Thursday, May 03, 2012

Jensen on g and genius

I found the excerpt below in the comments here.

Mega Questions for Renowned Psychologist Dr._Arthur R. Jensen - Interview by Christopher Michael Langan and Dr. Gina LoSasso and members of the Mega Foundation, Mega Society East and Ultranet 
Question #1: 
Christopher Langan for the Mega Foundation: It is reported that one of this century’s greatest physicists, Nobelist Richard Feynman, had an IQ of 125 or so. Yet, a careful reading of his work reveals amazing powers of concentration and analysis…powers of thought far in excess of those suggested by a z score of well under two standard deviations above the population mean. Could this be evidence that something might be wrong with the way intelligence is tested? Could it mean that early crystallization of intelligence, or specialization of intelligence in a specific set of (sub-g) factors – i.e., a narrow investment of g based on a lopsided combination of opportunity and proclivity - might put it beyond the reach of g-loaded tests weak in those specific factors, leading to deceptive results?

Arthur Jensen: I don’t take anecdotal report of the IQs of famous persons at all seriously. They are often fictitious and are used to make a point - typically a put-down of IQ test and the whole idea that individual differences in intelligence can be ranked or measured. James Watson once claimed an IQ of 115; the daughter of another very famous Nobelist claimed that her father would absolutely “flunk” any IQ test. It’s all ridiculous. 
Furthermore, the outstanding feature of any famous and accomplished person, especially a reputed genius, such as Feynman, is never their level of g (or their IQ), but some special talent and some other traits (e.g., zeal, persistence). Outstanding achievements(s) depend on these other qualities besides high intelligence. The special talents, such as mathematical musical, artistic, literary, or any other of the various “multiple intelligences” that have been mentioned by Howard Gardner and others are more salient in the achievements of geniuses than is their typically high level of g. Most very high-IQ people, of course, are not recognized as geniuses, because they haven’t any very outstanding creative achievements to their credit. However, there is a threshold property of IQ, or g, below which few if any individuals are even able to develop high-level complex talents or become known for socially significant intellectual or artistic achievements. This bare minimum threshold is probably somewhere between about +1.5 sigma and +2 sigma from the population mean on highly g-loaded tests. 
Childhood IQs that are at least above this threshold can also be misleading. There are two famous scientific geniuses, both Nobelists in physics, whose childhood IQs are very well authenticated to have been in the mid-130s. They are on record and were tested by none other than Lewis Terman himself, in his search for subjects in his well-known study of gifted children with IQs of 140 or above on the Stanford-Binet intelligence test. Although these two boys were brought to Terman’s attention because they were mathematical prodigies, they failed by a few IQ points to meet the one and only criterion (IQ > 139) for inclusion in Terman’s study. Although Terman was impressed by them, as a good scientist he had to exclude them from his sample of high-IQ kids. Yet none of the 1,500+ subjects in the study ever won a Nobel Prize or has a biography in the Encyclopedia Britannica as these two fellows did. Not only were they gifted mathematically, they had a combination of other traits without which they probably would not have become generally recognized as scientific and inventive geniuses. So-called intelligence tests, or IQ, are not intended to assess these special abilities unrelated to IQ or any other traits involved in outstanding achievement. It would be undesirable for IQ tests to attempt to do so, as it would be undesirable for a clinical thermometer to measure not just temperature but some combination of temperature, blood count, metabolic rate, etc. A good IQ test attempts to estimate the g factor, which isn’t a mixture, but a distillate of the one factor (i.e., a unitary source of individual differences variance) that is common to all cognitive tests, however diverse. 
I have had personal encounters with three Nobelists in science, including Feynman, who attended a lecture I gave at Cal Tech and later discussed it with me. He, like the other two Nobelists I’ve known (Francis Crick and William Shockley), not only came across as extremely sharp, especially in mathematical reasoning, but they were also rather obsessive about making sure they thoroughly understood the topic under immediate discussion. They at times transformed my verbal statements into graphical or mathematical forms and relationships. Two of these men knew each other very well and often discussed problems with each other. Each thought the other was very smart. I got a chance to test one of these Nobelists with Terman’s Concept Mastery Test, which was developed to test the Terman gifted group as adults, and he obtained an exceptionally high score even compared to the Terman group all with IQ > 139 and a mean of 152. 
I have written an essay relevant to this whole question: “Giftedness and genius: Crucial differences.” In C. P. Benbow & D. Lubinski (Eds.) Intellectual Talent: Psychometric and Social Issues, pp. 393-411. Baltimore: Johns Hopkins University Press.
See here for data relevant to this topic and the discussion in the comments.

146 comments:

Kevin Rose said...

Perfect. So the sub-factors of intelligence, like mathematical or verbal, etc, are what is decisive for top level achievement, and g is merely necessary up to a fairly basic point.  Makes total sense, and fits the evidence much better than all the blather about the centrality of g we have been hearing up till now.

I am kind of laughing though, this is what I have been saying all along. Of course now that Jensen says it, it will suddenly be "credible". In reality, anyone with a bit of confidence in his own intelligence could easily have seen that the way everyone was talking about how g is everything was all wrong. But so few people have confidence in what they clearly see with their own eyes, if it contradicts the experts.

steve hsu said...

> g is merely necessary up to a fairly basic point <

I think it's more complicated than that. For example, perhaps someone who is +2 has a 1% chance of somehow developing into a "genius", whereas someone who is +3 has a 10% chance, etc. Jensen himself is not very mathematical -- he expresses himself in the excerpt rather imprecisely. Note his point about the Nobelist he tested disfavors the idea that the probability of developing into a "genius" is independent of g once the basic threshold is reached. If that were the case almost all of the identified "geniuses" would have IQ pretty close to the threshold since there are so many more individuals there than further out on the tail. If you search around on the blog for SMPY or Roe data you'll see evidence against this. 

Pincher said...

I agree your inverted pyramid of potential genius is more precise when dealing with the hard sciences, but Jensen spoke of genius in a broader way.  What level g would one need to write Ulysses or Beethoven's Ode to Joy

Outside of the hard sciences, math, and games like chess, is one really helped in the creation of a notable achievement by being, say, three SDs above the mean instead of just one-and-a-half or two?

steve hsu said...

> Outside of the hard sciences, math ...

In these areas I'd guess the evidence is pretty strong that returns to g continue to be positive above +2 threshold. But in other areas I just don't know.

Kevin Rose said...

If that were the case almost all of the identified "geniuses" would
have IQ pretty close to the threshold since there are so many more
individuals there than further out on the tail.

Not necessarily.  An extremely high score on only one or two sub-factors might yield a high overall IQ, with a score for g only somewhere near the threshold Jensen mentions. I think you might be confusing g with IQ score, something which bedevils discussions on this issue.

The questions is do most geniuses come from the tail end of g, not from the tail end of IQ, and I don't know if there is any evidence for that at all. Nor do I know if there is evidence that most geniuses come even from the tail end of IQ. As you yourself note, there are fewer total numbers of people at the tail end, so the increase in probability would have to be pretty large to offset the massive decrease in absolute numbers. And lots of the -admittedly incomplete - evidence in this excerpt of yours militates against the idea that the gap in probability is huge.

Finally, at best this would show that the precise importance to assign to g is more than my dismissive minimum-threshold one, but the point that the sub-factors often play a decisive role stands.

Yan Shen said...

When a famous giant of science supposedly scored low on an IQ test, it has typically be a verbally loaded one. Shockley and Alvarez are amongst the most famous examples, both having failed to meet the threshold for further tracking on the verbally-loaded Terman exam.

 http://www.scientificamerican.com/article.cfm?id=recognizing-spatial-intel

"Ninety years ago, Stanford psychologist Lewis Terman began an
ambitious search for the brightest kids in California, administering IQ
tests to several thousand of children across the state. Those scoring
above an IQ of 135 (approximately the top 1 percent of scores) were
tracked for further study. There were two young boys, Luis Alvarez and
William Shockley, who were among the many who took Terman’s tests but
missed the cutoff score. Despite their exclusion from a study of young
“geniuses,” both went on to study physics, earn PhDs, and win the Nobel
prize.


How could these two minds, both with great potential for scientific
innovation, slip under the radar of IQ tests? One explanation is
that many items on Terman’s Stanford-Binet IQ test, as with many modern
assessments, fail to tap into a cognitive ability known as spatial
ability. Recent research on cognitive abilities is reinforcing what some
psychologists suggested decades ago: spatial ability, also known as
spatial visualization, plays a critical role in engineering and
scientific disciplines. Yet more verbally-loaded IQ tests, as well as
many popular standardized tests used today, do not adequately measure
this trait, especially in those who are most gifted with it."

http://etd.library.vanderbilt.edu/available/etd-07202009-203601/unrestricted/Wai.Disseration.07.20.2009.FINAL.pdf

"Stanley went beyond IQ or general intelligence to examine specific abilities (i.e., verbal and mathematical), which was an important advance in talent identification. Indeed Terman had missed two Nobel Laureates, William Shockley and Luis Alvarez, by utilizing only the highly verbal Stanford-Binet (Shurkin, 1992); had these two mathematically talented individuals also been given a quantitative reasoning measure they most likely would not have been missed."

Pincher said...

"Not necessarily.  An extremely high score on only one or two sub-factors might yield a high overall IQ, with a score for g only somewhere near the threshold Jensen mentions. I think you might be confusing g with IQ score, something which bedevils discussions on this issue."

When Jensen talks about IQ (and IQ scores) in the above passage, he's talking about g.  He's just being colloquial in using the terms interchangeably, which is appropriate for an interview.  Jensen doesn't care about an IQ score except to the degree it measures g.

