Editore"s Note
Tilting at Windmills

Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

July 25, 2006
By: Kevin Drum

NATURE, NURTURE, ETC....The gold standard of IQ research has long been studies of identical twins adopted into different homes. Twins are born with the same genes, so if twins brought up by different parents end up with similar IQs anyway it's evidence that IQ is primarily influenced by heredity. If they end up with different IQs it's evidence that upbringing is important.

Most of the research in this area has suggested that genes are more important than upbringing. But there's a flaw: poor people don't adopt very often, which means the research tells us only about children brought up in middle class environments. But what about twins brought up in poor homes? Does environment play a larger role there? In the New York Times Magazine, David Kirp reports on a well-known French study published in 1996:

To answer that question, two psychologists, Christiane Capron and Michel Duyme, combed through thousands of records from French public and private adoption agencies....The average I.Q. of children from well-to-do parents who were placed with families from the same social stratum was 119.6. But when such infants were adopted by poor families, their average I.Q. was 107.5 12 points lower. The same holds true for children born into impoverished families: youngsters adopted by parents of similarly modest means had average I.Q.s of 92.4, while the I.Q.s of those placed with well-off parents averaged 103.6. These studies confirm that environment matters the only, and crucial, difference between these children is the lives they have led.

Kirp also discusses some research by Eric Turkheimer, who noticed that even in more mundane research of twins brought up by their parents, it was mainly middle-class homes that were studied. Turkheimer went looking for more wide-ranging data:

He found what he needed in a sample from the 1970s of more than 50,000 American infants, many from poor families, who had taken I.Q. tests at age 7. In a widely-discussed 2003 article, he found that, as anticipated, virtually all the variation in I.Q. scores for twins in the sample with wealthy parents can be attributed to genetics. The big surprise is among the poorest families. Contrary to what you might expect, for those children, the I.Q.s of identical twins vary just as much as the I.Q.s of fraternal twins. The impact of growing up impoverished overwhelms these childrens genetic capacities. In other words, home life is the critical factor for youngsters at the bottom of the economic barrel.

As with all of this research, take it with a grain of salt. There are plenty of things we don't know yet, and the vast bulk of IQ research indicates pretty clearly that genes have a very strong influence on intelligence.

But it's nonetheless worthwhile to point out what ought to be obvious in any case: even if heredity has a strong effect on intelligence, so does upbringing. And even if upbringing only accounts for 30-40% of the variance, that's still a lot and it's probably at its highest in cases where home life is the worst.

Of course, it's still unclear what to do about this. Intensive educational interventions are the most obvious possibility, but results on this front haven't been very promising. Still, research like this suggests pretty strongly that we shouldn't give up. Biology isn't chickenfeed, but it's not destiny either. Especially among poor children, education and upbringing can have a considerable impact.

Kevin Drum 1:21 PM Permalink | Trackbacks | Comments (102)

Bookmark and Share
 
Comments

"Especially among poor children, education and upbringing can have a considerable impact."

Well, Duh.

Frist?

Posted by: brewmn on July 25, 2006 at 1:25 PM | PERMALINK

What if Blair were forced out of office due to his subservient relationship with Bush? That would be the Spanish PM, the Italian PM (arguably), and the British PM. While the Decider appears that he will stay in office - by popular consensus - until he is termed out. What a strange thought.

Posted by: dano3006 on July 25, 2006 at 1:27 PM | PERMALINK

Great Post Kevin -- I guess this means you now support breaking the unions' monopoly control of public schools.

Are you going to come out in support of Charter Schools?

Posted by: Norman Rogers on July 25, 2006 at 1:30 PM | PERMALINK

*** Insert Snarky comment comparing the rearing of William J. Clinton and George W. Bush Here ***

Posted by: Robert on July 25, 2006 at 1:31 PM | PERMALINK
There are plenty of things we don't know yet, and the vast bulk of IQ research indicates pretty clearly that genes have a very strong influence on intelligence.

I think the best summary, from all the reports I've seen, is that the evidence suggests that:

1) Genes play an overwhelming role in setting potential intelligence.

2) Environmental factors, particularly early childhood environment, play a major role in whether or not that potential is realized.

Posted by: cmdicely on July 25, 2006 at 1:31 PM | PERMALINK

I agree with just about everything you write. I would note, however, that while some people seem to think this is controversial among scientists, my belief is that for the vast majority it isnt.

The most radical views are not usually those actually expressed, but rather those attributed to the "other side" in on going debates on the relative impact of nature vs. nurture. Even this is a misnomer, since its not nature vs. nuture, but rather nature and nuture.

Of course, it's still unclear what to do about this. Intensive educational interventions are the most obvious possibility, but results on this front haven't been very promising.
Are you sure about that? I thought that results on this front have in fact been very promising, perhaps you could be more specific?

See, e.g. The Economics of Investing in Universal Preschool Education in California
The authors baseline estimate, which is arguably conservative, is that every dollar invested by the public sector beyond current spending will generate $2.62 in returns. And this estimate does not account for an array of other benefits not captured in their analysis because of data limitations. Those other potential benefits include lower intangible losses from crime and child abuse and neglect averted, reduced reliance on public welfare programs, improved labor market outcomes for parents of preschoolers, improved health and well-being of preschool participants, and the intergenerational transmission of favorable benefits. Broader economic and noneconomic benefits may accrue in other areas as well, including labor force recruitment and participation rates, workforce performance, economic growth, international competitiveness, and the distribution of economic and social well-being.

http://www.rand.org/pubs/monographs/MG349/index.html

Posted by: Catch22 on July 25, 2006 at 1:32 PM | PERMALINK

This is the kind of Big Idea Democrats need to rally behind:

Early Childhood Development

Posted by: WPB on July 25, 2006 at 1:33 PM | PERMALINK

Still the final word on this subject

Posted by: craigie on July 25, 2006 at 1:35 PM | PERMALINK

brewmn stole my DUH!!! Those of us who spent our careers in the classroom have known that for a long time. Especially important are the "early, formative years", but as with the concept of a low student to teacher ratio being critical to success in classrooms these ideas seem to slide right off those "experts" who think they know how to fix our public school system. The reason so many leave their teaching careers defeated, discouraged, and depressed!

Posted by: Dancer on July 25, 2006 at 1:35 PM | PERMALINK

I worry about attempts to raise the intelligence of backward children; else where would the GOP find future political candidates?

Posted by: Hedley Lamarr on July 25, 2006 at 1:36 PM | PERMALINK

This off topic but I thought it was interesting that Dan Froomkin has writen this about Specter's bill today:

Specter's so-called compromise with the White House, which he defended in a Washington Post op-ed yesterday, would gut the law that made Bush's program illegal in the first place, while sending all lawsuits about the program to a secret court.

A secret court. Think about that for a moment. University of Texas at El Paso government professor Bill Weaver has. He writes on NiemanWatchdog.org that a secret court that only hears one side of the argument and doesn't disclose its decisions is not the place for important, precedent-settings constitutional decisions.

Kevin hasn't offer any opinion on the issue but here is betting that Kevin is waiting for the GOP's side in order to find some way in order to to tell his readers why it really not so bad a bill or say thats its a loser augment for Dems because itll make them look weak on terrorism.

Any takers on this bet about Kevin and the Washington Monthly's real position on some of Bush's worse and most damaging policies?

Posted by: Cheryl on July 25, 2006 at 1:36 PM | PERMALINK

Great Post Kevin -- I guess this means you now support breaking the unions' monopoly control of public schools.

Are you going to come out in support of Charter Schools?

Right, because saving a few children via a lottery system and leaving the rest in schools that the state has already deemed as failing is a *great* way to improve public education.

Posted by: zoe kentucky on July 25, 2006 at 1:37 PM | PERMALINK

Great Post Kevin -- I guess this means you now support breaking the unions' monopoly control of public schools.

