco from being Oakland: lots of cops.

No, what keeps Oakland from being an ordinary locus is not enough cops and also cack-handed efforts to collect property taxes in depressed neighborhoods.

Dismaying the extent to which alt-right partisans have no interest in addressing ordinary social problems through ordinary means because the genetic-determinist circle jerk is just so amusing to them.

@Rayward If I thought that it would work I would support it. Have read up on the "Moving to Opportunity" study results?
The Congressional mandate authorizing the demonstration directed evaluation of its impacts on the housing, earnings, and education of the family members in the treatment groups. Researchers found that voucher recipients lived in lower-crime neighborhoods and generally had better units than the control group families, but the experiment had no impact on earnings or educational attainment.[5] It did, however, have unexpected results in health and happiness. Parents in families who moved to low-poverty areas had lower rates of obesity and depression,[6] and positive impacts on behavior and outlook among young women (but not young men) were also noted.[7]

I think most people above, and the title, are wrong to view "number of books" as a purely dependent variable, especially those who go so far as to view number of books as a proxy for race.

While I am sure "books" correlates with family intelligence, there is still the strong possibility that some parents are guessing wrong about the number of books they should have, and that some children are below potential as a result.

I believe that "read to your child" promotions have in fact raised student performance.

Notably, do e-books count? Do they count if you downloaded 1,000 of them in a bootlegged torrent?
Number of books is shortly to become a measure not of intelligence but of hoarding behavior.

My intuition is that "parents reading" either to the child, or as a demonstrated behavior, is what helps orient the child towards academic achievement.

As I say below I have no idea what the tech impact is and will be. We'll get the "toddler with an iPad" generation into the standard tests when?

Does it work if a robot reads to your child ? And in theory, how is that different from plopping you kid in front of the TV?
Do they need to see the written words? What if you put closed captioning on?

No seriously, I think the reading to your kid thing is overrated, or more accurately it is an indirect measure of something else. Either children acquire spoken language faster the more they are exposed to parents talking (which would mean that TV/Radio exposure should also be beneficial), or children tend to be acculturated to reading books by seeing their parents do it, which is something else entirely. Training kids to value book-reading isn't necessarily going to raise their intelligence level, it's more of an signifier of social class. In the internet era, it might make more sense to train them to write code or perform efficient web searches.

All I can tell you is that my dad had a master's in child development, and he paid me a quarter for every Doctor Suess book I could read. That was a lot of money in those days.

I guess your dad wasn't aware of all the research showing that external rewards destroy internal motivation. What the heck did your dad learn in that Masters program?

USC, and perhaps he had me pegged as a capitalist early on

Oh, well, I agree that actually getting kids to read would likely have benefits. It's the benefits of reading TO kids, especially those too young to understand the stories, that I'm not sure of. Some people promote reading to infants. I can't see how reading the tales of Peter Rabbit to a 1 month old is going to have any benefit over putting that 1 month old in front of a TV or a radio. Once the kid gets old enough to learn to read, you read to them and point at the words in the book and thus teach them to read. But some people seem to argue that there's a magical effect from merely seeing the parent reading aloud from a book, without seeing the words, that somehow makes kids magically smarter.

Pretty sure there is research on this

I'm another who has been pointing out we do a pretty good job once we adjust for race, which is the closest thing we have to a proxy for IQ, which is what Carnoy is getting at with social status.

A couple points to add:

While Steve is correct that KIPP has had some success, it's worth recognizing its limited nature. What KIPP does well is help a homogenous group of black and Hispanic middle schoolers get closer to 8th grade ability levels by the time they leave 8th grade. They do this by drill and kill, but also--and this is key--by adding something like an entire month to the school year with the additional funds they get from philanthropists. And, of course, by attrition of the less motivated.

However, there's very little evidence that KIPP alumni do better with that "head start" (heh) once they get to high school, or that their results are remarkably different from motivated black and Hispanic kids in public schools. KIPP alumni have about a 30% college graduation rate. That sounds impressive, but the motivated kids in even the most appalling urban high schools are doing that, too. It's far too easy to look at really bad schools and conclude, from the outside, that all kids are just wilting and giving up. But in fact, even the worst public schools have teachers and counsellors educating the kids who want to be educated. Even MIchael Brown's high school had kids scoring proficient on exams and going to school.

In fact, a recent study (which I can't find right now, will dig up later) showed that KIPP high schools (which have nowhere near the happy story of KIPP middle schools) had very little impact on KIPP middle school alumni. Kids new to KIPP in high school had a big boost, up to KIPP alumni level--and then nothing.

Note that all of this suggests capacity issues, which is a theme running all through our education efforts. Sitting kids in focused schools and drilling them endlessly, or just focusing on test scores at the elementary school level, charters or not, has had tremendous impact on 4th and 8th grade scores--but no impact on high school scores.