I also think you're too focused on the subtests, which is a partial but critical distortion of Jensen's point.  He focuses on characteristics, talents, and interests which bring g into full creative play and allow it to blossom into something we call genius.  It's true that two of the "talents" Jensen mentions (mathematical and literary) can be tested with more or less precision, but in context it's clear that Jensen wasn't generally talking about things that can be measured on an IQ subtest.

BlackRoseML said...

 Everyone would agree that the returns to g are monotonic in any intellectual domain, but my main concern is the marginal returns of g in different domains. It would seem that g matters little above SD in most fields, except physics, philosophy, and literature. As for myself, I have fairly high verbal intelligence, but it is obvious that people like Saint Augustine and Dante have more verbal ability (estimated >760V GRE based on their works) than me. 

Kevin Rose said...

I think this is about as clear as it gets. He quite clearly says that specific aptitudes for math or other fields rather than g are more important in determining genius than g.

Furthermore, the outstanding feature of any famous and
accomplished person, especially a reputed genius, such as Feynman, is
never their level of g (or their IQ), but some special talent and some
other traits (e.g., zeal, persistence). Outstanding achievements(s)
depend on these other qualities besides high intelligence. The special
talents, such as mathematical musical, artistic, literary, or any other of
the various “multiple intelligences” that have been mentioned by Howard
Gardner and others are more salient in the achievements of geniuses than
is their typically high level of g. Most very high-IQ people, of course,
are not recognized as geniuses, because they haven’t any very outstanding
creative achievements to their credit. However, there is a threshold
property of IQ, or g, below which few if any individuals are even able to
develop high-level complex talents or become known for socially
significant intellectual or artistic achievements. This bare minimum
threshold is probably somewhere between about +1.5 sigma and +2 sigma from
the population mean on highly g-loaded tests.

Everything else that both you and Steve are saying is just 1) Reading into it things you wish to see 2) Speculation

I am always amazed at how people try so hard to avoid simple clarity and precision when discussing IQ. When discussing IQ, everyone loves fudging, conflating dissimilar things (like g and IQ), and generally speaking in the vaguest terms possible (g itself is an extremely vague hypothese with barely any meaning)

David Coughlin said...

That excerpt argues for high-g as a necessity but not for sufficiency for genius.  I don't recall that Steve has ever argued more than that.  Though, from these comments, you would guess that Steve is interested in parameterizing the likelihood of genius by g.

Pincher said...

Kevin,

You quote, but do not read with any comprehension what you quote.

The first characteristics Jensen mentions are "zeal" and "persistence".  Can those traits be measured by an IQ subtest?

Jensen then mentions the "talents" or "multiple intelligences" (mathematical, musical, artistic, literary, etc.) made famous by Howard Gardner.  Pray tell, Kevin, what IQ subtest accurately measures musical or artistic ability?

No?  Then what was the point of you saying in your first post, "...the sub-factors of intelligence, like mathematical or verbal, etc, are what is decisive for top level achievement..." that you later claim might skew an IQ test score?

Given the importance you put on precision and clarity, I'm very interested in hearing your explanation.

BlackRoseML said...

 I think this is about as clear as it gets. He quite clearly says that
specific aptitudes for math or other fields rather than g are more
important in determining genius than g.


But g is the main reason why most people cannot be geniuses as their is a g threshold that precludes most people from attaining genius, since by statistical necessity, most people cannot possess 1.5-2.0 SD g. In the general population, g plays a predominant role in determine how much ability one has in a given intellectual domain, but, perhaps, for those possessing higher g, they would be less constrained by g, as abilities become less intercorrelated and more differentiated (Spearman's law of diminishing returns).

Pincher said...

"As for myself, I have fairly high verbal intelligence, but it is obvious that people like Saint Augustine, John Milton, and Dante have more verbal ability (estimated >760V GRE based on their works) than me."

I'm not sure I agree that the potential for literary genius can be captured by hairsplitting high GRE scores.  I'm sure Edmund Wilson could have scored much higher on the verbal section of a standardized test than his friend F Scott Fitzgerald had he taken one, but he couldn't write a decent line of prose to save his life.  His friend, on the other hand, produced a small gem of literature that will endure as long as people read.  

When going over the literary giants of the last century, it's difficult to see any correlation between literary genius and verbal test scores would sort out on the higher end.  I can think of a few writers like Joyce, Eliot, Nabokov, and Bellow who might have done quite well on the verbal section of a standardized test score, but I can also think of quite a few famous literary figures (Hemingway, Faulkner, Dreiser, Woolf, Conrad, etc.) who probably would not have scored impressively at all. 

Your three examples are also problematic because they are such extreme anachronisms.  All three lived well before the age of universal literacy, and two didn't write in English.  It's hard enough to guess how someone like Hemingway would have scored on a standardized test, but how is one to seriously speculate how Dante would have scored on the GRE?  At least Hemingway was alive when the tests were being given out.  He might have even taken one when he was in the U.S. military during WW1.

Kevin Rose said...

Pincher, the FIRST thing Jensen is mentioned is talent, followed by zeal. His discussion then focuses on talents. So your very fist comment is off to a very poor start.

Of the rest of what Jensen mentions, some are talents corresponding to IQ sub-factors, some are not. That's simply because not all genius - like musical aptitude - is a function of something measured by IQ tests. I would have thought if obvious that when I mentioned that sub-factors of intelligence are the determining factors, that was only in the particular fields measured by those sub-factors. But if you feel that a clarification is necessary, I have no problem amending my statement.

What Jensen is saying is that achievement in fields measured by IQ sub-factors, scores on those sub-factors are more important than  g in determining genius-level accomplishment. Obviously there are fields for which we have no precise measures, like music, and for which g is a very poor predictor of top level achievement.

Is that clearer?

Kevin Rose said...

Yes, if there is a minimum threshold for g that most people fall beneath, then that alone explains why most people will not be geniuses. But +1.5 g is not really all that high.

The point I was have always made and that Jensen now seems to be endorsing is simply that the g, while important, is very far from being the central factor people often take it to be in top level achievement. Often people will act as if extremely high g alone is sufficient to make a genius. But that never seemed to me consistent with the facts. Sure, a high-ish general intelligence is important, but then specific intellectual aptitudes, like verbal, mathematical, or spatial, which are not a function of g, will make or un-make a genius.

Look at Jews, for instance, who have a verbal of 122 and a spatial of around 100 I believe. Their general intelligence will be reasonably high, but they will have specific aptitude for certain things. Asians, on the other hand, have a spatial of 110 and a verbal a bit under 100, I believe. Their general intelligence is reasonably high, but they will clearly have specific aptitudes that will make them much better at some things. Focusing on g alone and conflating it with "intelligence" is hugely misleading and will make one misunderstand the nature of the world.

Kevin Rose said...

But the important novelty here is that up till now, it was g+ some nebulous personality trait. Now it seems that specific aptitudes - often sub-factors of IQ tests - are the determining factors in genius. This is the first I am hearing of anyone saying that specific aptitude are more important than the "general aptitude" represented by g. That IS novel. Up till now the attitude was that g is God.

Steve is speculating about the parameters here of likelihood. It might be a 1-10 percent range, or it might be 1-2 percent difference, which would yield a very different picture.

Pincher said...

Kevin,

"Pincher, the FIRST thing Jensen is mentioned is talent, followed by zeal. His discussion then focuses on talents."

I wasn't trying to be comprehensive in my chronology.  The "special talents" (i.e., multiple intelligences) Jensen mentioned were parenthetically listed after he mentioned "zeal" and "persistence."  I followed suit.

"Of the rest of what Jensen mentions, some are talents corresponding to IQ sub-factors, some are not. That's simply because not all genius - like musical aptitude - is a function of something measured by IQ tests. I would have thought if obvious that when I mentioned that sub-factors of intelligence are the determining factors, that was only in the particular fields measured by those sub-factors. But if you feel that a clarification is necessary, I have no problem amending my statement."

Okay. Glad we're on the same page on this.

"What Jensen is saying is that achievement in fields measured by IQ sub-factors, scores on those sub-factors are more important than  g in determining genius-level accomplishment. Obviously there are fields for which we have no precise measures, like music, and for which g is a very poor predictor of top level achievement."

Jensen never talks about how scores in one sub-category might affect a person's overall IQ score.  That's entirely your inference.  And given the context of Jensen's quote (which mentions several items that don't affect a person's IQ score), I don't think it's a valid interpretation.  

When Jensen mentions Howard Gardner, that's a dead giveaway Jensen is not talking about g or IQ scores.  Gardner's theory doesn't require but the weakest of correlations between the different intelligences.

BlackRoseML said...

 Steve's interest in intelligence leads him to believe that, regarding general intelligence, the more the better, at least, in the attainment of intellectual excellence and accomplishment scientific fields. Steve doesn't not believe in a sufficiency threshold (or one at placed at a "mundane" level, around 2-3.5 SD) where any additional g would fail to yield significant benefits in the pursuit of excellence. The hegemony of g applies to > 99.9% of the population, according to him. 

Stephen Zhang said...

'Outstanding achievements(s) depend on these other qualities besides high intelligence.'
What are these interesting qualities? Are these qualities teachable?

'He (Feynman), like the other two Nobelists I’ve known (Francis Crick and William Shockley), not only came across as extremely sharp, especially in mathematical reasoning, but they were also rather obsessive about making sure they thoroughly understood the topic under immediate discussion. They at times transformed my verbal statements into graphical or mathematical forms and relationships.' This tendency to discuss seems to me a learnable trait. Do you agree?

tractal said...

 I don't recall anyone directly saying G=everything, except maybe for some comments about high end theoretical physics which Mr. Jensen is not really in a position to dispute. There's obviously a lot else going into genius level creativity, especially when we are talking about artistic creativity. But the thing is, we don't really know how to measure that other stuff very well. Maybe Beethoven was +2 SD and loaded with composer alleles and a killer work ethic, but so what? What can we really go out and do with that information? Treating G as very important makes sense for a lot of domains because we are actually able to measure it. That said I agree with your point: there are problems with trying to understand top level performance as a simple manifestation of G.