Talk about non-sequiturs. Talk about unable to process information.

"poverty matters more than anything for childhood development."

"agreed, so let's get rid of unions."

Ah, the inner workings of the fixated conservative.

Posted by: craigie on July 25, 2006 at 1:38 PM | PERMALINK

In response to the glib comment above about public school performance and private schools:

The Education Department reported recently that children in public schools generally performed as well or better in reading and mathematics than comparable children in private schools....Additionally, it found that students in conservative Christian schools lagged significantly behind their counterparts in public schools on eighth-grade math.

Posted by: Catch22 on July 25, 2006 at 1:38 PM | PERMALINK

How come the left is willing to talk about IQ, except when it involves the never-refuted scientific research of the Bell Curve?

Posted by: American Hawk on July 25, 2006 at 1:39 PM | PERMALINK

Yes; but are IQ tests biased in favor of middle/upper class languages making the test itself suspect?

Posted by: Ellen1910 on July 25, 2006 at 1:41 PM | PERMALINK

I find this to be an absolutely fascinating topic. It is probably because I am an identical twin.

Genetics has a controlling factor in determining the potential for high IQ. However, IQ tests will also vary according to cultural and environmental conditions. One need only to look at specific questions on the test to determine a subtle (although unintentional) income bias. Those from the lower economic strata of society just would not have the cultural or educational exposure to do as well.

Why is this important? It goes to the heart of the myth that this is a country in which one can get ahead merely by using his/her intelligence and hard drive. If you fail, it is because you are inadquate. The progressives have to fight against this conservative propaganda. We have to attack the right wing agenda of dragging this country back to the late 1800's. There is no better example of the failure of their outlook than Texas. This should be the shining example of the Republican myth along with the ever widening gap between rich and poor. The public is beginning to feel the effects of these policies. Now is the time to educate them to "land of opportunity" myth of the Republicans.

Posted by: Jmohr on July 25, 2006 at 1:43 PM | PERMALINK

I can't help but wonder what Andrew Sullivan's take on this will be. He promoted in TNR Charles Murray's The Bell Curve, which essentially suggested nature was the overwhelming determinant. Rightwingers loved this since it suggested that a) racism is not such a bad idea and b) liberalism doesn't work because genetics is destiny. True, many if not most observers debunked Murray's unwieldy amalgam of dubious science and sociology. But the idea got out there, which was the main thing. And as we've seen, bad ideas can have very long half lives.

Posted by: walt on July 25, 2006 at 1:44 PM | PERMALINK
Of course, it's still unclear what to do about this. Intensive educational interventions are the most obvious possibility, but results on this front haven't been very promising.

How is that the most obvious possibility? If the problem is demonstrably the material conditions in which the child is raised, how is it that "intensive educational interventions", which do not address the identified problem, are the "most obvious possibility"?

Clearly, the only type of action which even addresses the problem is action directed at improving the material conditions in which children at the lower end of the economic spectrum live.

Posted by: cmdicely on July 25, 2006 at 1:47 PM | PERMALINK

I can't help but wonder what Andrew Sullivan's take on this will be. He promoted in TNR Charles Murray's The Bell Curve, which essentially suggested nature was the overwhelming determinant. Rightwingers loved this since it suggested that a) racism is not such a bad idea and b) liberalism doesn't work because genetics is destiny. True, many if not most observers debunked Murray's unwieldy amalgam of dubious science and sociology. But the idea got out there, which was the main thing. And as we've seen, bad ideas can have very long half lives.

Posted by: walt on July 25, 2006 at 1:47 PM | PERMALINK

Once you dive into poverty, the size of the overall family becomes important.

Kids who are well-off simply grow up, one way or another.

Kids who are poor have to cope. Once that process starts, you have reaction to it. Once one of a batch of siblings takes things into his own hands, other siblings react to that & can become dependent (needy) on him. Played out among twins, one figures out how to do things (like standardized tests), the other figures out how to relate to others. Which is an ability that can be tested, but usually falls outside the scope of most studies.

A more interesting study is the extent to which twins are mutually telepathic.

Posted by: Dave of Maryland on July 25, 2006 at 1:51 PM | PERMALINK

As I think of it, it also depends on what age you run the tests.

It would be interesting to see how twins differ at 2, 5, 10, 20, 30, 40, 50, 60 & 70.

Posted by: Dave of Maryland on July 25, 2006 at 1:55 PM | PERMALINK

Somehow or other we have not mentioned HeadStart, which was a success, geared to low income children. This program has been severely cut by the Republicans, even though it was a success.

Compassionate conservatism indeed.

Posted by: Carol on July 25, 2006 at 2:03 PM | PERMALINK

I agree with Turkheimer's conclusion. I tend to suspect that the lower median scores of adopted children in low income homes probably have less to do with a lack of stimulation or early education and an excess of negative stimulation, as in stress.

PS Whoever wrote those "Head On" ads that air ceaselessly, and mercilessly on cable news these days deserves to have their skull crushed.

Posted by: Linus on July 25, 2006 at 2:06 PM | PERMALINK

I'm just going to ignore the bell curve comment...

I would like to point out that there are other issues to consider in the "nature v nurture" argument. We also need to consider neonatal environment as part of the nurture equation. While identical twins may be identical genetically, they rarely experience equality in the womb. They will most likely be sharing a single placenta, and invariably one fetus gets a larger section to gather it's nutrition. In fraternal twins the considerations are greater, since while the twins may not share the same placenta, they are still competing for resources from the parent, interfering with any effect the genes may be displaying. Diet of the parent is another factor, particularly if it falls short at crucial moments in development.

Posted by: zed on July 25, 2006 at 2:06 PM | PERMALINK

I like thinking about the fact that we have a number that means  "intelligence" when we don't have an understanding of what "intelligence" means.

Posted by: Jeffrey Davis on July 25, 2006 at 2:11 PM | PERMALINK

"Especially among poor children, education and upbringing can have a considerable impact."

You need to add nutrition to education and upbringing. Just getting poor children the proper nutrition will probably increase the IQ, especially those B vitamins which help the brain.

Posted by: Mazurka on July 25, 2006 at 2:12 PM | PERMALINK

Another issue to consider is that there's a well-known phenomenon in which CHILDHOOD IQ scores are much more strongly influenced by environment, but the environmental effect appears to "wear off" with time: children brought up in an enriched environment get "dumber" and children brought up in an impoverished environment actually get "smarter" with age. If these studies were looking at childhood IQ (it looks like they were), then it may not be anything new or surprising.

Posted by: BRussell on July 25, 2006 at 2:14 PM | PERMALINK
PS Whoever wrote those "Head On" ads that air ceaselessly, and mercilessly on cable news these days deserves to have their skull crushed.

...by someone swinging a sledgehammer while yelling "HeadOn! Apply directly to the scalp!" over and over...

Posted by: cmdicely on July 25, 2006 at 2:14 PM | PERMALINK

I thought anything associated with IQs are inherently racist. Can we just agree with the consensus that everyone is equally intelligent and just settle the issue already?

Posted by: Freedom Fighter on July 25, 2006 at 2:20 PM | PERMALINK

The Right's compassion for life seems to stop after the blastocyst stage of development. So don't expect this study to mobilize "the culture of life" to support any programs to help the post-blastocyst population.

But seriously, this isn't anything new. Research has shown this over and over again for the past 20 years. Sure intelligence (however you define that term) is hereditary. But it will only blossom in the proper environment. Underfeed a child, raise him an environment deprived of any intellectual stimuli, and deprive of him of any meaningful social interaction, and you've got a tragedy. Mulitply that by millions and you've got a national disgrace.

