Monday, March 21, 2016
An open letter to the Virginia Tech community
By Charles Murray
Last week, the president of Virginia Tech, Tim Sands, published an “open letter to the Virginia Tech community” defending lectures delivered by deplorable people like me (I’m speaking on the themes of Coming Apart on March 25). Bravo for President Sands’s defense of intellectual freedom. But I confess that I was not entirely satisfied with his characterization of my work. So I’m writing an open letter of my own.
Dear Virginia Tech community,
Since President Sands has just published an open letter making a serious allegation against me, it seems appropriate to respond. The allegation: “Dr. Murray is well known for his controversial and largely discredited work linking measures of intelligence to heredity, and specifically to race and ethnicity — a flawed socioeconomic theory that has been used by some to justify fascism, racism and eugenics.”
Let me make an allegation of my own. President Sands is unfamiliar either with the actual content of The Bell Curve — the book I wrote with Richard J. Herrnstein to which he alludes — or with the state of knowledge in psychometrics.
The Bell Curve and Charles Murray
I should begin by pointing out that the topic of the The Bell Curve was not race, but, as the book’s subtitle says, “Intelligence and Class Structure in American Life.” Our thesis was that over the last half of the 20th century, American society has become cognitively stratified. At the beginning of the penultimate chapter, Herrnstein and I summarized our message:
Predicting the course of society is chancy, but certain tendencies seem strong enough to worry about:
An increasingly isolated cognitive elite.
A merging of the cognitive elite with the affluent.
A deteriorating quality of life for people at the bottom end of the cognitive distribution.
Unchecked, these trends will lead the U.S. toward something resembling a caste society, with the underclass mired ever more firmly at the bottom and the cognitive elite ever more firmly anchored at the top, restructuring the rules of society so that it becomes harder and harder for them to lose. [p. 509].
It is obvious that these conclusions have not been discredited in the twenty-two years since they were written. They may be more accurately described as prescient.
Now to the substance of President Sands’s allegation.
The heritability of intelligence
Richard Herrnstein and I wrote that cognitive ability as measured by IQ tests is heritable, somewhere in the range of 40% to 80% [pp. 105–110], and that heritability tends to rise as people get older. This was not a scientifically controversial statement when we wrote it; that President Sands thinks it has been discredited as of 2016 is amazing.
You needn’t take my word for it. In the wake of the uproar over The Bell Curve, the American Psychological Association (APA) assembled a Task Force on Intelligence consisting of eleven of the most distinguished psychometricians in the United States. Their report, titled “Intelligence: Knowns and Unknowns,” was published in the February 1996 issue of the APA’s peer-reviewed journal, American Psychologist. Regarding the magnitude of heritability (represented by h2), here is the Task Force’s relevant paragraph. For purposes of readability, I have omitted the citations embedded in the original paragraph:
If one simply combines all available correlations in a single analysis, the heritability (h2) works out to about .50 and the between-family variance (c2) to about .25. These overall figures are misleading, however, because most of the relevant studies have been done with children. We now know that the heritability of IQ changes with age: h2 goes up and c2 goes down from infancy to adulthood. In childhood h2 and c2 for IQ are of the order of .45 and .35; by late adolescence h2 is around .75 and c2 is quite low (zero in some studies) [p. 85].
The position we took on heritability was squarely within the consensus state of knowledge. Since The Bell Curve was published, the range of estimates has narrowed somewhat, tending toward modestly higher estimates of heritability.
Intelligence and race
There’s no doubt that discussing intelligence and race was asking for trouble in 1994, as it still is in 2016. But that’s for political reasons, not scientific ones. Once again, the state of knowledge about the basics is not particularly controversial. The mean scores for all kinds of mental tests vary by ethnicity. No one familiar with the data disputes that most elemental statement.
Regarding the most sensitive difference, between Blacks and Whites, Herrnstein and I followed the usual estimate of one standard deviation (15 IQ points), but pointed out that the magnitude varied depending on the test, sample, and where and how it was administered. What did the APA Task Force conclude?