Kevin Rose said...

There's obviously a lot else going into genius level creativity,
especially when we are talking about artistic creativity. But the thing
is, we don't really know how to measure that other stuff very well.

But that's just it, for the first time that I can recall someone is saying that genius-level achievement is a function of ability, and one that we can in many cases measure, like math or verbal. Up till now it was precisely stuff we "could not measure well", like personality or outspokenness, which Asians are supposed to lack, etc. While Jensen includes a bit of that mumbo-jumbo here with his talk of zeal, etc, this is the first instance I can recall when he specifically makes genius a function of talent, and often ones that we can measure, like math and verbal.

The most common explanation of genius up till now was g +some vague personality attribute, like boldness or whatever. Now we are seeing genius might be a mere ability after all? That is new. (yes, I know Jensen includes appeals to vague personality attributes as well, but his focus here seems mostly on ability)

It has always seemed to me that the principle of parsimony required us to first see if we could not explain genius in terms of intelligence alone before appealing to things like personality. I am glad to see that kind of thinking showing up elsewhere.

Kevin Rose said...

Jensen never talks about how scores in one sub-category might affect a person's overall IQ score.

Huh? I am a bit lost.

Jensen is talking about aptitude for math or verbal ability (among other aptitudes), and how this is more important to genius in these domains than general intelligence, or g.

That sounds very much like saying that for those mental aptitudes for which there exist IQ sub-tests, your scores on them are more important than g - so long as your g passes a certain minimum point - in determining if you will be a genius in that specific domain.

No?

Kevin Rose said...

Yet according to Jensen it can be as low as 1.5 SD. I wonder what grounds Steve has for his belief beyond speculation.

BlackRoseML said...

 Perhaps the g threshold is merely the level where Spearman's law of diminishing returns starts to exert huge effects on specific aptitudes and g's influence weakens. 

tractal said...

Sure, but you should be really cautious. Just because Jensen says its "ability" doesn't mean it is. His only evidence for genius being some kind of trait is that he can't measure it well on IQ tests. There's nothing really here which says genius is not just boldness or work ethic etc. Of course it almost certainly is more than that, but all we have here is Jensen speculating.

Kevin Rose said...

I agree with Pincher below. The mistake I think comes from this belief that IQ and other standardized tests really capture everything there is about human intelligence. I personally think that there are entire dimensions of human intelligence that are simply not captured by tests as we now have them. IQ tests capture something important, but not everything, not by a long shot. And I am talking pure mental functioning here that is not being captured. Just look at the world and the mean IQs of various countries and how much it fails to explain about their respective performances, etc. And before one turns to personality to explain anomalies, consider the principle of parsimony. Anyways. 

This is the perennial temptation of scientism - to believe that what cannot be measured does not exist. It's not as if IQ tests are complete, or infallible!

Pincher said...

"That sounds very much like saying that for those mental aptitudes for which there exist IQ sub-tests, your scores on them are more important than g - so long as your g passes a certain minimum point - in determining if you will be a genius in that specific domain."  No?

In everything but math, the answer would be no.  

And since a math sub-test on a general IQ test is likely to be too general to tease out the dedicated overachiever from the actual genius with special talent, I'm guessing even that sub-test is likely to be insufficient for the task.  And that's your best chance.

An SAT test is an IQ test, for example.  Do you think you can tell the math geniuses from the merely very bright students in math by just looking at their math scores on the test?

Jensen is talking about very specialized talents that don't show up well in standardized tests. 

Kevin Rose said...

Agreed. We don't know. But the principle of parsimony requires us to first exhaust the possibility that genius is purely a function of ability before turning to combinations of personality+ability.

I am just pleased to see for the fist time that I can think of someone important saying that genius might not be just g+personality, and to downplay g, even if it is just speculation.

BlackRoseML said...

And since a math sub-test on a general IQ test is likely to be too
general to tease out the dedicated overachiever from the actual genius
with special talent, I'm guessing even that sub-test is likely to be
insufficient for the task.  And that's your best chanc


I think Hsu knows this, but he would attribute it to a low ceiling, perhaps due to the limitations of a math test based on geometry, algebra, and in the case of the GRE, rudimentary statistics and probability. That's why the cutoff for his IQ studies is 800Q/M on the GRE/SAT.

Kevin Rose said...

The issue is not what we can measure, but the role Jensen has in mind for specific aptitudes vs general intelligence. It does not matter if we cannot measure these specific aptitudes. This is the first I have heard that specific aptitudes for math or verbal might be more important in producing a genius than general intelligence.

It has always been obvious to me that the contribution of g to high level performance in any specific field is overshadowed by the contribution of specific aptitudes for that field. It explains so much more about the world. I am glad to finally see this idea getting more attention.

Yan Shen said...

I think you raise a good point. Take the examples of Shockley and Alvarez, or the anecdotes related to Feynman or Einstein. Probably all were significantly skewed towards M and S and away from V. Or for that matter re-read what I posted before about Terence Tao.

http://www.davidsongifted.org/db/Articles_id_10116.aspx

"There's no doubt that Terry Tao reasons almost
incredibly well, mathematically, and learns mathematics and related
subjects astonishingly fast. His performance in mathematics competitions
in Australia and on the mathematical portion of the College Board
Scholastic Aptitude Test (SAT-M) at age 8 is phenomenal. He was taking
the 60-item 60-minute multiple-choice SAT-M for the first time. On it,
only 1 percent of college-bound male 12th-graders in the United States
score 750 or more (College Board, 1985). He scored 760. Only one other
8-year-old of whom I am aware has done as well. That boy, who lives in a
suburb of Chicago, was taking the test for the fifth time! He managed
to score 800 before becoming 10 years old. Terry was not retested on
SAT-M at age 9, because that seemed unnecessary.

Yet at age 8 years 10 months, when he took both the
SAT-M and the SAT-Verbal, Terry scored only 290 on the latter. Just 9%
of college-bound male 12th-graders score 290 or less on SAT-V; a chance
score is about 230. The discrepancy between being 10 points above the
minimum 99th percentile on M and at the 9th percentile on V represents a
gap of about 3.7 standard deviations. Clearly, Terry did far better
with the mathematical reasoning items (please see the Appendix for
examples) than he did reading paragraphs and answering comprehension
questions about them or figuring out antonyms, verbal analogies, or
sentences with missing words.

Was the "lowness" of the verbal score (excellent
for one his age, of course) due to his lack of motivation on that part
of the test and/or surprise at its content? A year later, while this
altogether charming boy was spending four days at my home during early
May of 1985, I administered another form of the SAT-V to him under the
best possible conditions. His score rose to 380, which is the 31st
percentile. That's a fine gain, but the M vs. V discrepancy was probably
as great as before. Quite likely, on the SAT score scale his ability
had risen appreciably above the 800 ceiling of SAT-M.

How could Terry possibly learn mathematics and physical and computer
sciences so well with only 290-380V development? We of the Study of
Mathematically Precocious Youth (SMPY) at Johns Hopkins have discovered,
chiefly by testing able 12-year-olds, that when the examinee's SAT-M
score vastly exceeds his or her SAT-V score the youth is almost certain
to score high on a difficult test of nonverbal reasoning ability such as
the Advanced Form of the Raven Progressive Matrices, often higher than a
high-M high-V examinee does. To test this out, on 6 May 1985 I
administered to Terry the RPM-Advanced, an untimed test. He completed
its 36 8-option items in about 45 minutes. Whereas the average British
university student scores 21, Terry scored 32. He did not miss any of
the last, most difficult, 4 items. Also, when told which 4 items he had
not answered correctly, he was quickly able to find the correct response
to each. Few of SMPY's ablest protégés, members of its "700-800 on
SAT-M Before Age 13" group, could do as well."

Most likely people who are high at M and S have a tendency to skew towards fields such as mathematics, computer science, engineering, and the physical sciences, i.e. physics and chemistry and people who excel at V skew more towards the social sciences, humanities, and the life sciences.

Yan Shen said...

"While Jensen includes a bit of that mumbo-jumbo here with his talk of
zeal, etc, this is the first instance I can recall when he specifically
makes genius a function of talent, and often ones that we can measure, like math and verbal."

"I personally think that there are entire dimensions of human
intelligence that are simply not captured by tests as we now have them.
IQ tests capture something important, but not everything, not by a long
shot."

Aren't these two statements more or less contradictory?

Yan Shen said...

By the way Kevin, I read this complaint of yours directed towards Half Sigma on his blog and I agree that you've been treated somewhat unfairly in the blogosphere. As you complain to Half Sigma...

"Sigh. I
explicitly acknowledge that the average Asian is smarter than the
average white, but seriously, cant we discuss this like intelligent
adults? What difference does it make what my supposed motivations are?
Maybe your unspoken imputation is right, maybe I am just some white
nationalist who is just jealous that Asians are so smart - what
difference does it make to my arguments?"

Pincher said...

Kevin,

"It does not matter if we cannot measure these specific aptitudes."

Of course it matters.  If we can't measure something, then it might as well not exist.  We are left with only our speculations.  How would we know if our guesses are right if we can't measure them?  

Steve Hsu has at least given us a tangible and potentially testable model for thinking about this question using g. His probabilities might be off, and he might be entirely wrong, in which case it's back to the drawing board, but he's still given us something concrete to work with.