--Beo

Posted by: beowulf888 on July 25, 2006 at 2:21 PM | PERMALINK

You can change a child's poverty level, environment, and upbringing.

You can't change a child's genetics. (yet).

Posted by: Osama_Been_Forgotten on July 25, 2006 at 2:22 PM | PERMALINK

"You need to add nutrition to education and upbringing. Just getting poor children the proper nutrition will probably increase the IQ, especially those B vitamins which help the brain."

And, we must add, getting good nutrition and relatively low stress to the mother while she is pregnant. Environment matters from before the time the child is born. IQ has been increasing in America over the past decades (though everyone who ever has to deal with Americans in customer service wishes that intelligence and knowledge would also be increasing).

"Are you going to come out in support of Charter Schools?"

Are you going to support a comprehensive program to eliminate poverty and the negative biological and sociological results of it?

Posted by: freelunch on July 25, 2006 at 2:26 PM | PERMALINK
The Right's compassion for life seems to stop after the blastocyst stage of development.

No, its just fake from the beginning; there are undoubtedly many innocent zygotes being destroyed in Israel's efforts at collective punishment in Lebanon, and the Right doesn't care.

Posted by: cmdicely on July 25, 2006 at 2:26 PM | PERMALINK

Adding to what Linus, Zed and Mazurka have said, stability is another thing that's often missing. Poor families lose their housing, forcing the family to move and often the kids to change schools. The kids grow up without the safe haven of home, and the absence of that safety zone has a big impact on their cognitive development.

That's just one piece to the puzzle, but it's an important one. It's not just a lack of material possessions but the lack of stability.

Posted by: Kurzleg on July 25, 2006 at 2:30 PM | PERMALINK

Ack. You're a smart guy Kevin, but this post pretty much captures everything that is wrong with how the general public thinks of both IQ and of nature/nurture.

Nobody who has come within 100 yards of research on any of the above holds the simplistic and wrong ideas you describe.

Posted by: alfred binet on July 25, 2006 at 2:32 PM | PERMALINK

"I thought anything associated with IQs are inherently racist. Can we just agree with the consensus that everyone is equally intelligent and just settle the issue already?"

I realize that the Republicans think that anyone can be smart enough to be President, but reality doesn't buy that doctrine. Of course, if Republicans had to take responsibility for the facts it might cost them some money, and as it is quite clear the Republican commitment to their 'culture of life' ends whenever it would start to cost them anything.

Posted by: freelunch on July 25, 2006 at 2:34 PM | PERMALINK

Another issue to consider is that there's a well-known phenomenon in which CHILDHOOD IQ scores are much more strongly influenced by environment, but the environmental effect appears to "wear off" with time: children brought up in an enriched environment get "dumber" and children brought up in an impoverished environment actually get "smarter" with age.

It's important to remember what IQ tests are supposed to measure. They're not supposed to measure "intelligence" since nobody knows what that is. IQ tests were originally designed to measure incapcity. Incapcity isn't the opposite of capacity. There is no Bell Shaped Curve of "intelligence" so that 115 and 130 and 145 etc are each a standard deviation more "intelligent" than their predecessors. Adults living their lives, caring for children, painting their decks, having affairs or remaining faithful, fussing about bills, skipping sermons on Sunday to go fish, etc
are all "intelligent". They're living the lives they've tried to fashion for themselves and the "intelligence" which chose and fashioned that is as shut off from us as everything else which is distinctly and privately human.

IQ tests were only designed to indicate the incapcities of people who couldn't care for themselves. Big difference. Intelligence is a metaphysical construct. No number is going to measure it since measuring it would imply that there was some psychological Archimedes who could stand outside the human condition to set his fulcrum and lever appropriately.

This IQ flummery is just gas and hijinx.

Posted by: Jeffrey Davis on July 25, 2006 at 2:35 PM | PERMALINK

"Nobody who has come within 100 yards of research on any of the above holds the simplistic and wrong ideas you describe."

Your 'criticism' was meaningless.

I've heard 'did not' - 'did so' banter that made more sense.

Posted by: freelunch on July 25, 2006 at 2:36 PM | PERMALINK

it found that students in conservative Christian schools lagged significantly behind their counterparts in public schools on eighth-grade math.

Must be that biblicly mandated pi=3.0 thing.

Posted by: Disputo on July 25, 2006 at 2:36 PM | PERMALINK

alfred binet: Can you explain to what you're referring, specifically? I didn't find anything wrong with what he posted, and I'm a psychologist, not in this field, but in cognitive psychology, and I consider myself pretty knowledgeable about the state of IQ research. Twin research is the gold standard, it shows a stronger genetic influence than environmental, but environment still plays an important role.

Posted by: BRussell on July 25, 2006 at 2:39 PM | PERMALINK

genetics as we know it is kind of old-fashioned.

the new genetics deals with proteins and how their production is turned on and off.

so, for example, a person may have a gene for alcoholism, but it is never expressed because the production of the proteins that the DNA codes is blocked.

this would explain how both various environmental factors and "genetics" can affect a person various characteristics.

many geneticists are now calling for a new Human Genome project to figure out gene expression. it is a project that is considerably more complicated than its predecessor.

Posted by: Michele on July 25, 2006 at 2:43 PM | PERMALINK

Jeffrey Davis: IQ tests are basically standardized tests designed to aid in educational placement. That's how Binet used them, and that's essentially how they're used today. It's not the be-all end-all of humanity, that's for sure (I think breast cup size is much more important), but it's not flim-flammery either.

Posted by: BRussell on July 25, 2006 at 2:45 PM | PERMALINK

I would think that reducing poverty might be the mostestest obvious solution.

Personally I'd be interested in studies of identical twins and their intelligence: 1) at age 35 with and without an infant, 2) at age 40 with and without credit card debt, 3) at age 80 looking at multiple nutritional and environmental variables, etc.

Posted by: B on July 25, 2006 at 2:50 PM | PERMALINK

BRussell: IQ tests are basically standardized tests designed to aid in educational placement. That's how Binet used them, and that's essentially how they're used today.

That may be true in clinical and educational settings, but unfortunately the IQ fetish is used to promote various political agendas.

I think breast cup size is much more important

Certainly a far more interesting fetish.

Posted by: alex on July 25, 2006 at 2:53 PM | PERMALINK

"Of course, if Republicans had to take responsibility for the facts it might cost them some money, and as it is quite clear the Republican commitment to their 'culture of life' ends whenever it would start to cost them anything."

What is that supposed to mean exactly?

Posted by: Freedom Fighter on July 25, 2006 at 2:56 PM | PERMALINK

Define poor?

Seriously, what is the definition of poverty? What are the characteristics of a poor family? What are the characteristics of a well to do family? In the context of America does poverty mean exposure to hunger? Does it mean exposure to drugs? Does it mean exposure to preoccupied parents? Just how do you define poverty? By the same token how do you define well to do? I know of some very wealthy parents who expose their kids to drugs, who are very preoccupied, who don't provide their kids with an age appropriate diet. Do you define Paris Hilton's parents as poor or well to do?

Since we are talking about a French study, how do the French define poverty?

Posted by: Ron Byers on July 25, 2006 at 3:01 PM | PERMALINK

I think breast cup size is much more important

And here also there is a positive correlation with disposable income.

Ron -- good point about being a French study. Seems like poverty make more of a difference where there is no universal health care.

Posted by: B on July 25, 2006 at 3:09 PM | PERMALINK

freedom fighter:

Republicans talk a good game about caring, but prove by their behavior that they have no compassion. They like to blame genes because it lets them get a free pass. They don't want people to know that many of the bad results of poverty, such as increases in chronic illness or a lesser ability to learn, can be ameliorated because they are too selfish to pay to eradicate poverty.