“Although studies using different tests and samples yield a range of results, the Black mean is typically about one standard deviation (about 15 points) below that of Whites. The difference is largest on those tests (verbal or nonverbal) that best represent the general intelligence factor g” [p. 93].
Is the Black/White differential diminishing? In The Bell Curve, we discussed at length the evidence that the Black/White differential has narrowed [pp. 289–295], concluding that “The answer is yes with (as usual) some qualifications.” The Task Force’s treatment of the question paralleled ours, concluding with “[l]arger and more definitive studies are needed before this trend can be regarded as established” [p. 93].
Can the Black/White differential be explained by test bias? In a long discussion [pp. 280–286], Herrnstein and I presented the massive evidence that the predictive validity of mental tests is similar for Blacks and Whites and that cultural bias in the test items or their administration do not explain the Black/White differential. The Task Force’s conclusions regarding predictive validity: “Considered as predictors of future performance, the tests do not seem to be biased against African Americans” [p. 93]. Regarding cultural bias and testing conditions: “Controlled studies [of these potential sources of bias] have shown, however, that none of them contributes substantially to the Black/White differential under discussion here” [p. 94].
Can the Black/White differential be explained by socioeconomic status? We pointed out that the question has two answers: Statistically controlling for socioeconomic status (SES) narrows the gap. But the gap does not narrow as SES goes up — i.e., measured in standard deviations, the differential between Blacks and Whites with high SES is not narrower than the differential between those with low SES [pp. 286–289]. Here’s the APA Task Force on this topic:
Several considerations suggest that [SES] cannot be the whole explanation. For one thing, the Black/White differential in test scores is not eliminated when groups or individuals are matched for SES. Moreover, the data reviewed in Section 4 suggest that—if we exclude extreme conditions—nutrition and other biological factors that may vary with SES account for relatively little of the variance in such scores [p. 94].
The notion that Herrnstein and I made claims about ethnic differences in IQ that have been scientifically rejected is simply wrong.
And so on. The notion that Herrnstein and I made claims about ethnic differences in IQ that have been scientifically rejected is simply wrong. We deliberately remained well within the mainstream of what was confidently known when we wrote. None of those descriptions have changed much in the subsequent twenty-two years, except to be reinforced as more has been learned. I have no idea what countervailing evidence President Sands could have in mind.
At this point, some readers may be saying to themselves, “But wasn’t The Bell Curve the book that tried to prove blacks were genetically inferior to whites?” I gather that was President Sands’ impression as well. It has no basis in fact. Knowing that people are preoccupied with genes and race (it was always the first topic that came up when we told people we were writing a book about IQ), Herrnstein and I offered a seventeen-page discussion of genes, race, and IQ [pp. 295–311]. The first five pages were devoted to explaining the context of the issue — why, for example, the heritability of IQ among humans does not necessarily mean that differences between groups are also heritable. Four pages were devoted to the technical literature arguing that genes were implicated in the Black/White differential. Eight pages were devoted to arguments that the causes were environmental. Then we wrote:
"If the reader is now convinced that either the genetic or environmental explanation has won out to the exclusion of the other, we have not done a sufficiently good job of presenting one side or the other. It seems highly likely to us that both genes and the environment have something to do with racial differences. What might the mix be? We are resolutely agnostic on that issue; as far as we can determine, the evidence does not yet justify an estimate". [p. 311].
That’s it—the sum total of every wild-eyed claim that The Bell Curve makes about genes and race. There’s nothing else. Herrnstein and I were guilty of refusing to say that the evidence justified a conclusion that the differential had to be entirely environmental. On this issue, I have a minor quibble with the APA Task Force, which wrote “There is not much direct evidence on [a genetic component], but what little there is fails to support the genetic hypothesis” [p. 95]. Actually there was no direct evidence at all as of the mid-1990s, but the Task Force chose not to mention a considerable body of indirect evidence that did in fact support the genetic hypothesis. No matter. The Task Force did not reject the possibility of a genetic component. As of 2016, geneticists are within a few years of knowing the answer for sure, and I am content to wait for their findings.