It's easy to claim that a math genius needs to be damn good at a math subtest to qualify as a math genius.  This is, of course, true (because it's circular) and yet not very illuminating. A math genius might also need a critical element that is not directly tested in many IQ subtests for math -- working memory, for example -- but would be tested in more comprehensive IQ tests for g

It's also possible that literary genius needs something other than a facility with words to reach its full potential.  Think of the complex structure in some novels, for example, that still manage to cohere.  The mind that can write Lolita is not just a wordsmith, but a careful builder of elaborate plots.

Yan Shen said...

 Regarding musical ability and perfect pitch, is it possible that people of Chinese extraction have an advantage, possibly genetic? Which way does the arrow of causation run?

http://ucsdnews.ucsd.edu/newsrel/soc/DeutschTone.asp

"Could it be that
cellist Yo-Yo Ma owes his perfect musical pitch to his Chinese
parents? While we may never know the definitive answer, new
research from the University of California, San Diego has found
a strong link between speaking a tone language – such
as Mandarin – and having perfect pitch, the ability once
thought to be the rare province of super-talented musicians.

The first large-scale,
direct-test study to be conducted on perfect pitch, led by psychology
professor Diana Deutsch of UC San Diego, has found that native
tone language speakers are almost nine times more likely to
have the ability. 

Perfect, or absolute,
pitch is the ability to name or produce a musical note of particular
pitch without the benefit of a reference note. The visual equivalent
is calling a red apple “red.” While most people
do this effortlessly, without, for example, having to compare
a red to a green apple, perfect pitch is extremely rare in the
U.S. and Europe, with an estimated prevalence in the general
population of less than one in 10,000.

Tone languages –
Mandarin and Vietnamese, among many others – are those
in which words take on entirely different meanings depending
on the tones in which they are enunciated. In Mandarin, for
example, the word “ma” means “mother”
when spoken in the first tone, “hemp” when spoken
in the second tone, “horse” in the third and a reproach
in the fourth. (Tone is not to be confused with shades of meaning
imparted by intonation; saying something sarcastically, for
instance, or rising at the end of a sentence to indicate a question.)

Deutsch and her co-authors
measured the prevalence of perfect pitch by means of a direct,
on-site test in two populations of music students: a group of
88 first-year students enrolled at the prestigious Central Conservatory
of Music in Beijing, China, all of whom spoke Mandarin, and
a group of 115 first-years at the Eastman School of Music in
Rochester, New York, none of whom spoke a tone language.

“We found a very
clear difference between the two populations,” Deutsch
said. “In Mandarin speakers, perfect pitch appears to
be not rare, but rather a readily acquired ability."

Kevin Rose said...

Of course it matters.  If we can't measure something, then it might as well not exist.

Ah, the follies of scientism!

You seem to be saying things that are increasingly remote from the discussion - that a math genius might need working memory (although why then would it not be captured on the math sub-test)? That writing a complex novel might need more than a mere facility with words (though the verbal sub-test measures complex verbal reasoning, not facility with words)? That what is needed for being a genius in math IS something captured on g (isn't the whole point that it is NOT something captured by g)?

Your main sticking point seems to be my insistence that Jensen was talking about the sub-factors of IQ. But I already backed away from that insistence in an extreme sense. OK, although it seems that at least some of what Jensen is talking about corresponds directly to the sub-factors, some does not. The point is not important to what I am trying to say.

For the first time specific abilities (not personality traits) other than just general intelligence are considered more important in producing a genius. Up till now all I had ever heard was that a genius is a product of a very high g with some very minor role played by specific aptitudes, and the rest is all personality. It was always g is God and the rest is all a bold willingness to speak up for oneself, blah blah blah (thus sweeping under the rug all the shy reitiring geniuses that litter the history books)

Get it?

(Incidentally, the less of a role g plays in high level ability in math or any other kind of cognition, the less coherence the concept of a general intelligence has. Which is why I say g is a barely coherent concept anyways.)

So fine, these abilities Jensen talks about do not correspond in straightforward ways to the IQ sub-factors. OK. So they cannot be measured. OK. For the first time that I am aware of, in principle, specific ability is considered more important than general intelligence in top level accomplishment in any field.

Of course, mere recognition of this fact undermines the very notion of a general intelligence. I predict the next evolution in the crude field of IQ testing will be a gradual moving away from the importance of the concept of a general intelligence.

Kevin Rose said...

If you can afford it, I seriously recommend that you seek the services of a trained therapist.

Kevin Rose said...

Are you aware that no one was talking about Asians or how they compare to whites on this thread except you? You are pretty much the first - and the only - person to even bring up that subject. Most people here seem genuinely interested in understanding something about our world.

A piece of advice my young friend; life is more than just Asians Are Just The Bestest People Ever. Go outside. Get some sunshine. Take a walk in the park. Get a girlfriend.

I am sorry for whatever defeats and humiliations whites have inflicted on you, but they are not me. Time to let it go. Life is more than that.

Asians are cool. We like you guys. You don't have to constantly prove yourself to us.

Yan Shen said...

You are truly a strange and hilarious individual Kevin.

I've been following your comments on this blog and also on sites like Half Sigma. You, along with people like Tractal and Sineruse, persistently whine about Asians, so it was pretty obvious where you were coming from. Given that this is Infoproc, probably most of the others commenters such as Steve Hsu, David Budd, etc also know that this is what you're trying to drive at.

Yan Shen said...

This advice is coming from a guy who whines about East Asians obsessively on HBD web sites right? And who after being criticized by Half Sigma complained that...

""Sigh. I explicitly acknowledge that the average Asian is smarter than
the average white, but seriously, cant we discuss this like intelligent
adults? What difference does it make what my supposed motivations are?
Maybe your unspoken imputation is right, maybe I am just some white
nationalist who is just jealous that Asians are so smart - what
difference does it make to my arguments?"

And later one was referred to as a nutcase by another Half Sigma reader.

"Kevin sounds like a nutcase. No one accused you of being a white
nationalist, no one is trying to "silence" you. But you seem to think
that is the case.

Posted by: Dextrology | February 26, 2012 at 10:22 PM"

Pincher said...

"Ah, the follies of scientism!"

I didn't say something we don't measure doesn't exist. I said it might as well not exist.  We are left only to guess what defines and makes it.

 "That what is needed for being a genius in math IS something captured on g (isn't the whole point that it is NOT something captured by g)?"

I don't think we have a method to define, capture, and predict creative genius at the top echelons of talent. All we have is something to point us in the right general direction for the hard sciences and math.  That something is called g.  Hsu has provided the framework for how to think about it with some precision.  I don't think the same method works as well outside of those narrow fields. I don't think we can use IQ to winnow down the field to pick the next great doctor or novelist or musician or artist or social scientist except in the broadest of terms.  (In general, it's better to be noticeably above the mean. After that, it's more of a crapshoot.) These areas are less hierarchical than the hard sciences in sorting out the top talent.Would Paul Krugman score higher on an IQ test than your average state college economics professor?  Certainly.  But it's not at all clear to me that he would score higher than most of the people he attended graduate school with at MIT.  (And if he did, he would probably gain his advantage in the V, which seems counterintuitive when one thinks of how mathematical modern-day economics has become.) Yet today he has the Nobel Prize and many intellectual accomplishments, and Arnold Kling has a blog. Even in the hard sciences, there is probably a good deal of superficial randomness to the results given the fact that most smart scientists at any level of IQ never create anything substantial enough that will define them as a "genius". 

Yan Shen said...

I've never seen a white nationalist as insecure as your friend Kevin, Pincher. He has this incredibly bizarre habit of going off into random outbursts about how East Asian Americans all hate whites because they feel humiliated by them. And if you point out to him that most East Asian Americans really don't care that much about whites, or for that matter about blacks and Hispanics, and that many of them are more or less focused on their own life and are cognizant of their status as a market dominant minority, Kevin just becomes angrier and starts declaring more emphatically that East Asian Americans all hate whites because they feel humiliated by them.

I linked to an article in the WSJ from way back in 2005 about how the students at Monta Vista High believed that if you were Asian you had to confirm that you were smart, but if you were white you had to prove it. That in particular seemed to set him off. I think that as a white nationalist the idea that East Asian Americans feel a sense of intellectual superiority over white Americans absolutely drives him insane.

David Coughlin said...

 It's interesting to me when people talk about things being 'teachable', as opposed to 'trainable'.  The former is the encumbent responsibility of one person to relate to many, the latter of one person to relate to one person [specifically, one's self].

Kevin Rose said...

I actually agree with you that we don't have a method to define and capture top level ability, really in any field. It seems to me that there is lots we don't yet know about human intelligence, and that IQ is a crude metric. It is useful in a broad way across large populations, but there are entire dimensions of cognition I think it fails to capture.

If you look at the achievement profile of various countries and their variable fertility in producing geniuses, it becomes clear just how inadequate IQ is as a metric. It fails to capture what may be the most interesting and important aspect of human mental functioning - genius.

Of course, IQ tests were always meant to be crude instruments with limited predictive value, and to be used with caution. The problem is that not many advances in psychometrics have taken place in the last century and the instrument remains quite crude, yet people have come to regard this instrument - through increased use, familiarity, and a fetishization of what can be measured and the consequent downgrading of what cannot be, a creeping illness in our culture - as almost infallible.

I consider psychometrics to be a science in a state of infancy at this point. There is so much much about the world its current results fail to explain (of course, there is much about the world it succeeds in explaining).

Pincher said...

Kevin Rose, (continuation from discussion below)

I think psychometrics is one of the best scientific tools we have in all of social science.  If only the rest of psychology could rise to such a "state of infancy".I also think you gravely misdiagnose what you call the "illness" of the public's belief in IQ.  With the exception of a few scholars, websites, and blogs, almost no one thinks seriously about IQ and very little in U.S. society is directly based on the science.  What little is based on it (standardized testing, for example) is highly controversial in many influential quarters.  The reigning orthodoxy in public affairs is human mental equality.Compare what John Q. Public hears about IQ to what he thinks he hears about, say, economics.  How much respect and influence do economists have when talking about the economy compared to psychometricians who talk about IQ?  Think of all the babble that is spoken every day about economics on TV -- most of it concerning silly unanswerable questions like predicting the next recession or the effect of an increase in marginal tax rates on the economy.  Now tell me when's the last time you even heard a psychometrician allowed to talk at length about his subject on TV?  If there is a social science we fetishize in the U.S., it's economics.