Posted by: freelunch on July 25, 2006 at 3:11 PM | PERMALINK

Kevin: And even if upbringing only accounts for 30-40% of the variance, that's a lot and it's probably at its highest in cases where home life is the worst.

Of course, it's still unclear what to do about this. Intensive educational interventions are the most obvious possibility, but results on this front haven't been very promising.

Having trouble connecting the dots eh? Just look at what the studies (and you) said about the IQs of children from the lower income strata. It's not the genes or even education (as you point out). It's the money. Raising incomes means being able to pay for optional programs for child enrichment. It also means parents are less stressed and can have more energy and patience with their kids.

Posted by: NeoLotus on July 25, 2006 at 3:23 PM | PERMALINK

We seem to be equating IQ score with intelligence. Just because poor kids get lower grades on this test needn't mean that the poor kids are less intelligent. If we consider the possibility that these tests are skewed toward the middle class (and there is a lot of reason to think that they are) then the results may be more an indictment of the test, than an accurate reflection of the intelligence of the child.

Posted by: catherineD on July 25, 2006 at 4:09 PM | PERMALINK

NeoLotus

IQ is supposed to be a measure of an individual base line ability. It is not supposed to be influenced by enrichment programs. At least that is what IQ is supposed to be about. What the various IQ tests actually measure is anybody's guess.

While enrichement programs are not supposed to increase IQ other things can. For example, real starvation during a developmental period can permanently lower IQ. So can lead and other environmental poisons. So can drug use during developmental periods.

When somebody says that IQ is affected by poverty what they are really saying is that poor kids taking the test score lower than rich kids taking the test. What does that mean? Well, if you are right and the test outcome can be influenced by enrichment programs, the IQ test isn't neasuring what it is supposed to measure. It is more of a performance test like the SAT. On the other hand, if we are really talking about inate ability lower scores among "poor" people could mean a lot of things. Exposure to environmental contamination. Poor diet. Bad genetics. All sorts of things. What does it all mean? Hell I don't know, but I do know that the summaries of the studies of twins provided don't give us enough information to draw any conclusions. The only variable they seem to control is genetics. Genetics is an important variable. Probably the most important. It probably sets the high side limit. Genetics, however, is not the only variable. The hard part is pulling conclusions out of the remaining possible causes of variation. Good luck trying.

Posted by: Ron Byers on July 25, 2006 at 4:13 PM | PERMALINK

Early Childhood Interventions
Proven Results, Future Promise
Rand 2005

Parents, policymakers, business leaders, and the general public increasingly recognize the importance of the first few years in the life of a child for promoting healthy physical, emotional, social, and intellectual development. Nonetheless, many children face deficiencies between ages 0 and 5 that can impede their ability to develop to their fullest potential. The PNC Grow Up Great initiative, a program financed by PNC Financial, Inc., asked RAND to prepare a thorough, objective review and synthesis of current research that addresses the potential for various forms of early childhood intervention to improve outcomes for participating children and their families. The authors consider the potential consequences of not investing additional resources in the lives of children, the range of early intervention programs, the demonstrated benefits of interventions with high-quality evaluations, the features associated with successful programs, and the returns to society associated with investing early in the lives of disadvantaged children. Their findings indicate that a body of sound research exists that can guide resource allocation decisions. This evidence base sheds light on the types of programs that have been demonstrated to be effective, the features associated with effective programs, and the potential for returns to society that exceed the resources invested in program delivery.

http://www.rand.org/pubs/monographs/MG341/

Posted by: Catch22 on July 25, 2006 at 4:14 PM | PERMALINK

I didn't mean increase IQ. I meant "influence." I am tired.

Posted by: Ron Byers on July 25, 2006 at 4:15 PM | PERMALINK

Even zookeepers can tell you that the amount of environmental enrichment an animal experiences has a major impact on its ability to enjoy its life, solve problems, and behave in intelligent ways.

When a family lacks many of the basic needs, it spends much of its time trying to survive, and little or no time and money are spent on enriching things (books, playthings, art) and enriching behavior (playtime with elders). The purchase of a hardback book can mean doing without a week of lunch. Time that could have been spent reading to a child is spent working an evening shift. Elegant, complex meals are eschewed in favor of ones that fill you up, regardless of nutritional value.

That's what it means to be poor.

Posted by: Riggsveda on July 25, 2006 at 4:22 PM | PERMALINK

This doesn't come as news to the GOP. They've been trying to make more and more people poor for a long time because they know poor people tend dumber and dumb people tend Republican.

Posted by: The Tim on July 25, 2006 at 4:24 PM | PERMALINK

How come the left is willing to talk about IQ, except when it involves the never-refuted scientific research of the Bell Curve?>/i>

Don't you ever get tired of being an idiot?

Posted by: The Tim on July 25, 2006 at 4:26 PM | PERMALINK

I think the best summary, from all the reports I've seen, is that the evidence suggests that:
1) Genes play an overwhelming role in setting potential intelligence.
2) Environmental factors, particularly early childhood environment, play a major role in whether or not that potential is realized.

Isn't this true for pretty much everything? Genes play an overwhelming role in propensity to be overweight, an alcoholic, live to 90, etc, and environment plays a role in whether or not you end up a fat, old drunk, regardless of propensity.

You might have been born with an incredible talent for music, but if you never pick up and instrument, it's unlikely that you're going to play many contracts just because you've got good genes. I'm amazed that people even talk about this anymore.

It's obvious that being poor sucks, and that poor kids do worse in school and on tests. Also, grass is green and water is wet.

Posted by: TWAndrews on July 25, 2006 at 4:51 PM | PERMALINK

The solution is not letting poor people adopt. Even better if we could take away their kids too.

Posted by: Compassionate Con on July 25, 2006 at 4:57 PM | PERMALINK

Kevin: you really, really need to read Judith Rich Harris' book The Nurture Assumption. And her follow up, No Two Alike.

These two books are extremely accessible discussions of how children become whatever they become. A hint: nurture probably isn't what you thought it was...

Posted by: nat on July 25, 2006 at 4:58 PM | PERMALINK

See what economists from the Minneapolis Federal Reserve Bank say:

www.minneapolisfed.org/research/studies/earlychild/

plenty of impliations for social security debates

Posted by: aha on July 25, 2006 at 5:01 PM | PERMALINK

That's how Binet used them

Binet designed his tests for people of "substandard intelligence". He recognized that beyond a certain point, intelligence cannot be measured. (Tough to measure something you can't define.) Since then, we've certainly felled forests assigning numbers to test takers, but the meaning of IQ numbers beyond the artificial one of success with the test is all just gas and gaiters.

Posted by: Jeffrey Davis on July 25, 2006 at 5:21 PM | PERMALINK

The Turkheimer and the French Cross Fostering adoption study by Capron and Duyme are remarkable in that their conclusions stand apart from the body of literature, so the question arises as to whether they are accurately measuring heretofore untested aspects of IQ phenomona or whether they are just reporting artifacts of some sorts.

Consider the Turkheimer study. They separated out, and studied separately, the cohort of low IQ subjects and determined that nature played a smaller role in determining IQ than it did when compared to the larger group from which these subjects were drawn. Anyone familiar with statistical methods will recognize that this study is prone to restriction of range issues.