But I cannot leave the issue of genes without mentioning how strongly Herrnstein and I rejected the importance of whether genes are involved. This passage from The Bell Curve reveals how very, very different the book is from the characterization of it that has become so widespread:
In sum: If tomorrow you knew beyond a shadow of a doubt that all the cognitive differences between races were 100 percent genetic in origin, nothing of any significance should change. The knowledge would give you no reason to treat individuals differently than if ethnic differences were 100 percent environmental. By the same token, knowing that the differences are 100 percent environmental in origin would not suggest a single program or policy that is not already being tried. It would justify no optimism about the time it will take to narrow the existing gaps. It would not even justify confidence that genetically based differences will not be upon us within a few generations. The impulse to think that environmental sources of difference are less threatening than genetic ones is natural but illusory.
In any case, you are not going to learn tomorrow that all the cognitive differences between races are 100 percent genetic in origin, because the scientific state of knowledge, unfinished as it is, already gives ample evidence that environment is part of the story. But the evidence eventually may become unequivocal that genes are also part of the story. We are worried that the elite wisdom on this issue, for years almost hysterically in denial about that possibility, will snap too far in the other direction. It is possible to face all the facts on ethnic and race differences on intelligence and not run screaming from the room. That is the essential message [pp. 314-315].
I have been reluctant to spend so much space discussing The Bell Curve’s treatment of race and intelligence because it was such an ancillary topic in the book. Focusing on it in this letter has probably made it sound as if it was as important as President Sands’s open letter implied.
But I had to do it. For two decades, I have had to put up with misrepresentations of The Bell Curve. It is annoying. After so long, when so many of the book’s main arguments have been so dramatically vindicated by events, and when our presentations of the meaning and role of IQ have been so steadily reinforced by subsequent research in the social sciences, not to mention developments in neuroscience and genetics, President Sands’s casual accusation that our work has been “largely discredited” was especially exasperating. The president of a distinguished university should take more care.
It is in that context that I came to the end of President Sands’s indictment, accusing me of promulgating “a flawed socioeconomic theory that has been used by some to justify fascism, racism and eugenics.” At that point, President Sands went beyond the kind of statement that merely reflects his unfamiliarity with The Bell Curve and/or psychometrics. He engaged in intellectual McCarthyism.
See you next week.
Posted by jonjayray at 10:52 PM
Monday, February 29, 2016
Ho Hum! More Leftist nonsense about IQ
As with Leftists in general you have to look past what the author below says to what he doesn't say. It is true that tracking down a particular gene for any given type of behavior is in its infancy, though some progress has been made with IQ. But we don't need to know that. We can assess inheritance by twin studies. And for many years now we have found that identical twins reared apart are amazingly similar whereas non-identical twins reared apart can be quite different. And that shows how much we owe to our genes. In the case of IQ the twin studies indicate that about two thirds of it is inherited.
The author below, Oliver James, refers to Prof. Robert Plomin, a leading behaviour geneticist, but he totallly misrepresents what Plomin says. Plomin is a very active researcher and I read his papers frequently. He is the last person to deny genetic influences on behaviour. He studies them all the time. There is no point in listing his academic articles here but you can find here an article in which he discusses his research and conclusions. Believe Plomin on Plomin, not some Leftist nutter.
See also my recent comment on Plomin's work here. It gives the link to Plomin's own comprehensive study.
You would not guess it from Mr James's deceptions but there is in fact a steady stream of findings coming out all the time about IQ and its genetic base. I have collected my various posts over the last couple of years on the subject into a single blog, an IQ blog. I have done that mainly for my personal ease of reference but I think anybody browsing through the entries there will be amazed at the wide-ranging influence of IQ.
Mr James is just a liar. He says he had a difficult childhood. I believe it
When I was ten, my parents were informed by my headmaster that I was born stupid, and would have to move to a school for the congenitally defective.
To be fair, I was a badly behaved slacker who was always at or near the bottom of every class (the weekly beatings did not help). But the interesting thing is that it was not my genes that made me a thicko.
Although hardly anyone outside the world of science is aware of it, research in the past decade has proved for the first time that no one is made dim or bright by their genes, or for that matter, mad or sane.
It’s finally being established that your character and mentality is not in your genes. The age-old nature-nurture debate is over, and nurture has won.