I think the proper comparison for the science of IQ is against the rest of social science, and it is there that psychometrics stands out for the quality of its results.  If you are going to use an absolute standard that uses g to predict everything about the human condition, well then of course psychometrics will fail that test.  But if you use my standard, then it looks pretty damn good.

Pincher said...

See my answer to you above.

David Coughlin said...

It would be fascinating to see the in-common genetics of the Vietnamese and Chinese.

Yan Shen said...

Tractal is another guy who whines hysterically on white nationalist blogs like Half Sigma about East Asians.

Yan Shen said...

 I don't know if you're utterly stupid or if you think everyone else here is utterly stupid. 1) You've been complaining about East Asians not only on HBD blogs, but also here. and 2) Pretty much everyone knows your real criticism of IQ testing is that it favors East Asians relative to whites. or 3) As you complained to Half Sigma

""Sigh. I explicitly acknowledge that the average Asian is smarter than
the average white, but seriously, cant we discuss this like intelligent
adults? What difference does it make what my supposed motivations are?
Maybe your unspoken imputation is right, maybe I am just some white
nationalist who is just jealous that Asians are so smart - what
difference does it make to my arguments?""

As long as you keep commenting on this blog, I'm going to keep pointing out the obvious fact that you're a pathetic joke.

tractal said...

I would respond, but I've already committed myself not to. And in any case I have a rule: When someone is just too damn dumb, ignore them. 

tractal said...

Interestingly, GRE-V seems to have high predictive power in English Phd track performance. I recall reading that in one of Roe's papers. Maybe something differentiates the geniuses from the Yale PhDs but that is a tiny population.

Pincher said...

"Interestingly, GRE-V seems to have high predictive power in English Phd track performance."

I'm unaware of the study, but having a PhD in literature from Yale doesn't make you a literary genius.  Is there a single Yale PhD who is as renowned a literary writer today as Jonathan Franzen (Swathmore, majored in German) or Michael Chabon (UC Irvine, majored in Creative Writing)? 

In fact, I would agree with the many critics outside of academe who think that the quality of literature has mostly suffered at the hands of the growing number of academics who have felt the need to produce something literary for the public.  

Unlike great scientists or mathematicians, great writers tend to pop up in unexpected places.  You never know where one will come from.  For the most part, literature programs produce professors, not writers.

Yan Shen said...

I also have a rule. When someone drones on and on about the importance of verbal ability and how important it is to know what an ad hominem is, but turns out not to actually understand what an ad hominem is, ignore them.

Yan Shen said...

Look, I'm not trying to start World War 3 here. I just want Tractal or Matthew Carnegie or Pincher or someone here to explain to me this white guy obsession with East Asians. I've seen it with Steve Sailer, with Sineruse, with Matthew Carnegie, with Tractal, with BenG, etc. How can it be that every other day there's a 1000 word entry on some HBD blog angrily railing against East Asians, with 50 million equally angry comments from whiny, insecure white guys. How is this even humanly possible?

And to the few sane people here like Steve Hsu and Chris Chang or David Budd and David Coughlin, is it reasonable or fair to expect East Asian Americans to just sit there and take this one sided bashing Shouldn't respect be a two way street? How can there be harmonious race relations in this country when one group is more or less having a field day bashing the other, while the other group more or less takes the high road and goes about its own business. Am I right to be annoyed and upset here? How did Infoproc turn into iSteve2.0, with hordes of Tractals, David Versaces, Matthew Carnegies, BenGs, Kevin Roses flooding the comments section time and time again railing hysterically against East Asian Americans?

I've pointed out before that this angry, obsessive East Asian bashing isn't just limited to the white nationalist/neo-Nazi blogosphere. It's even affected mainstream white America. Just look at this comment on the HuffingtonPost, which clearly has a left wing bias both in its reporting and its audience, over a story about how the majority of IMO competitors and Intel Science Talent Search semi-finalists were children of immigrants.

http://www.huffingtonpost.com/2011/05/24/immigrants-science_n_866321.html

"Well, I can tell u how it works.



My boy and girl attend/ed the elite govt selective school here in sydney
oz. 90% asian. They have no life outside coaching, school &
homework (set by the extended family too), minding shop."

Just the same shit over and over again. No respect. No sense of self-criticism or self-improvement. Just inane East Asian bashing over and over and over again. This is kind of garbage that's infected even mainstream white America. I've had first-hand experience with more than one white parent who has voiced the kinds of sentiments above.

What the hell is going on here? Can someone please let me know when the one-sided East Asian bashing that's par for the course in this country going to end?

Kevin Rose said...

I think psychometrics is one of the best scientific tools we have in all of social science.

Ah well, that's not saying much is it :)

The "illness" I was talking about was the tendency to downgrade anything that cannot be measured. The problem with this is that it might lead us to ignore some of the most important aspects of reality or our human nature simply because they cannot - yet -  be reliably quantified. It's a tendency that really picked up steam in the 20th century as physics began to be the model for all serious intellectual effort, and I think it might be impeding interesting discoveries as people fail to follow up lines of thought because they think they cannot accurately quantify their results, and they know that only that would win them any respect. While I am a huge fan of precision and clarity, I think one can present interesting findings about our world with clarity and precision even if we cannot quantify them.

Much of the talk about IQ and genius struck me as asinine because everyone acted as if what could not be measured did not exist, so that IQ must surely capture everything about human cognition, and genius needed to be explained by an appeal to something beyond intellectual ability, like personality. The tide seems to be turning on that.

I think you are right that the public gets treated to lots of "science" that is much fluffier than the science of IQ, but the experts who discuss IQ seem to me to be committing the intellectual error I just described, so that when you read their writings on human intelligence and IQ you are left with the strange sense of having read something that simply ignores glaring realities.

Yan Shen said...

 Kevin Rose complaining to Half Sigma on his blog...

"Sigh. I explicitly acknowledge that the average Asian is smarter than
the average white, but seriously, cant we discuss this like intelligent
adults? What difference does it make what my supposed motivations are?
Maybe your unspoken imputation is right, maybe I am just some white
nationalist who is just jealous that Asians are so smart - what
difference does it make to my arguments?"

Another commenter to Kevin Rose

"Kevin sounds like a nutcase. No one accused you of being a white
nationalist, no one is trying to "silence" you. But you seem to think
that is the case.


Posted by: Dextrology | February 26, 2012 at 10:22 PM"

Kevin Rose said...

Wise. The guy is having a very embarrassing public melt down.

Yan Shen said...

 Of course only Pincher would spend hours arguing with someone who says...

"While Jensen includes a bit of that mumbo-jumbo here with his talk
of zeal, etc, this is the first instance I can recall when he
specifically makes genius a function of talent, and often ones that we
can measure, like math and verbal."

And 2 minutes later says...



"I personally think that there are entire dimensions of human
intelligence that are simply not captured by tests as we now have them.
IQ tests capture something important, but not everything, not by a long
shot."

Yan Shen said...

Kevin Rose complaining to Half Sigma on his blog...



"Sigh. I explicitly acknowledge that the average Asian is smarter
than the average white, but seriously, cant we discuss this like
intelligent adults? What difference does it make what my supposed
motivations are? Maybe your unspoken imputation is right, maybe I am
just some white nationalist who is just jealous that Asians are so smart
- what difference does it make to my arguments?"



Another commenter to Kevin Rose



"Kevin sounds like a nutcase. No one accused you of being a white
nationalist, no one is trying to "silence" you. But you seem to think
that is the case.


Posted by: Dextrology | February 26, 2012 at 10:22 PM"

Somewhere in between those two comments in the same thread, Tractal whining hysterically about East Asians.
 

botti said...

How did this comments thread turn into a discussion about what people say on other blogs? Why not stick to discussing the post about genius?

Pincher said...

Kevin Rose,

What books, specifically, have you read about IQ?  Because I don't recognize the validity of your criticism in the scholars I've read.

"While I am a huge fan of precision and clarity, I think one can present interesting findings about our world with clarity and precision even if we cannot quantify them."

Like what?  Could you give me a couple of scientific examples?

Scientists measure things in order to avoid that very same lack of precision you claim is important.  The g factor, for example, is a result of psychometricians noticing a correlation in student test scores, trying to measure that correlation precisely, and then checking to see if that result had both consistency and empirical validity in other areas.  The discovery that it did was a major finding in social science. It's still widely unappreciated today.  The general public believes IQ just refers to book smarts, and many social scientists avoid using IQ as a control or variable in their empirical work. 

Does IQ explain everything or even most things about the human mind?  No.  But then no scholar or expert I know believes it does.  The science of IQ is still underutilized.  And like all good science that's underutilized, it needs promoters more than detractors.

sineruse said...

There is a lot of discussion of "M, V and S" on your blog.  Since you have read a lot of material on psychometry I was wondering whether you could help sharpen the discussion by affirming that there is no evidence for the existence of any neural "M" dimension orthogonal to the use of V/S/g/conscientiousness/etc capacities to acquire and develop mathematical skills.   For example, it is nonexistent on the WAIS which has only two principal components, "g" and V/S asymmetry and no evidence of additional dimensionality, and factor analyses of other tests such as Ravens APM and SPM either do not agree with each other as to what the factors are, or do not identify anything particulary mathematical as a factor, or do find some quasi-mathematical ability but do not include anything beyond basic number sense and primitive reckoning instinct as the rough interpretation of the factor (the "numerical" factor, when there is one).    Note that many animals have a counting/reckoning capacity -- rats can handle surprisingly large numbers approximately -- but do not have mathematics as such.