All we're really doing is time shifting. In the past, even low-achieving kids would eventually hit their maximum ability level by high school. These days, more of them are hitting max ability at a lower age.

But it ain't doing a thing to improve algebra scores.

This is consistent with the other conversation above, about the stagnation of reading scores relative to math. It's much easier to move the needle on basic arithmetic skills in elementary school. You can teach kids fractions, math facts, number lines, they'll vaguely remember more of it at test time. But teaching kids vocabulary is very difficult. It may be possible if we drilled and killed on vocabulary directly, but that's a hell of a lot more soulless than doing it in math.

Anecdotally, I grew up in a home with one slightly above average and one slightly below average IQ parent (say 110, 95), neither of whom went to college. I have a very high IQ. I devoured everything I could read: the backs of Clorox bottles, John D. McDonald potboilers, whatever. The idea that kids with the necessary IQs are sitting around passively, opportunity lost because their parents aren't providing the proper environment, is just not credible. Research doesn't back it up either, or preschool studies would be more compelling.

I've written about some of this here:

Teachers can't raise IQ, but adoption can? Explain.

Can you explain your question?

Googling casually I see studies that show 9 to 14 IQ points gained in adoption. I am not sure the details, but those are respectable, and certainly support two ideas: that IQ is malleable and that parental commitment matters. This result is also in line with the main article, that for a given starting place in parental educational resources, our schools do well.

There are many studies that show adoption does not raise IQ so I'm not sure how to explain what you say except that in most social science areas there will be studies on both sides. I have not read the studies you refer to but I think the balance of the literature is certainly on the no-IQ gain to adoption side.

I think part of this might be that indirect studies are applied the wrong way.

Here is a direct study: Kids who are adopted get a boost in IQ

When people just look at biological/adoptive parent correlations they may miss the simpler and direct question, what does a change in environment do?

The first link is interesting and seems like a good study, although it has a lower effect (<5 IQ points- still very significant). The latter relates to children who were abused or neglected, which I certainly can believe lowers IQ.

I do wonder how to square it with the twin adoption studies and other evidence that parenting does not affect IQ/outcome

Perhaps twin studies have too many happy households to show the net effect of unhappy to happy movements.

In the Swedish study they are just looking at lower to higher education though. You have to think that would be the case most of the time with adoption.

If parents not having enough Mozart CDs and literary classics lying around the house is the problem, then yes, adoption into affluent, intelligent families would be the solution.

I tend to discourage this idea, because I don't want my descendants taxed for reparations to future black and latino "Stolen Generations" who were removed from their biological parents and put in foster homes in an attempt to improve their test scores.

"Oh no, this might imply a tax" is a limiting world view. I can certainly think of many volunteer and charitable efforts that expand parental educational resources.

There's this one device called the Internet, that can beam classics of literature and mathematics lectures right into your home. We'll see how that works out.

The impact of the Internet is really interesting, but I'm not sure it is clear cut. It provides both reference and distraction. I am not sure that it improves test scores as much as it improves the ability to goof off and then scramble to get stuff done. Perhaps the next generation will have less in their heads, but be better able to find answers. One hopes.

30% college graduation is about the same rate as the general public

I think you are a bit too confident in your conclusions

Actually, KIPP results are up to 45% now (updated), and 34% is the rate for all students. But what I'm saying is that if you controlled for ability and interest, KIPP isn't giving that much of a boost to college completion. I don't question at all that it helps some kids who are borderline interested and would be pulled off path by kids in the usual urban school environment. But motivated kids do better than is generally known in high poverty schools. (I've taught at two Title I high schools, one 65% Hispanic, one 30-30-30-10).

Also, it's not unusual for black/Hispanic kids in suburban schools to go to KIPP, because they will get better grades and face less competition.

Sadly, there's a difference between completing college and academic achievement these days. That is, the KIPP kids finishing college (as well as the non-KIPP kids) aren't necessarily doing anything approaching top 30% work. Again, I'm absolutely not dismissing KIPP's achievements. But there's a difference between raising academic achievement and corralling kids with average ability, drilling them, supporting them, and then sending them off to colleges who are primed to likewise encourage black/Hispanic kids. To say nothing of the schools that give preference to KIPP alumni.

I've wondered about how much intensive drilling on basic math would help in more advanced math.
I figured it would help some. If you're using up mental energy remembering how to do things like add fractions, then you may not have much mental energy left for the higher-level thinking.
There may be some similarity to the situation with music. To get good at playing a musical instrument, the basic techniques have to be automatic, and it seems to help if those basics are acquired at an early age.
One of my kids did EPGY math up through pre-calc. The elementary school level math had "math races" to help make basic calculations automatic. Later he did Mathcounts, and higher-level stuff got pretty automatic.
But it sounds like, for a lot of people, the higher-level stuff is just beyond their abilities, even if the low-level stuff is automatic.