Let me set up an anology to aid in my analysis. Let's look at the genetic-environmental influence on skin cancer. We take a large cross section of society and we note that some particular people are more prone to skin cancer than others. After some study we conclude that genetics plays a significant role, skin pigmentation as a first order effect, in the onset of cancer. For some reason however this conclusion is deemed insufficient and we want to downgrade the influence of genetics, let's say the Sunscreen Manufacturers fund the study because they hope to show that ALL PEOPLE have equal need to use sunscreen and they want the sales of their products to increase. So we separate out the fair skinned redheads and blondes from the main group which encompassed peoples from all racial backgrounds because we hypothesize that this group, which has a disproportionate incidence of skin cancer, may have distinct behavioral traits that increase their risk for developing skin cancer. Now, within this restricted range of subjects we also parse by people who've had family history of skin cancer and we chose to study only them. Within this group of subjects there are office workers and there are people who work in the outdoors. There are people who use sunscreen religiously and cover themselves while others do not use suncreen and don't take any particular care about shielding themselves from the sun. So, what will we expect to find? Of course, we'll see that the behavior has an extremely large role to play in whether one develops skin cancer and that genetics plays a diminished role for all of the subjects were drawn from a restricted range of subjects, namely those who were fair skinned and had a family history of skin cancer. We've restricted the genetic range considerably and thus increased the influence of environmental factors.

What Turkheimer's study does is kind of the same thing. We know that SES is highly correlated to IQ, same with race. By restricting his range in the way he did he minimized the role of genetics and inflated the role of environment. He wasn't dealing with twins whose IQs ranged from 70-140, rather he was likely dealing with twins whose IQs ranged from 70-100, for keep in mind the fact that IQ, as an independent variable is correlated significantly to the dependent variable of SES, therefore the children born to low SES parents are also inheriting their, generally, lower IQ. By studying subjects in such a narrow range of course we would expect the influence of environment to rocket upwards.

As for the French study, this follows up on what we already know from studies done on Head Start children. While in their programs, when they are young of age, the influence of environment is immense, but as they get older they have more control of their environment and thus the influence of genetics increases and by the time the subjects are adults it becomes almost impossible to impose a standardized environment on the entire group. The French subjects were tested when they were only 14, unlike the adult testing of adopted children, thus we would expect that the subjects fall somewhere in between the results we would see from pre-schoolers and adults. My co-blogger has a very technical analysis of this study here.

Lastly, what the NYT reporter neglected to mention was the far larger Random Placement Korean Adoption study, which had a sample size of 1,117 compared to the French sample size of 38. In the Korean study children were randomly assigned to American families so we immediately can rule out selection bias. Further, the families were chosen from a wide SES range, with 20% of the adoptive parents not having completed high school. The Korean adoptees tended to perform to the mean for the Korean population in the US rather than to the mean of their adopted family and their siblings, despite being raised in the same environment. The SES of the adopted home had far less influence on the adoptees than it did on the children born to the parents.

Posted by: TangoMan on July 25, 2006 at 5:25 PM | PERMALINK

I should add, in case it's not obvious, that unlike the Korean study where the children were randomly assigned to the families on a take-it-or-leave-it basis, the French adoption process wasn't random, so we're likely dealing with a selection bias of some sorts and it's quite plausible that the adoptive parents were selecting children, of various ages, based on personality and demeanor, which have some relationship to intelligence. It's also likely that the troubled children were placed in homes of parents of lower SES.

Of course this is only a hypothesis on the selection bias that was present, but it should be addressed by the advocates of the research.

Posted by: TangoMan on July 25, 2006 at 6:08 PM | PERMALINK

I.Q. tests almost always call upon inculcated thinking patterns and never truly reflect innate intellectual abilities. Thus, it must follow that those who are exposed to the intellectual culture fare better.

Posted by: kntrla on July 25, 2006 at 6:45 PM | PERMALINK

Uh ... why do we assume that twins have the same genes? I know you would think so, but couldn't a mutation in genetic structure occur with any discrete individual member of a species? Has there been a systematic study which matches all genes between a twin population?

Posted by: Piehole on July 25, 2006 at 6:46 PM | PERMALINK
Uh ... why do we assume that twins have the same genes?

Because, by and large, they do.

I know you would think so, but couldn't a mutation in genetic structure occur with any discrete individual member of a species?

It really doesn't matter to the conclusions drawn from studies of numbers of pairs of twins raised in differing environments, unless, of course, there is a plausible systematic causal connection where those random mutations would cause the different environmental conditions, so that the systematic differences that show up based on the environmental conditions can be credibly explained by the genetic mutations.

Posted by: cmdicely on July 25, 2006 at 7:04 PM | PERMALINK
Lastly, what the NYT reporter neglected to mention was the far larger Random Placement Korean Adoption study, which had a sample size of 1,117 compared to the French sample size of 38. In the Korean study children were randomly assigned to American families so we immediately can rule out selection bias. Further, the families were chosen from a wide SES range, with 20% of the adoptive parents not having completed high school. The Korean adoptees tended to perform to the mean for the Korean population in the US rather than to the mean of their adopted family and their siblings, despite being raised in the same environment.

Being a recognizable racial/ethnic minority is, arguably, an important component of environment; at least, one cannot rule it out as such without presuming that substantial response to perceived race/ethnicity does not exist.

Posted by: cmdicely on July 25, 2006 at 7:07 PM | PERMALINK

Being a recognizable racial/ethnic minority is, arguably, an important component of environment;

Let's explore that a bit. So what environmental practice were the Korean adoptees exposed to that led them to perform closer to the Korean mean than to the performance of their adopted siblings? Did teachers hold them to higher standards than their white peers? Did the parents inflict stricter homework requirements on their Korean adopted children than they did their natural children? Did the parents do math drills with their Korean adopted children but let their natural children escape that activity?

Also, if we're to model a large effect on the "minority" status with respect to interactions with the "majority" then how to explain the minimal performance variance from populations where the "minority" as we term them in the US, is in fact, the majority population, and thus the interaction effect is completely absent?

Posted by: TangoMan on July 25, 2006 at 7:22 PM | PERMALINK

I.Q. tests almost always call upon inculcated thinking patterns and never truly reflect innate intellectual abilities. Thus, it must follow that those who are exposed to the intellectual culture fare better.

I wonder how you model "inculcated thinking patterns" for a 1 year old?

The correlation between the IQs of a sample of 20 year olds and their intelligence measured before the age of 1 is about 0.60.

Posted by: TangoMan on July 25, 2006 at 7:25 PM | PERMALINK

Nice, informative posts TangoMan, thanks.

Posted by: BRussell on July 25, 2006 at 7:33 PM | PERMALINK

TangoMan,

I should have made clear I'm not questioning the significance of the finding, only the characterization of the conditions as "the same environment".

(I'll comment on the significance of the study findings, if at all, once I've got a useful reference to the study.)

Posted by: cmdicely on July 25, 2006 at 7:36 PM | PERMALINK

Of course, it's still unclear what to do about this.


What is society for if not health care?

Posted by: cld on July 25, 2006 at 7:54 PM | PERMALINK

Dear Kevin:

That's a good summary. Thanks.

Posted by: Steve Sailer on July 25, 2006 at 7:55 PM | PERMALINK

only the characterization of the conditions as "the same environment".

OK, thanks for the clarification. Keep in mind this very obvious, but often overlooked, point. Even identicial twins, which have the greatest genetic similarity of all study subjects, do not share the same environment. The environment for TWIN A has the significant prescence of TWIN B, and the environment for TWIN B has the significant presence of TWIN A. Their environments are not identical. So we're never really dealing with "the same environment" and we always have to contend with the issue of how much significance to attach to the variations in the environment. Usually, we can't find much measurable evidence that TWIN A's presence in TWIN B's life is more or less significant than the reverse. Same with many other variables which lead to environmental variance.

The trick here is that some variables seem to have a lot of intuitive appeal, like racial perceptions for instance. However, since the late 60's - early 70's we've known that when we compare white, asian, black, and hispanic men at the same age and same IQ level, their incomes are no more than $1,000 (2005 level) apart. So, that take us to the question of whether unmeasurable variables are acting upon IQ levels and raising or depressing IQ. We know that IQ has significant stability throughout life and it's hard to conceive of factors that work differently on children on their first years of life. I've already linked to the IQ correlation starting at less than 1 year of life. The correlation increases in significance when the child is 3 years old. The range of environment that a 3 year old child experiences is far narrower than that of older children, so the rational investigative analysis should concentrate on determining what parental activities are at work, for they are far more significant than pre-school, grade school, televised cultural messages, peer influence, etc.