Matthew Carnegie said...

g is focused on and discussed rather than specific subfactors, as far as I can see because

1) it explains much more variance than the combined subfactors
2) IIRC within groups at least it has higher heritability than both raw IQ scores and subfactors
3) it is more interesting to try to raise (or understand how to raise or prevent decreasing) than subfactors because it would be better to raise all abilities rather than debate which tradeoffs to make amongst human qualities which
4) talking about it avoids the discussion of "how much role does culture and personality" play when people talk about specific culturally mediated and perhaps personality mediated abilities like mathematics or literature or art

But if you're looking for e.g. a mathematical or scientific high achiever, I would think (although I could be wrong), yes, it would be better to look for the child who is good at maths (when it actually is maths, rather than rote learning and calculation that small children do), rather than find a child with high g (does well on a test which has been highly correlated with the g factor produced by some IQ battery), unless there is huge swathe of the child population that just has no experience with maths at all, which is just not the case in most (almost all) of the world.

Yan Shen said...

 Every time Steve Hsu blogs about affirmative action, 50 angry white guys show up on Infoproc unprovoked. I think you are too smart Matthew to somehow focus on this one thread. You have been following this blog for quite a while now.

Can you explain to me why East Asians are such a popular topic in HBD blogs like iSteve or Half Sigma and why you, along with many other white guys, often whine hysterically about them in the comments section there?

Yan Shen said...

 "The blogs you mention seem more concerned about crime, politics and the media."

Whenever someone blogs about East Asians there, you end with like 5000 comments, many of the them angry and hysterical denunciations of East Asians.

Yan Shen said...

Matthew, I'd like to hear your thoughts about something. I argued before that I was disappointed by anti-East Asian bashing in the United States because I believed that if East Asians were ever to be academically out-competed in their own countries, the most likely response would be for many East Asians to work even harder than before. I stated that East Asian parents would criticize their children for being lazy and unmotivated and East Asian children would take that criticism to heart and look to improve themselves. I expressed frustration at the fact that most common white American response in the United States to East Asian academic over-performance has been simply to bash East Asians unprovoked.

I linked to something here recently that I believe sheds some light on this matter.

http://www.apa.org/monitor/feb03/overestimate.aspx

"Cross-cultural comparisons
Regardless of how
pervasive the phenomenon is, it is clear from Dunning's and others' work
that many Americans, at least sometimes and under some conditions, have
a tendency to inflate their worth. It is interesting, therefore, to see
the phenomenon's mirror opposite in another culture. In research
comparing North American and East Asian self-assessments, Heine of the
University of British Columbia finds that East Asians tend to
underestimate their abilities, with an aim toward improving the self and
getting along with others.

These differences are highlighted in a
meta-analysis Heine is now completing of 70 studies that examine the
degree of self-enhancement or self-criticism in China, Japan and Korea
versus the United States and Canada. Sixty-nine of the 70 studies reveal
significant differences between the two cultures in the degree to which
individuals hold these tendencies, he finds.

In another article in the October 2001 Journal of Personality and Social Psychology
(Vol. 81, No. 4), Heine's team looks more closely at how this occurs.
First, Japanese and American participants performed a task at which they
either succeeded or failed. Then they were timed as they worked on
another version of the task. "The results made a symmetrical X," says
Heine: Americans worked longer if they succeeded at the first task,
while Japanese worked longer if they failed.

There are cultural,
social and individual motives behind these tendencies, Heine and
colleagues observe in a paper in the October 1999 Psychological Review
(Vol. 106, No. 4). "As Western society becomes more individualistic, a
successful life has come to be equated with having high self-esteem,"
Heine says. "Inflating one's sense of self creates positive emotions and
feelings of self-efficacy, but the downside is that people don't really
like self-enhancers very much."

Conversely, East Asians'
self-improving or self-critical stance helps them maintain their "face,"
or reputation, and as a result, their interpersonal network. But the
cost is they don't feel as good about themselves, he says. Because
people in these cultures have different motivations, they make very
different choices, Heine adds. If Americans perceive they're not doing
well at something, they'll look for something else to do instead. "If
you're bad at volleyball, well fine, you won't play volleyball," as
Heine puts it. East Asians, though, view a poor performance as an
invitation to try harder."

Is this obsession with self-esteem why so many white Americans are so incredibly angry and rude towards East Asian Americans, and why most white Americans have no conception of working harder to improve themselves?

Kevin Rose said...

I have not read any full length book on the subject, but have read a ton of stuff on the web, papers, discussions, summaries, interviews with the big names, you name it. I do not claim to be any kind of expert, but I do claim that my reading has given me a good sense of the contours of the discussion, which is what I am talking about here. I have a pretty good sense of how experts talk about IQ on the web . Heck, even Steve who has read reams of stuff on the subject - and way more technical papers than I - discusses IQ in the terms that exemplify what I am criticizing.

Re ways in which an obsession with quantification can damage thinking -

Well, IQ is a good example of what I am talking about. Every discussion of human genius I have seen so far focuses on personality as the X factor. It can't be something about intelligence, because of course that is all taken care of by IQ. It's a number, after all, a measurement, so we must know all about intelligence. So the elusive X factor in genius must be about personality.

You see, it's a question of emphasis. Of course, no psychometrician will say outright that IQ captures everything about intelligence, but he will allow such an assumption to guide his thinking. How else to explain that the X factor of genius is invariably sought in personality if it is not that people believe the search for it in intelligence has been exhausted, which only makes sense if you think you know everything of importance there is to know about intelligence?

This is an egregious violation of the principle of parsimony which can only be explained by the well known fact that people will publicly disavow principles which they secretly allow to guide their thoughts.

It's not just IQ, either. All sorts of interesting things about the world - observations, insights - that cannot be quantified will be downgraded in importance if you believe that numbers are the only thing that matter. Sometime in the early 20th century, the precision of physics became the model for all human thought. Physics was prestigious not because it was more intellectually difficult - in some ways it was intellectually less difficult than writing a complex literary work or treatise on political science containing insights into human nature - but because it worked, it gave us power over the world.

It's instructive to read any social science book of today side by side with one from early in the 20th century  - the earlier book will be bristling with fascinating facts and observations about the human condition or a particular society, many of them highly questionable but highly suggestive, while the newer one will be bristling with forumals and equations but will say nothing really intereswting or perceptive, simply because most interesting things that can be said about the subject cannot be put into a formula. An entire dimension of thought has been lost. It is why modern SS books make for such listless reading.

Now obviously I am not against precision - rather the complete opposite. I am against the attempt to force the precision of mathematics onto subjects that cannot be forced int such a straightjacket - AND - I am against the tendency to ignore and not pursue those ideas that cannot beexpressed with the precision of mathematics.

Kevin Rose said...

This has to have been the most epic Yan Shen public melt down ever.

Guy_Brodude said...

I think Jensen's first point is extremely valid; Feynman in particular strikes me as somebody who wouldn't let facts get in the way of a good story.

Pincher said...

Kevin,

I don't think you get a good sense of the science of IQ from the web.  Too many unschooled partisans abound.  Discussions are easily hijacked and derailed. Distinctions are lost and never reclaimed.  You need to sit down and read through several volumes of books before you get a sense of it.  

"Heck, even Steve who has read reams of stuff on the subject - and way more technical papers than I - discusses IQ in the terms that exemplify what I am criticizing."

Some of this is explained by Hsu's career in the hard sciences, and some of it is probably because you misunderstand his points.

For obvious reasons, the proprietor of this blog is much more interested in how g relates to accomplishments in physics than, say, Nobel Prizes awarded in literature.  And I think he is on strong ground in believing the relationship between g and the hard sciences and math is quite strong well beyond 2 SDs, while at the same time being agnostic about the relationship between g and intellectual accomplishments outside of those areas.  That's a quite reasonable view, and it's quite contrary to your description of him.

"It's instructive to read any social science book of today side by side with one from early in the 20th century  - the earlier book will be bristling with fascinating facts and observations about the human condition or a particular society, many of them highly questionable but highly suggestive, while the newer one will be bristling with forumals and equations but will say nothing really intereswting or perceptive, simply because most interesting things that can be said about the subject cannot be put into a formula."

I think you romanticize the social science of the early twentieth century.  You think of literary masterpieces, but I think mainly of Freudian horseshit, Marxist tracts on the plight of the working man, and progressive fantasies that were all being peddled in the name of science.  That the science of IQ could be born in such an intellectual morass speaks well of how it has since developed.  It's like a poor kid born in the ghetto who has successfully strove for respectability.

Freud supposedly wrote beautiful German, but many social scientists were no better prose stylists than you typically find today. The journalist H.L. Mencken wrote a hilarious sendup of the Thorstein Veblen's prose, concluding that it was "a cent's worth of learning wrapped in a bale of polysyllables."

Don't hate reality because it often conforms to math.  That's neither reality nor math's fault.

"Now obviously I am not against precision - rather the complete opposite. I am against the attempt to force the precision of mathematics onto subjects that cannot be forced int such a straightjacket - AND - I am against the tendency to ignore and not pursue those ideas that cannot beexpressed with the precision of mathematics."

Yes, but how do you judge?  The answer should be obvious.  You test it.  If the numbers work empirically, then they are valid.  If not, then you move on.

You conflate Jensen's hedging to mean something that it's not.  All he's doing is telling you what he knows the numbers work for and what he knows they can't explain.  It's that simple.

Iamexpert said...

Why do people think there's something magical about the 2 SD level. If g matters in a field below this level, it matters just as much above it. We just don't notice because there are so few people significantly above the 2 SD level.