Don’t take my word for it: Professor Robert Plomin, a behavioural geneticist at King’s College, London, one of the world’s leading experts in this field, said last year: ‘I’ve been looking for these genes for 15 years and I don’t have any.’
Or look at the huge 2013 study of the genes of twins, whose title told you all you need to know: ‘No genetic influence for childhood behaviour problems from DNA analysis’. Many other studies have had similar findings.
Yes, significant genes for differences in physical traits, like height or eye colour, have been identified by the international quest for genes known as the Human Genome Project.
But no genes that matter have been found for psychological traits.
Posted by jonjayray at 9:36 PM
Friday, February 5, 2016
Some new/old findings about IQ
A paper titled "Top 10 Replicated Findings From Behavioral Genetics" has just come out with Robert Plomin as lead author. The finding of the paper is an embarrassment to most psychologists. We now know that most findings from psychological research are NOT replicable. They are a flash in the pan with no generalizability. They tell us nothing. So the fact that findings about the influence of genetics on behavior ARE replicable makes them stand out from other research. It is putting it a little to strongly to say that it is the only sort of psychological research that it worth bothering with, but it gets close to that. I say here why I gave up on survey research after 20 years of doing it.
I have always noted that the heritability of IQ is by far the best replicated finding in psychology but Plomin shows that other effects of genetics on behaviour are highy replicable too. Leftists hate all mention of genetics so on that issue, as on many others, they are on the wrong side of history. And how ironic that is precisely the most well substantiated findings in psychology that are too politically incorrect for general mention.
So why are studies in the genetics of behaviour so robust? Plomin suggests five sensible reasons but let me give a more impressionist reason: It is because genetic effects on behaviour are REAL. There really is something going on there. And, as Plomin's other findings show, what is going on is that genetics have a strong and pervasive effect on ALL behaviour. As Plomin points out, even family environment is not an influence in its own right. It too is affected by genetics. I am reminded of something Hans Eysenck said to me around a quarter of a century ago: "It's ALL genetics". Already in his time, he had seen how pervasive genetic influences were.
My days as an active psychological researcher are long gone and I read very little in the psychological research literature these days. I have however kept a watching brief on research on IQ. So I was well aware of one of Plomin's more surprising findings: The influence of IQ GROWS as the person grows up. IQ is only a small influence of behaviour in early childhood but a large influence on the same person's behaviour in adulthood. The genetic infuence in fact seems to keep growing until about age 30. That can be seen as rather counterintuitive. One would think that a small child had ONLY genetic influences to guide his behaviours but as he grew up he would come under all sorts of additional influences on his behaviour.
Plomin explores some possible reasons behind that finding but I think he misses the obvious: A child is very heavily regulated whilst growing up. He is pushed in all sorts of directions by parents, teachers and others. It is only in adulthood that he is reasonably free to "be himself". And that is exactly what happens. He throws off most of his environmental influences and behaves in a way that feels good or right to himself.
Posted by jonjayray at 8:32 AM
Friday, January 29, 2016
IQ and health: More discomfort for the Left
The Left hate IQ because it conflicts with their crazy "all men are equal" gospel. So they never give up hope of discrediting the whole concept. They have tried all sorts of arguments but the phenomenon is so robust and so pervasive that it has easily survived all the assaults aimed at it. It cannot be explained away.
But the unending stream of scientific findings showing how important IQ is to one's life chances has now mostly driven Leftists back to their most basic defence-mechanism: Denial. They just refuse to think or talk about it. They act as it doesn't exist -- with results that range from the hilarious to the disastrous.
For those of us who think reality is important, however, the recent report below should be of interest. The basic finding -- that high IQ people are healthier -- actually dates back to the 1920s but it is nice to see current research coming to the same conclusion. That's the pesky thing about IQ: Careful research into it always leads to the same conclusion
The findings below in fact support something I have been saying for a long time: That high IQ is an index of general biological fitness. The brain is just one of the body's organs and if it is functioning well, it is likely that the rest of the body is functioning well too. As a great Rabbi once said: "To him that hath, more will be given him" (Matthew 13:12). Jesus was not an egalitarian
Clever people are more likely to be healthier than those with a lower IQ due to a genetic link between how our bodies manage diseases and intelligence.