No one from my side of this issue is arguing that environment doesn't play a role. In fact I agree that genetics does set the limits on potential and that environment is the mechanism to insure that potential is reached. However, where we differ from the highlighted studies is that we do not operationalize the null hypothesis that subjects with low IQ simply haven't reached their potential due to environmental impoverishment. I don't see any support for that contention. It's really hard to explain why the children of successful Black professionals who have high IQ somehow do not benefit from their enriched environments yet the children of poor white families tend to outscore them on SATs. This brings us back to genetics and regression to the mean issues but know we're really straying far afield. And one other point on this contrapoint - the high IQ Black parents are obviously older than their children and thus were raised in earlier times. Unless someone is prepared to argue that racism today is more virulent, or destructive, than is was in yesteryear, then how to explain how the parents achieved their levels of intelligence in a more hostile, and deprived, environment and the children are regressing back to the mean despite the benefits of the enriched environment within which they are raised.

These two studies stand out from the body of literature in their counter-intuitive conclusions and the question that arises is why this is so.

Posted by: TangoMan on July 25, 2006 at 8:18 PM | PERMALINK

What about the miscreant retard boys sired by Poppy and Bar Bush? Not a brain in the litter. And those assholes were born with a silver coke spoon up their noses!

Posted by: Fred Flintrock on July 25, 2006 at 8:55 PM | PERMALINK

TangoMan,

Damn I wish I had the smarts to say what you said.:) Seriously. well done.

Posted by: Ron Byers on July 25, 2006 at 9:52 PM | PERMALINK

Shorter Kevin Drum: Poor people are stupid.

Posted by: e. nonee moose on July 25, 2006 at 10:18 PM | PERMALINK

Is there any correlation between intelligence and psychological well-being? Does being smarter make you any happier?

Posted by: SLJ on July 25, 2006 at 10:35 PM | PERMALINK

Interesting: The average I.Q. of children from well-to-do parents who were placed with families from the same social stratum was 119.6. But when such infants were adopted by poor families, their average I.Q. was 107.5 12 points lower. The same holds true for children born into impoverished families: youngsters adopted by parents of similarly modest means had average I.Q.s of 92.4, while the I.Q.s of those placed with well-off parents averaged 103.6. These studies confirm that environment matters the only, and crucial, difference between these children is the lives they have led.

These studies do confirm that environment matters -- somewhat. They also confirm that genetics must play an awfully important role. Think about it: The rich kids adopted into poor families STILL had higher IQs (107.5) than poor kids adopted into rich families (103.6). Environment helps, but not enough to overcome the effect of bad genes.

Posted by: Anono on July 25, 2006 at 10:36 PM | PERMALINK

Environment helps, but not enough to overcome the effect of bad genes.

Can you change people's genes?

No. So who cares if environment helps "enough to overcome the effect of bad genes"? Environment is the factor we can control.

Irrigating the Great Plains makes them farmable, but still not as productive as farmland in Holland. So irrigation helps, but it's not enough to overcome the effects of climate and soil composition. But who cares? We can't control climate or soil composition. So we build irrigation systems. We can't change our children's genetic potential for intelligence. But we can give them early childhood education, and redistribute income to reduce poverty. If we give a flying fuck about whether our children turn into idiots or not, that is.

Posted by: brooksfoe on July 25, 2006 at 11:23 PM | PERMALINK

But who cares? . . . We can't change our children's genetic potential for intelligence. But we can give them early childhood education, and redistribute income to reduce poverty.

Who cares? How about people who think that policies should be set in response to supportable premises. Your policy prescriptions are based on an ideologically pleasant, but evidence poor, premise.

Reducing poverty through income redistribution won't solve these problems. People trapped in poverty are there for reasons beyond not having enough money. Consider the cases of newly arrived immigrants who find themselves in poverty then slowly climb the SES ladder and their children fare much better. For these people poverty was not a permanent generational stasis. Compare their experience to the families who have a multigenerational presence within the ranks of poverty. They had the advantage of speaking the language, knowing the customs, having extensive social and family networks, etc at their disposal which the immigrants didn't have yet they stay trapped in the bottom ranks while newcomers climb upwards.

If you redistribute income to aid the people who are only temporarily hindered by poverty then they will soon not be a burden to society, however the people who are in poverty for reasons not having to do with lack of resources will not be aided by increased resources. The effect of IQ on SES is far greater than the reverse effect of SES on IQ.

As for early childhood education, that does little to raise IQ, though it can help with other personality metrics, which are important in life. A far better strategy is to enable more breast-feeding, micronutrient supplementation and enriched home-life. By the time the child enters pre-school a lot of IQ plasticity has been set.

If you want to increase the welfare of the disadvantaged then you need to craft policies based on accurate assessments of reality. This may result in a more natalist-centric series of policies rather than class-based policies. Whatever the policies advocated they should reflect reality, rather than ideologically appealing nostrums.

Posted by: TangoMan on July 25, 2006 at 11:51 PM | PERMALINK

TangoMan, nothing in your post has anything whatsoever to do with genetics. There are early childhood interventions which improve IQ. (Not to mention late childhood ones.) Who cares whether the kids they help don't end up smarter than the smartest genetically gifted kids? Why is that a rational or relevant comparison to make?

The rational news article would look like this: "Programs to change kids' environments to improve their IQs have resulted in some improvement.

"In comparison, programs to alter kids' genes to improve their IQs have...uh...

"Meanwhile, programs to allow only high-IQ people to breed in order to improve the overall IQ of the population have...yeah. Forget it."

Posted by: brooksfoe on July 26, 2006 at 12:44 AM | PERMALINK
These studies do confirm that environment matters -- somewhat. They also confirm that genetics must play an awfully important role. Think about it: The rich kids adopted into poor families STILL had higher IQs (107.5) than poor kids adopted into rich families (103.6). Environment helps, but not enough to overcome the effect of bad genes.

Children born into poor families, wherever they are raised, have pre- and, depending on the exact conditions of adoption, likely some unspecified duration of post-natal environment influenced by conditions, aside from genetics, that apply to the poor family.

Even at birth, children are not a pure product of genes, but of the combination of genes and environment. Heck, that's true even at conception.

So the conclusion you suggest is not warranted by the facts you cite.

Posted by: cmdicely on July 26, 2006 at 1:12 AM | PERMALINK

The trouble isn't so much stupid people as legislating according to the worst behaviors of stupid people, which is pretty much how Republicans govern the country. They've turned America into a grade school where everyone is punished because some little brat stole the principal's car keys, or might steal the principal's car keys.

PS Head On: apply directly to the forehead! Head On: apply directly to the forehead! Head On: apply directly to the forehead!

Posted by: Linus on July 26, 2006 at 1:13 AM | PERMALINK
The trick here is that some variables seem to have a lot of intuitive appeal, like racial perceptions for instance. However, since the late 60's - early 70's we've known that when we compare white, asian, black, and hispanic men at the same age and same IQ level, their incomes are no more than $1,000 (2005 level) apart.

I assume you mean the median or some (probably arithmetic) mean is only $1,000 apart in some study controlling for those factors, not that if you take people from the groups their income is really "no more than $1,000 apart".

Not, of course, that in any case that's more than tangentially relevant to the issues at hand, which concerns factors contributing to IQ, not differences in utility of IQ attributable to other factors.