David Coughlin said...

 We already have a recent remake of the Three Stooges, Curly.

Iamexpert said...

Well no test could possibly measure all of intelligence, but if you can measure g, the common factor of all mental abilities, then that's an excellent approximation.

Yan Shen said...

Any thoughts David on whether or not I am correct in pointing out that white Americans often bash East Asian Americans unprovoked? Am I right or wrong about this. If I'm correct, do East Asian Americans have a right to be angry at the one sided bashing?

botti said...

***They have no life outside coaching***
 
Yes, well I can see why that kind of stereotyping and whining would grate. You get similar observations in NZ. There was a current affairs show a few years ago that looked at the academic success of asians in NZ. I remember they interviewed some asian students who suggested it was probably due to their culture, and hard work etc.. The impression I get in NZ is that this success is generally admired and given as an example for groups like Maori and polynesians when they (or white academics) complain they are failed by the european system.
 
And there are also going to be people who are envious. I haven't read Amy Chua's 'World on Fire' but Chinese obviously encounter some resentment in parts of South East Asia where they are a "market dominant minority". Also, near to NZ you see this with Fijian indians and the native fijians http://www.justpacific.com/fiji/fijicoup2000/indofijianspeaks.pdf. And in NZ with some polynesian and Maori attitudes towards the "privileged" Pakeha (NZ whites).  A agree it would be nice if people adopted the approach you mention of working harder to improve themselves, rather than complaining about others. 

http://en.wikipedia.org/wiki/World_on_Fire

tractal said...

 By incessant whining YS means that one time I posted on HS that 70% Asian American representation at a NYC elite college can't be explained entirely by G because of math.

Yan Shen said...

You're just one part of the problem, Tractal. When Steve Sailer and Half Sigma write a post about East Asians, you usually end up with 5000 comments. We can post some of them here if you'd like, to back up my assertion that many of these comments are whiny and hysterical in nature and extremely derogatory towards East Asians.

Iamexpert said...

JD Salinger had an IQ of 104. And that was back when IQ tests were easy (the Flynn Effect). So I disagree with Jensen that you need an IQ above the +1.5 SD level to contribute significantly to the arts. People tend to overestimate the amount of intelligence required (in all fields and all aspects of life)

Yan Shen said...

This blog is full of super smart people right? So I want someone to either agree with my assertion that there's an incredible amount of East Asian bashing going on in the United States that's more or less tolerated by mainstream white America or tell me that I'm just crazy, that I'm seeing something that doesn't actually exist.

Commenter RKU thinks that I'm too quarrelsome and that I should mellow out and not get so worked up over these matters. Is that the future of race relations in this country? Should we just accept one sided bashing of East Asian Americans by white Americans as the status quo, as something not worth getting worked up over? Or should we condemn this barbarity in the strongest possible terms purely as a matter of principle?

tractal said...

Sineruse, if there is no independent M factor, how can we make sense of SAT-M vs SAT-V predictive power? What, for instance, can we make of a High M, low V, low S profile? Are those kids just more trained? Seems sort of unlikely to me given how much predictive power SAT-M has. 

BlackRoseML said...

Because I am a 2 SD individual, and it seems that I have the raw intellectual ability (although no conscientiousness) to master any intellectual domain besides the aforementioned fields. ;)

Iamexpert said...

Well if you were 4 SD you wouldn't need conscientiousness. The point is there's no magic threshold beyond which IQ stops being relevant.

Iamexpert said...

This kind of anecdotal evidence proves little. Actual studies (see steve's latest post) show high IQ is increasingly advantageous up to even the 1 in 10,000 level and beyond.

BlackRoseML said...

 Look, I am simply denying the primacy of g at high abilities levels (due to Spearman's law of diminishing returns) while attributing achievement, success, attainment to specific abilities. Conscientiousness is also important too.

For instance, I read some of The City of God and I could say that I don't have Augustine's verbal ability. He has high g relative to modern whites, but it is his profoundly superior verbal ability (not g) that enables him to write such beautiful and abstract prose. 

Iamexpert said...

Spearman's law of diminishing returns appears to have been falsified.

As for augistine's verbal ability, special talents are important at all IQ levels, not above the +2 SD level only. jD Salinger had a 104 IQ but had an amazing verbal talent. Muhammad Ali had an IQ of 78 but was a kinesthetic super genius not to mention a charismatic pop cultural poet. There are always going to be those at all IQ levels with amazing wild talents, but for the vast majority of people, in the vast majority of situations, overall intelligence is what matters most.

Pincher said...

Salinger was a so-so writer.  He didn't have "amazing verbal talent."  He wrote one incredibly famous novel, which is today mainly read by teens and almost no one else, and some decent short stories.

I also question whether Salinger had a 104 IQ.  That sounds like one of those anecdotal reports of a famous person's IQ that Jensen wants us about taking seriously.

tractal said...

You are literally delusional man, and you are distracting the worthwhile discussions.

Iamexpert said...

A so-so writer? His book sold 65 million copies and was ranked by prestigious authoritative Time magazine as one of the 100 best English novels ever written between 1923 and 2005 and was voted one of the 100 best novels ever by the French public. His literary genius is established fact.

His IQ is also well established. His biography clearly states he was a B student with 104 IQ. Another biography I just found states he graduated from military academy where his IQ was recorded at 115. So we have two sources citing two different tests both confirming he had an IQ in the upper average range.

Yan Shen said...

 Oh okay. I am completely wrong about the fact that Asian bashing exists in the comments section of this blog, i.e. whenever Steve Hsu decides to write a post about Asian Americans and affirmative action, 50 angry white guys flood the comments section and start bashing away.)

Richard Seiter said...

Yan Shen, we don't know each other and I don't frequent many of the blogs you mention so I am probably missing much background, but here are my thoughts.  Please take this in the spirit it is meant--friendly feedback to someone I respect but who I think is behaving badly at the moment.

I have appreciated many of the insights you have offered in Infoproc.  In the comments for this blog post two that stand out for me are your posts on Terry Tao and on the prevalence of perfect pitch in ethnic Chinese and speakers of tonal languages.  Both of those posts taught me something.  Thank you.

I agree there is Asian (I won't subdivide because I don't think the average basher does) bashing in the United States.  In some circles I would agree that is tolerated where other ethnic bashing might not be.  I would disagree that it is as widespread as you indicate (that may be because I am less sensitive to it, a friend of mine in an interracial couple has opened my eyes a bit on this topic).  I think you are conflating anti-intellectualism and/or anti-work ethic with Asian bashing to some degree (the prevalence of both of the former in the US cause me much distress).

I agree with RKU that you are being quarrelsome.  I agree with you that you should not accept one sided bashing, but that is not what I see happening in this thread.  I see you generating most of the divisive white/asian labeling and commentary in this thread.  Based solely on reading this thread I think it would be more accurate to characterize your statements as "Asian supremacist" than those of your opponents as "white supremacist" (I don't think I agree with either characterization though).  I don't doubt that there might be history (both in other blogs and in past Infoproc threads) that lends support to your case.  I just have to ask if your responses in this thread have been proportionate and constructive (and for me the answer for too many of them is no) to the posts made in this thread.Thanks for reading.  For whatever it's worth I am a "white guy" who has a great appreciation for the intellectual abilities (and personal qualities) of Asians (going to MIT will do that ;-).  I look forward to your future contributions here.

Pincher said...

Salinger was a one-hit wonder.  He is known today for The Catcher in the Rye and nothing else.

I strongly suggest you don't follow Time magazine's literary rankings for judging the best writers.  Look instead to Salinger's peers.  The Catcher in the Rye wasn't a major literary accomplishment that caught their attention.  Most would have laughed at you if you told them that Salinger was a literary genius.  His is famous today for a book for kids that captured the zeitgeist of the fifties and sixties in a similar way that Jack Kerouac's book On the Road did for hippies.  (You'll notice that Kerouac's book, which is poorly written, is also on the Modern Library's best 100 list.)
Some books do that.  Another book from that same era that did it for the anti-war movement was Catch-22.  You should note that Joseph Heller, who was a better writer than Salinger, never wrote another decent book in his life.  These books capture a mood or a trend, not necessarily because of their literary skill, but because of their subject matter and timing.  Holden Caulfield became an icon for alienated teens and the youth rebellion just when teens were starting to question authority.  

One way you can measure true literary skill, instead of just fortuitous timing, is to note the consistency and range of  output of an author's high-quality work.  Salinger is remembered today for one book.  So is Heller.  So is Kerouac.  Saul Bellow, on the other hand, is remembered for half-a-dozen.  Robert Penn Warren is remembered as the only man to win a Pulitzer Prize in both the categories of poetry (twice) and fiction.  Warren and Bellow are true literary masters with "amazing verbal talent"; Salinger is just a drive-by with a book.  Neither Bellow nor Warren as as well-known today as Salinger, however, because their books are not usually assigned reading in high school.  You should also avoid going to the French public for their views on books written in English.  Novels are best judged in their own language.

I wouldn't use sales figures, either.  Otherwise, you might confuse Danielle Steel as a literary genius.

"His IQ is also well established."

If you don't know whether it's 104 or 115, then it's not well-established.  There's a nontrivial difference between those two figures.

Kevin Rose said...

By the way, for all you white people who are treating Yan Shen with kid gloves and trying to "sympathize" with him, you are displaying an incredibly odious brand of paternalistic "soft" racism that educated whites sometimes have when dealing with misbehaving minorities. If Yan Shen was a white guy he would simply be considered a lunatic and an idiot, and be told to go shove it. It is obvious to anyone that he is a bully and a spoiled brat, and the biggest racist on this site. But since he is Asian, we feel we have to show him "understanding" and "sympathy". But holding Asians to lower intellectual and emotional standards than whites is not doing them any favors. We should expect Yan Shen to behave like an adult and not say retarded things and throw tantrums and try to bully people and in general act like a spoiled brat, just as we would expect of any white person.