Researchers from Scotland analysed data from around 100,000 people held in the UK Biobank.
They compared each person's mental test data with their genome and found that traits linked to disease and thinking skills shared the same genetic influences.
In particular, the international team of scientists led by the University of Edinburgh found 'significant negative genetic correlations' between a person's education and verbal-numerical reasoning skills and Alzheimer's disease, coronary artery disease and strokes.
In other words, well-educated people who excel at problem solving are less likely to contract the conditions.
Clever people were also less likely to be overweight. [I like that one]
The team found there was a negative genetic correlation between body mass index and verbal-numerical reasoning, while a greater risk of high blood pressure was associated with lower education.
The researchers explained: 'Our results provide comprehensive new findings on the overlaps between cognitive ability levels, genetic bases for health-related characteristics such as height and blood pressure, and physical and psychiatric disorders even in mostly healthy, non-diagnosed individuals.
'They make important steps toward understanding the specific patterns of overlap between biological influences on health and their consequences for key cognitive abilities.
'For example, some of the association between educational attainment - often used as a social background indicator - and health appears to have a genetic [cause].'
However, the team added: 'It has not escaped our notice that there are multiple possible interpretations of these genetic correlations.
The results of the latest Edinburgh-based study build on previous research that found 95% of the link between intelligence and life expectancy is genetic.
Using a study on twins, experts from the London School of Economics found brighter twins tend to live longer and noted the pattern was much more pronounced in fraternal - non identical - twins, than identical pairs.
By looking at both fraternal twins - who only share half their twin's DNA - with identical twins, researchers were also able to distinguish between genetic effects and environmental factors, including housing, schooling and childhood nutrition.
'Not only might particular genes contribute both to cognitive and health-related traits, but genetic variants relating to health conditions could have indirect effects on cognitive ability and vice versa, [on] lifestyle choices.'
As an example, poorly educated people may be less likely to make informed choices about what they eat and how much they drink.
The study is not all good news for intelligent people, though.
The team did find that the genetic variants associated with obtaining a degree were also related to a higher genetic risk of schizophrenia, bipolar disorder and autism.
Edinburgh University Professor Ian Deary, who led the research, said the study could help in understanding some of the links between low levels of cognitive function and poor health.
Psychologist Saskia Hagenaars, who worked on the research, added: 'The study supports an existing theory which says that those with better overall health are likely to have higher levels of intelligence.'
The UK Biobank, launched in 2007, is a major long-term investigation into the respective contributions of genetic predisposition and environmental exposure in the development of disease.
The findings are published in the journal Molecular Psychiatry.
Posted by jonjayray at 8:33 AM
Sunday, January 24, 2016
Poverty and IQ again
Charles Murray showed a couple of decades ago that the poor tend to have lower IQs. And it was hardly a surprise that being dumb might keep you poor. But the Left purport to love the poor so Murray was furiously attacked over his findings -- though he had not in fact said most of the things he was alleged to have said. It was a very cautious and scholarly book rather than any kind of polemic. The Leftist rage at Murray finally exhausted itself but Murray still has his marbles and is an active Facebooker so I imagine that he could give you more details of the "controversy".
Murray seems to have won the war, however. Leftists do now occasionally mention the inverse correlation between lower social class and IQ. Rather than say that low IQ causes poverty, however, they try to prove that poverty causes low IQ. I dealt with the latest such attempt a couple of weeks ago.
There was another attempt in that direction back in 2013 that I commented on at the time. It claimed that poverty was very stressful and that the stress prevented your brain from working properly. I would have thought that middle-class careerists were under the greatest stress but let's leave that for the moment. The title of the article was "Poverty Impedes Cognitive Function". There is a journalistic rendering of the claim here.
I put the findings in context at the time, showing that the conclusions did not follow from the reported evidence. I was not aware, however, that Jelte Wicherts also looked at the study around the same time. Now that J.P. Rushton is deceased, Wicherts is probably the man who knows the research on IQ better than anyone else. And he is fair. If someone puts up a celebratory claim about IQ, Wicherts will look at that critically, and if someone puts up claims that disrespect IQ Wicherts will look at that critically too. So I have a very favourable impression of Dr. Wicherts.