So, that take us to the question of whether unmeasurable variables are acting upon IQ levels and raising or depressing IQ. We know that IQ has significant stability throughout life

We know that on average IQ is relatively stable, there is also plenty of evidence that that stability is not constant across differently situated groups, however, and that environmental factors distinct from those that predict early IQ predict later IQ changes (notably, declines.) And that both the environmental factors that predict low early IQ and the different environmental factors that predict IQ decline are factors that tend, generally, to correlate with low SES.

and it's hard to conceive of factors that work differently on children on their first years of life.

Huh? No its not, its actually quite easy to think of factors that work differently on children early in life, and in fact a wide array of such factors from various types of chemical exposure to elements of family environment to, well, all kinds of things are not merely "conceived of", but well documented as having special and significant effects in the first years of life.

I've already linked to the IQ correlation starting at less than 1 year of life. The correlation increases in significance when the child is 3 years old. The range of environment that a 3 year old child experiences is far narrower than that of older children, so the rational investigative analysis should concentrate on determining what parental activities are at work, for they are far more significant than pre-school, grade school, televised cultural messages, peer influence, etc.

You seem to be falsely generalizing, again, from the average stability of early IQ to the assumption that early IQ is stable for all groups; this is not a valid conclusions, and, in fact, the research I've seems to suggest that IQ stability is related to, among other factors, enviromental stability.


It's really hard to explain why the children of successful Black professionals who have high IQ somehow do not benefit from their enriched environments yet the children of poor white families tend to outscore them on SATs.

Well, it would be hard to explain if you assumed that IQ was almost entirely genetic, since those children also ought to benefit from their enriched genetic heritage as well. But if you saw studies demonstrating that IQ decline from early childhood levels seems to be correlated (among other environmental factors) with living in an urban environment, and if you had some idea about population demographics in the US, you might have the beginning of a clue as to what kind of threads to follow to understand that result.

Posted by: cmdicely on July 26, 2006 at 2:05 AM | PERMALINK
Reducing poverty through income redistribution won't solve these problems. People trapped in poverty are there for reasons beyond not having enough money.

Not having enough money isn't a reason for poverty (except inasmuch as there is a significant positive feedback loop involved), it is the definition of poverty.

Consider the cases of newly arrived immigrants who find themselves in poverty then slowly climb the SES ladder and their children fare much better.

That reminds me: newly arrived immigrants are, incidentally, one group whose low IQ children tend to improve much more rapidly than the average results on IQ stability would suggest, demonstrating that IQ stability is not constant across populations, but itself a product of environmental factors.

Aside from that, clearly, recent immigrants (whether low SES by US standards or not when they arive) have experienced different material and social environments than those low SES in the US, so don't really say a lot about whether the missing factor for advancement for non-immigrants trapped in poverty is more material resources or whether something else is missing.

And immigrants who migrate to the US aren't even representative of people of similar SES in their own countries; as there are undoubtedly a number of personal characteristics that are related to the drive to take the risk involved in emigration (and to marshal the resources and navigate the system to successfully do so.)

For these people poverty was not a permanent generational stasis.

Yes, immigrants who are poor by US standards when they arrive in the US are not generally comparable in any other way to people who are born in the US poor by US standards.

Compare their experience to the families who have a multigenerational presence within the ranks of poverty. They had the advantage of speaking the language, knowing the customs, having extensive social and family networks, etc at their disposal which the immigrants didn't have

Er, what? Immigrants to the US generally have at least some family network (immigration law itself strongly favors this), and often extensive supportive ethnic communities as well. At the same time, poor people in the US are particularly likely not to have extensive family networks, with broken households, dead or incarcerate fathers, etc.

And, I don't know if you realize this or not, but different socioeconomic strata in the US have different customs.

If you redistribute income to aid the people who are only temporarily hindered by poverty then they will soon not be a burden to society, however the people who are in poverty for reasons not having to do with lack of resources will not be aided by increased resources.

Poverty is lack of resources.

Whatever the policies advocated they should reflect reality, rather than ideologically appealing nostrums.

Quite.

Posted by: cmdicely on July 26, 2006 at 2:33 AM | PERMALINK

"Consider the cases of newly arrived immigrants who find themselves in poverty then slowly climb the SES ladder and their children fare much better."

I don't know man. All the evidence seems to suggest that my family has been getting progressively more stupid since the first one (not counting the drunk indians) got here 398 years ago.

Posted by: Linus on July 26, 2006 at 3:04 AM | PERMALINK

there is also plenty of evidence that that stability is not constant across differently situated groups,

Link to some. Then we can argue specifics. I'd be particularly interested in studies which show consistent increases in IQ over time for specific groups.

elements of family environment to, well, all kinds of things are not merely "conceived of", but well documented as having special and significant effects in the first years of life.

Link them. Show me how gov't policies will affect these factors. How is gov't going to enforce breastfeeding or compliance with the most effective parent-child forms of interaction?

No its not, its actually quite easy to think of factors that work differently on children early in life,

What I was referring to was the standard liberal responses of racist attitudes, racist policies, stereotype threat, and a host of other societal signals. These factors wouldn't exert their influence on a child's development in the first year of their life. We're talking of governmental intervention here, not parenting styles or parental vocabulary. Perhaps you're prepared to argue for parenting licenses and gov't approved parenting styles, and if you are then I'll be happy to consider your arguments. What policies are going to aid in a solution?

But if you saw studies demonstrating that IQ decline from early childhood levels seems to be correlated (among other environmental factors) with living in an urban environment

I think it's quite safe to say that I've read far more IQ studies than you likely have, but I do appreciate your bluff. Perhaps you'd like to discuss this study in some detail? I thought it was quite interesting how the racial composition of their urban cohort was 84% Black and the suburban cohort was 95% White, but they didn't think to control for that and instead chose to attribute effects solely to urban environment. Further, I also found it interesting how they didn't control for racial differences for birth weight, gestational age, birth-weight-for-gestational age z score, and instead just focused on birth weight. They did however note that the IQ disparity was already wide at 6 years of age. As I wrote above, there certainly are factors which inhibit some children from reaching their IQ potential and I'm not arguing that being raised without paternal influence, taking on the values of a malignant street culture, and being raised under different parental philosophies won't have a depressing effect on IQ, what I'm arguing is that there's little we can do about those personal problems. We already know from existing studies that improving a child's neighborhood doesn't yield an improvement in academic performance:

Families originally living in public housing were assigned housing vouchers by lottery, encouraging moves to neighborhoods with lower poverty rates. Although we had hypothesized that reading and math test scores would be higher among children in families offered vouchers (with larger effects among younger children), the results show no significant effects on test scores for any age group among over 5000 children ages 6 to 20 in 2002 who were assessed four to seven years after randomization. Program impacts on school environments were considerably smaller than impacts on neighborhoods, suggesting that achievement-related benefits from improved neighborhood environments are alone small.

you might have the beginning of a clue as to what kind of threads to follow to understand that result.

Give me a break. Your tactic of argument by assertion doesn't carry much weight with folks who recognize blowhardery when it's presented in a dressed up form. Try arguing with evidence.

Posted by: TangoMan on July 26, 2006 at 4:17 AM | PERMALINK

This has been an interesting discussion and I thank Kevin Drum and several of the commenters for it. A few observations.

First, its important to keep in mind the Jeffrey Davis point that IQ scores mainly measure success taking IQ tests and at best only very indirectly and highly imperfectly measure overall intellectual ability to function in society. Moreover, intellectual ability is surely multidimensional, as illustrated by the separate tests for mathematical and verbal skills on the SAT and the fact that females tend to score better on the latter than males while the reverse is true of the math test. The father of modern intellectual skill testing, the French scientist Binet, always resisted aggregating the results of the individual sub-tests into a single score. The question of whether IQ can be attributed to a 'single (statistical) factor' has been sharply debated and the evidence is inconclusive (or so I have read in Stephen J. Gould's book, The mismeasure of man). Given the complexity of intellectual skills, the likelihood that that they could be determined primarily by genetics -- to the extent to produce significant differences among different sub-groups of humans -- also seems quite implausible, especially given how young the human race is and how close, as a result, its members are genetically. My impression (others may correct me) is that race, for example, is regarded by most biologists and zoologists as largely meaningless in biological terms, the distinction being based mainly on cultural and other historical criteria.