Ok, so its not that American kids are stupider than French kids. It's that American parents are stupider that French parents. If you adjust for having an overall dumber population, American schools are doing pretty good.

This is a continual shuffle among "the parents," "the teachers," "the culture," "lead," "sleep deprivation," "taciturn negro households," and on and on and on.

I'm not sure why the idea that mind is a function of brain, and brain is formed via genetic coding, and that genetics can cluster to produce people with differing characteristics (and yes, distributions within those clusters) is so repulsive to people who idolize Science! Does anybody seriously believe that anyone can be Isaac Newton given the same academic inputs?

I'm comfortable with the idea that intelligence is genetic, but I'm not convinced that those intelligence genes cluster together with genes for skin color and other physical attributes. There are many different genes governing skin tone and hair texture, they can't all be right next to an intelligence gene. Plus there are so many epigenetic effects that can profoundly impact intelligence all the way from fetal stages through infancy.

"...all the way from fetal stages [And not a day earlier!]through infancy."

Getting into phenotypes is distracting. But since you bring it up, the test score gaps among racial groups persist, from elementary school through graduate school and across income levels. Within income groups, the same disparities are observed for criminal behavior.

I don't get it. You understand that intelligence could be partially genetic. You understand that different human populations have differing genetics for skin color and other physical attributes. So why is it not possible for these populations to also have differing genetics relating to intelligence? There is no need for intelligence genes to be attached to skin color genes. The sickle-cell genes aren't connected to the skin color genes in black people, for example.

There is strong evidence that the average genetic IQ potential of a population can significantly change due to selection pressure or mating selection within just a few centuries. If that's the case, then it's hard to believe that human populations would all keep the same IQ if separated for thousands of years.

You are all over the map with nonequivalent terms.

India has great genetic diversity and diversity in skin tone. How many races does India have? Is there any reason intelligence should track the divisions you just created? We any of these races actually populations in an ethnographic sense?

The trick here is that people invent "races" for political purposes, for example to say who gets to be a slave, and then the legacy is self-referential.

There is strong evidence that the average genetic IQ potential of a population can significantly change due to selection pressure or mating selection within just a few centuries.

Can you summarize what form this evidence takes? Is it an extrapolation from the set of achievements of that population? Like adoption of agriculture, urbanization, metallurgy, etc?

If this is true, then it's not really the case that blacks on average are intellectually inferior to whites, is it? It's just that the breeding ratio of dumb blacks to smart blacks is higher than the breeding ratio of dumb whites to smart whites. So eugenics, or encouraging smart blacks to breed faster than their dumber brethren, ought to bring up their IQ to "white: levels within a few generations, right?


IQ levels in India seem to align well with the caste hierarchy, just like they align with the racial hierarchy in America. Status and self-respect seem to matter a lot, though genetics cannot be ruled out.

Kris, by invoking caste you highlight that race, or some equivalent, is always subjective, and more parochial than exponents are usually aware.

It is hard for an underclass however they get defined. What were the IQs of Russian imperial serfs compared to educated commies a generation later? I would guess somewhat lower on standardized tests.

To go all California on you, we can approach life with positive or negative energy. When we read these studies as an opportunity to improve people's lives, that's positive energy. If anyone waves away the data, sticks to biases that we can't help people because they are bad in one way or another, that's negative energy.

When you read "books in the home" as a "proxy" for your prior bias, that's negative energy.

People who do bad in school can have great lives and be great citizens in the USA. PISA is a silly competition that promotes national pride.

I did much bad in school.

Mostly I stole things and beat people up.

I am again struck by the following direction of public discussion:

"American healthcare is a terrible deal as measured by metrics x and y in comparison to other countries. We need to adopt their policies of price caps and centrally controlled national delivery."

"American education is a terrible deal as measured by PISA is a terrible deal in comparison to other countries. We should spend more money per head even though they all spend less than we currently do."

One possibility is the international comparisons are not all that great in either case and we aren't doing all that bad in either case. It is possible that inside the aggregate story is a great healthcare system for innovation and a great educational system for academic achievers. It is also possible these two things are not comparable to each other because reasons . I think the most likely way to resolve these things is to observe that they are both fundamentally aesthetic not objective claims.

This particular study is not at all about spending more in schools to improve scores, it is saying look to the home.

Or more broadly, books are underrated.

"Books in the home", eh?

Poor old French, when having books in the home counts against the efficacy of your education system, these people, inclined towards a positivity to intellect, amusement at the witty, and the general larding of their bourgeois residences with many of the works of Jean Sol Partre and the like (even if they do not read!), will not do so well.

On the other hand, Americans are quids in under such a measurement scheme! USA! USA!

Related: How the Talmud Became a Best-Seller in South Korea

"The reverend’s thesis is that the Jews have thrived for so many years because of certain educational and cultural practices, and that such benefits can be unlocked for Christians if those practices are taught to their children. "

Comments for this post are closed