I have now come across his criticism of the 2013 study and it does not disappoint. I reproduce the abstract below:
"Mani et al. (Research Articles, 30 August, p. 976) presented laboratory experiments that aimed to show that poverty-related worries impede cognitive functioning. A reanalysis without dichotomization of income fails to corroborate their findings and highlights spurious interactions between income and experimental manipulation due to ceiling effects caused by short and easy tests. This suggests that effects of financial worries are not limited to the poor"
Posted by jonjayray at 8:34 AM
Friday, January 8, 2016
Having a big family makes your children either badly behaved or low achievers at school, study claims
The academic article underlying the popular report below is "The Quantity-Quality Trade-off and the Formation of Cognitive and Non-cognitive Skills". In my usual pesky way, I have had a look at it
I don't have the time to look in great detail at this quite complex study so I will content myself with a couple of basic observations. For a start, the finding is unusual. Other studies have found no effect of family size.
The problem, if there is one, appears to be an artifact of social class, though the authors are not allowed to mention that naughty word, of course. The effect seems mostly found among the poor, who are also less bright and who are also more likely to have big families.
And here's the tricky bit: Mention of poverty in America immediately calls to mind the naughtiest word of all: race. Did the authors control for race? Would the effect drop to insignificance if you looked at whites only? The Abstract of their current paper does not mention that word. But here's the kicker. There is also online what appears to be a preprint of the paper. And that DOES mention the word. And they DID find that race had a big effect. The effect of family size was primarily seen among persons of sub-Saharan African ancestry ("blacks" in non-academic language).
And the effects overall were not large. The word "IQ" is another word that may not be mentioned in polite circles. It is too easily understood. But their statistics can be translated into IQ. And the result is that we are looking at only about an IQ change of 1.5 IQ points. So the whole thing hardly matters anyway.
The takeaway is that in most families parents can have as many children as they like without concern about dumbing their kids down
Finally: I don't like to do this but I feel that I must place this study in the context of the current uproar in psychology about the high rate of unreplicable results and the associated topic of research ethics. It is now clear that many scientists do not tell the full truth about their research results -- for various reasons.
In that context, any concealment of findings calls into question the integrity of the research and the integrity of its authors. And since scientific communication depends heavily on trust, any attempt at concealment of findings -- as we see in the published abstract of this study -- strongly suggests that the work was not honestly reported and should therefore be disregarded. I am not being cynical in saying that the abstract IS the article for most readers of academic journal articles. Only specialists in that field plough through the whole thing
In the circumstances it is open for one to conclude that the real findings concerned blacks only but that was too unpalatable ("politically incorrect") to publish. So that problem was "worked around" in one or more ways
A new study has found that for every additional child born, the others are more likely to suffer poor cognitive abilities and behavioural problems afterwards.
Boys were more likely to misbehave while girls saw their performance in maths and reading skills dip.
Using data from 1986 to 2012 taken by the National Longitudinal Survey of Youth (NLSY) and by the Children and Young Adult Survey, three economists analysed how older siblings performed before and after a younger sibling was born.
They looked at the number and timing of births into a family and matched these to various mental and behavioural traits.
Levels of parental engagement were also crucial - with factors like how often families eat meals together, one-on-one time with each child, affection and the safety of the home also affecting how a child performed.
As families got bigger, the time spent with each child reduced, which has been linked to worse outcomes for children, they found.
'Our fixed effect estimates indicate that the arrival of a younger sibling reduces measures of parental investment as well as cognitive and non-cognitive outcomes of older children by approximately one-tenth of a standard deviation,' the research paper said.
The study was conducted by economists Chinhui Juhn, Yona Rubinstein, and C. Andrew Zuppann, who questioned whether the 'quantity' of children would effect the 'quality' of their upbringing.
It discovered that parental investment in older kids fell by 3 percentile points after a young child is born, while cognitive scores fell by 2.8 percentile points and behavioural problems increased.