Second, and related to the first point, it is not hard to imagine that exceptionally high or low intellectual skills are more determined by genetics, indeed primarily by genetics (as in the case of the genes that are responsible for Downes syndrome and some other forms of mental retardation), while skill(s) levels 'in between the tails' of the distribution are not nearly as influenced by genetics wheras environment is much more important.

Third, it is much more difficult to separate out, even conceptually, the effects of factors that interact in producing a given set of effects than when the factors act independently. The evidence cited here seems to me to illustrate that 'nature' and 'nurture' do not act independently but rather interact in complex ways.

Finally, (even) more of a question. My impression is that average IQ test scores (in the U.S.) of both caucasians and African-Americans have risen noticeably since the first half of the last century while a significant gap has remained in the average scores of the two groups. If so, this seems to be most plausibly interpreted as illustrating the importance of environment. After all, significant genetic changes typically take much longer than a century to occur and express their effects. So if genetics were the main determinant of IQ (scores), there should not have been an improvement in the average scores of the population over such a short period. It is easier to imagine that the improvement reflects the substantial improvements in health, living standards, educational methods and access, and other such factors. Likewise, one would expect a significant gap to remain among the two groups given the still significant differences in environmental conditions between the two groujps.

Posted by: CP on July 26, 2006 at 5:27 AM | PERMALINK
I'd be particularly interested in studies which show consistent increases in IQ over time for specific groups.

The most important ones are differential decreases. Since you've already linked to one that illustrates my general point, I don't think I need to link to more.

I think it's quite safe to say that I've read far more IQ studies than you likely have, but I do appreciate your bluff.

Not a bluff.

Perhaps you'd like to discuss this study in some detail? I thought it was quite interesting how the racial composition of their urban cohort was 84% Black and the suburban cohort was 95% White, but they didn't think to control for that and instead chose to attribute effects solely to urban environment.

Well, its certainly not the only study I've seen on the subject, but it certainly (regardless of whether this error is significant) undermines all of your arguments based on the premise that IQ is uniformly stable, as it (whether or not it correctly identifies the cause) demonstrates that this is not true.

As I wrote above, there certainly are factors which inhibit some children from reaching their IQ potential and I'm not arguing that being raised without paternal influence, taking on the values of a malignant street culture, and being raised under different parental philosophies won't have a depressing effect on IQ, what I'm arguing is that there's little we can do about those personal problems.

Well, that's a pretty stupid argument, IMO, as those are not areas in which it is difficult to imagine policies.

ive me a break. Your tactic of argument by assertion doesn't carry much weight with folks who recognize blowhardery when it's presented in a dressed up form. Try arguing with evidence.

Um, I've been arguing with assertion no more than you have, and the only evidence you've presented validates my criticism of a central premise of many of your arguments.

Posted by: cmdicely on July 26, 2006 at 11:13 AM | PERMALINK

The twin data shows, and has shown repeatedly, for years, that 30-40 % of IQ is not inherited. How significant that percentage is depends on your point of view. BTW, IQ studies that focus on young children show more variance due to environment. As the children age, environmentally-caused differences decline.

Why are you writing about social science? You don't seem to know much about it.

Posted by: msf on July 26, 2006 at 1:31 PM | PERMALINK

Gah-can we please at least understand the basic methodology of these studies? Twin/adoption research uses behavior genetics (BG), a population-level approach (in contrast to molecular genetics, which actually looks at [or can look at] the biological substrates of individuals). The relevent data that BG studies yield is a heritability coefficient, which is the number that is most often bandied about (e.g., "70% of homosexuality is heritable"). Heritability is NOT synonymous with inherited (i.e., transmission via genetic processes). The heritability coefficient only tells you that 70% of the variance in "gayness" is attributable to variance in genetic similarity (where, on average, MZ twins = 100% similarity; DZ/full sibs = 50%; half-sibs = 25%; adopted siblings = 0%). Meaning that the greater the genetic similarity, the more likely the similarity in the dependent variable of interest. As you can see--this in no way provides an unequivocal assessment of the role of genes vis-a-vis environment, just that there is shared covariance.

Thus, it is simply flat-out incorrect to say that IQ is 30-40% genetic--heritability values cannot and do not address this (molecular) level of understanding. Say the heritability of height is 87%; I'm 6 foot--does this mean that 5'2.64" is attributed to genes, but the other 9.38 inches are due to my environmental experiences? Of course not--this statistic (heritability) via BG methods is useless in individual cases, and really only has meaning when looking at the variation within (and not across

Posted by: softpedant on July 26, 2006 at 11:24 PM | PERMALINK

cont'd

(and not across

Posted by: softpedant on July 26, 2006 at 11:27 PM | PERMALINK

grr...

(and not across, one of the major errors in Reynolds' and Herrnstein/Murray's work) populations.

Posted by: softpedant on July 26, 2006 at 11:28 PM | PERMALINK

I said:

These studies do confirm that environment matters -- somewhat. They also confirm that genetics must play an awfully important role. Think about it: The rich kids adopted into poor families STILL had higher IQs (107.5) than poor kids adopted into rich families (103.6). Environment helps, but not enough to overcome the effect of bad genes.

CM Dicely said:

Children born into poor families, wherever they are raised, have pre- and, depending on the exact conditions of adoption, likely some unspecified duration of post-natal environment influenced by conditions, aside from genetics, that apply to the poor family.

Even at birth, children are not a pure product of genes, but of the combination of genes and environment. Heck, that's true even at conception.

So the conclusion you suggest is not warranted by the facts you cite.

This is all true, but purely hypothetical. You seem to be suggesting that perhaps the "poor" kids all were adopted at age 3 after being abandoned in a mobile home, while the "rich" kids were adopted at birth after their entire families were killed in hospital accidents. If that were the case, then yes, you'd have a systematic difference in the environments experienced by the "rich" and "poor" kids prior to adoption.

But again, that's purely hypothetical on your part. You have no freaking idea whether there was any systematic difference here. All of the children in the study could have been adopted at birth. And it's just completely wishful thinking to suggest that "pre-natal" environments caused the staggering result that rich kids adopted into poor families have HIGHER IQs than poor kids adopted into rich families. Think about what you're saying: That the pre-natal environment could have more influence on IQ than many years of post-natal environment (including family structure, poverty or wealth, schooling, etc.). This is simply unbelievable.

Nice try, though.

Posted by: Anono on July 27, 2006 at 12:10 AM | PERMALINK

Malnutrition and exposure to some things such as lead, mercury, and some pesticides during the pre-natal and early childhood years can cause permanent decreases in IQ.
In experiments with rats or mice (can't remember which), when a mother was malnourished during pregnancy, not only were the resulting offspring cognitively impaired, so were the children of the daughters.
As far as the social environment, programs such as head start have been shown to have long-lasting positive benefits.

Posted by: Patricia Shannon on July 27, 2006 at 1:33 PM | PERMALINK

As far as the social environment, programs such as head start have been shown to have long-lasting positive benefits.

Not on IQ, they haven't.

Posted by: TangoMan on July 27, 2006 at 2:01 PM | PERMALINK

Not on IQ, they haven't.

Then why care about IQ?

Posted by: ABC on July 28, 2006 at 8:42 AM | PERMALINK




 

 

Read Jonathan Rowe remembrance and articles
Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

Advertise in WM



buy from Amazon and
support the Monthly