'We have documented a significant trade-off between quantity and quality of children for NLSY mothers and their children. 'On average, children in larger families have lowered parental investment and worse cognitive and non-cognitive outcomes. '
Other factors found to influence the outcomes was the mother's intelligence and economic well-being.
Mothers were asked to take the Armed Force Qualification Test (AFQT), used by the military to assess skills including reading and reasoning.
Those who scored badly saw a larger drop in cognitive scores when they had their second child.
Posted by jonjayray at 5:18 PM
Thursday, January 7, 2016
Does poverty reduce your IQ?
Only in America, allegedly. That poverty CORRELATES with low IQ has long been known. Poor people are often that way because they make dumb decisions -- spending all their money on beer, drugs and cigarettes, for instance. A much more interesting question, therefore is whether poverty CAUSES low IQ. We know that low IQ causes poverty but does poverty cause low IQ?
A recent very extensive and very sophisticated study set out to examine that -- and the results have been reported enthusiastically -- as showing that poverty DOES have an effect on IQ. I reproduce a popular report of it below.
I have however, in my usual pesky way, gone back to the underlying academic journal article and read it. I have even looked at the numbers! Despite its great methodological care and statistical complexity, it is an amusing example of failing to do something that the best journals now recommend: Pre-register your expectations. Studies that do not do that are very prone to data dredging effects -- looking for any correlation in the data that seems large and changing your hypothesis to say that's what you expected all along.
And the authors below did not pre-register their expectations. They data-dredged. After all the hard work they did in gathering and analysing their data, they initially found NO EFFECT of poverty on IQ. So they desperately looked at their data to see what was in fact going on. And they found that if they used U.S. data only, there was a weak effect in the direction expected.
Findings that were not pre-registered can of course still be accepted and there are long-standing procedures to allow for data dredging -- adopting an experiment-wise error-rate approach, for instance. That is however very rarely done in fact. It would take all the fun out of a lot of research. But some approach to allowing for that sort of thing is now being given emphasis in journal review policies. In simple words, a much stronger effect is required for an unplanned relationship to be taken seriously. A weak relationship could be just a random oscillation.
The effects reported below were however very slight so I think that by current academic standards we should accept the null hypothesis. We should conclude that poverty does NOT demonstrably affect IQ.
I don't like to flog a dead horse but a second defect in the study is that the findings were not controlled for race. Could race alone account for the aberrant U.S. results? Knowing as we do how atypical are the IQs of persons with sub-Saharan African ancestry ("blacks", to use non-academic language) the researchers should clearly have excluded blacks from all analyses on the grounds that they are a quite separate population requiring study in their own right. The authors admit this but did not do it
The original study is rather misleadingly titled: "Large Cross-National Differences in Gene × Socioeconomic Status Interaction on Intelligence"
Poverty has long been linked with lower levels of intelligence, especially among children, but a new study has suggested its impact may depend on where you live.
Scientists believe a person's intelligence is formed by a complex interplay between the genes they inherit from their parents and the environment they grow up in.
But a study of twins has determined that childhood poverty appears to 'dampen down' the potential contained within a person's genes - and the situation varies from country to country.
The study, conducted by researchers at University of Texas at Austin and the University of Edinburgh in Scotland, found people born in the US tend to suffer the effects of poverty more.
Elsewhere, the link between poverty and a lower IQ was less noticeable in Western Europe and Australia, and in fact the opposite may be true in the Netherlands.
The study, which is published in the journal Psychological Science, analysed the findings of 14 peer-reviewed papers.
Combined, they drew upon almost 25,000 sets of twins and siblings from the US, Australia, England, Sweden, Germany and the Netherlands.
The researchers said the differences between the US and European countries may be due in part to more universal access to healthcare, which has helped to close some socioeconomic gaps.
Differences in the education systems in the countries may also play a role.
The researchers behind the study added that the results could prove useful in helping to tackle gaps between socioeconomic groups.
They said that providing more uniform access to education and healthcare can counter and even reverse the negative effect of poverty on genes involved in IQ.
Posted by jonjayray at 8:35 AM