 

BlackRoseML said...

Advanced theology is high verbally loaded and intellectual demanding, but it not considered prestigious it fails to offer practical insights that allows humanity to understand and manipulate the natural world. As the Kevin Rose states:

Physics was prestigious not because it was more intellectually
difficult - in some ways it was intellectually less difficult than
writing a complex literary work or treatise on political science
containing insights into human nature - but because it worked, it gave us power over the world.

I could easily image that if Augustine were alive today, he would get an 800 CR and perhaps a high 600 low 700M on the SAT. (And he might be excluded from Terman's study based on his raw abilities.) He would then score a 178-180 on LSAT and waste his impressive verbal intellect and life working at Biglaw, although he'll make a respectable amount of money. Maybe Monica would successful persuade him to use his intellect for something better.

Again, Augustine's accomplishments are due to his verbal ability, not g. The examples of Walter Alvarez and William Shockley, two men excluded from Terman's study because they presumably lack enough g to do well on the verbal assessment of Terman's test, are a testimony to the power of specific abilities (in their case quantitative ability) rather g at high levels of accomplishments. 

Iamexpert said...

It sounds like your criteria for identifying literary genius is pretty arbitrary and thus unsatisfying. As for the 11 point discrepancy between Salinger's two IQ scores, that's pretty typical. Different IQ tests only correlate about 0.7 with one another so it's rare for anyone to get the same score every time they are tested. We can't measure intelligence with the precision with which we measure height and weight. It's too complex and consists of too many different parts. At best we can take a rough sample of the most representative (g loaded) aspects of the mind's intellectual repertoire.

BlackRoseML said...

You seem to conflate incorrectly, literary fame with literary genius. Salinger wasn't a genius. Would Jerry B Jenkins be considered a genius because his books are so popular?

Yan Shen said...

 "Based solely on reading this thread I think it would be more accurate to
characterize your statements as "Asian supremacist" than those of your
opponents as "white supremacist" (I don't think I agree with either
characterization though).  I don't doubt that there might be history
(both in other blogs and in past Infoproc threads) that lends support to
your case.  I just have to ask if your responses in this thread have
been proportionate and constructive (and for me the answer for too many
of them is no) to the posts made in this thread."

Thanks for the reply Richard. You're right that I'm the one who has forced this issue out into the open in this particular thread, precisely because there is an ugly history at Infoproc and elsewhere of one sided East Asian bashing. A previous thread on Asian Americans and affirmative action somehow turned into a discussion about the masculinity of East Asian males. Commenter Kevin Rose has a history of whining at white nationalist/neo-Nazi blogs about East Asians, as does Tractal. And Sineruse has a creepy and obsessive history that I don't even want to talk about. But that's only the beginning.

Part of the problem is that most of Steve Hsu's commenters are white. I've seen very few East Asians comment in this blog.  This tends to skew the balance of discussions. Unfortunately, it seems like more and more of his commenters are coming these days from blogs like iSteve, Half Sigma, etc.

You seem like a smart guy. Do you think my comments, which you've labelled as "Asian supremacist" in nature, might be motivated in part by the constant Asian bashing that goes on here? The moment Steve Hsu blogs about affirmative action, hordes of angry white nationalists start whining and bashing away. Do you see why the constant whining and East Asian bashing might annoy some of us? Can you see why eventually some of us might say something back? Do you see how hordes of angry white nationalists constantly bashing East Asians as grinds might eventually cause some of us to offer up alternative explanations, even though we might otherwise not say anything?

Pincher said...

I thought the typical spread in a person retaking an IQ test was about five points or a third of a standard deviation.  Your spread is more than two-thirds of a standard deviation. In any case, I don't doubt that Salinger was a man of modest intellectual caliber.  I just doubt we know the exact caliber.

As for Salinger's literary qualities, I can only restate that anyone who thinks he had "amazing verbal talent" because he wrote The Catcher in the Rye will have to invent an entire new dictionary of adjectives to praise the work of someone like Vladimir Nabokov or James Joyce. 

Pincher said...

I just don't think you can speak confidently of knowing a man's g or what his SAT scores would have been like when he lived and died before the fall of Rome.

Yan Shen said...

Richard, I have a couple of additional thoughts that I'd like to share with you. I'm extremely appreciative of your initial response, because I feel like you're the only person here who's taken the time to at least respond to some of my substantive points. I want to start off our dialogue by focusing on a particular comment made by a Huffington Post regular in response to a story about how the majority of Intel Science Talent Search Semi-Finalists and IMO competitors in the United States were the children of immigrants, primarily those from China and India. I believe that this comment more or less captures the essence of the kind of mean-spirited Asian bashing which I am referring to.

http://www.huffingtonpost.com/2011/05/24/immigrants-science_n_866321.html

"Well, I can tell u how it works.



My boy and girl attend/ed the elite govt selective school here in sydney
oz. 90% asian. They have no life outside coaching, school &
homework (set by the extended family too), minding shop.



My kids got there on ability, not coaching & rote learning. They have a huge social & sporting life.



Friends have employed these stars & found them useless outside the prescribed box.



Its just a gimmick. An old asian tradition (read on getting into
imperial bureaucracy - exams). The whole desperate family throws its
resources behind advancing the smartest among them, to gain influence
and pull the family up."

The commenter was from Australia, judging by her comments, but I've heard this kind of sentiment being voiced by many mainstream white Americans, parents and children alike. For years now, East Asian Americans have been put down in this kind of fashion. We've been subjected to crude and dehumanizing stereotypes and have been told that we are not deserving of the accomplishments which many of us have worked extremely hard for. Note that I am not talking about the white nationalist/neo-Nazi community. I am referring to mainstream white Americans. The fact that Huffington Post regulars would engage in this kind of behavior worries me, because I was under the impression that most white liberals considered themselves to be the paragon of moral virtue.

I just want to ask how I, as an Asian American, should respond to these kinds of comments. I can assure you that they are far more prevalent than you may believe, because I've heard these kinds of sentiments being voiced by far too many mainstream white Americans. I don't profess to be a saint Richard. I'm sure that I'm motivated by the same base impulses that drive other people to do and say things and behave in certain ways. In a sense, all of us are slaves to human nature. But I don't believe that I'm a bad person either, and I think you and others would be gravely mistaken to assume that somehow I'm just the East Asian equivalent of the white nationalists whom I often criticize. When Asian Americans are constantly being belittled and put down in the fashion above, don't you think that over time it might eventually anger some of us? Might not the constant denigration of everything that many of us have worked so hard to attain inflame the passions and cause some of us to eventually respond with what you might characterize as Asian supremacist sentiments? After all, all of us are human.

I've never heard an East Asian American parent go out of their way to denigrate the accomplishments of another black, Hispanic, or white American. I have heard many East Asian Americans parents be critical of their own children though. My mother was like that towards me growing up, and at times it put a strain in our relationship. But looking back on it, it helped to make me a better person overall. I express both frustration and confusion when I read comments like the one I linked to above, because the essence of this kind of mean-spirited bashing really is anathema to the Confucian culture I was raised under.

Iamexpert said...

When someone retakes the same IQ test there's a lot more consistency because reliability coefficients are so high, but when someone takes two different IQ tests, particularly two different IQ tests that emphasize different non-g components, there's a lot more spread. For example the average IQ at Harvard is 143 as measured by the SAT, but when tested on the WAIS or LSAT, Harvard students regress to an IQ of 130.

Iamexpert said...

I'm just looking for objective criteria for identifying literary Genius. I cited both popularity and critical praise, but apparently these are both the wrong yardsticks. My concern is that every time a Genius with an unspectacular IQ is identified, there are attempts to call either his reported IQ or his Genius into question. As a result, the notion that Genius requires very high IQ becomes unfalsifiable.

dwbudd said...

Yan, please take these comments for what they're worth - the opinion of one man.
I think you're right on in one respect: people in this country - and in my experience, it's hardly restricted to "white nationalists" (whatever you mean by that) - make denigrating comments about Asians (primarily, though not exclusively these days, ethnic Chinese) with an impunity that would not be extended them had they made similar remarks about Latinos, blacks, or certainly Jewish people.  I attribute this in part to a number of factors.  

One, as you've stated, there are not really many visible, vocal, activists.  Who is the Chinese version of Al Sharpton?  The purchasing power of Asian immigrants and Asian Americans is much larger than their demographic numbers, but the lack of willingness or interest to actually put some political muscle behind those numbers makes it easier for a lazy comedian, script writer, or politician to make a dismissive comment about Asians than he would dare make about other ethnic minorities.

Two, there is not, despite the actions of some fragmented people, a successful pan-Asian coalition.  In my own experience, part of this is due to the legacy of animosity amongst the ethnic groups themselves.  The Chinese resent and fear the Japanese, the Japanese think that Koreans and Chinese are dirty and ill-mannered.  All three would be offended to be grouped among the lower, southeast Asians (e.g., Vietnamese.)  And ALL look down on Filipinos.  Divide and rule is key here.

Third, well-to-do newcomers do not make sympathetic victims.  When ethnic activists complain about how Lucy Liu is portrayed on some brain-dead television sitcom, instead of focusing on racist college admissions policies, it looks a lot more like whining about a make-believe problem.  In my home state of California, WHERE were the Asian activists when Proposition 229 was being debated?  The weight of affirmative action fell disproportionately on Asians, and yet the only group that spoke up loudly about 227 was the ironic group "Chinese FOR Affirmative Action."  Talk about Stockholm Syndrome.  The data most people say are that Asians are over-represented in Berkeley, have incomes above the norm, are seen as living disproportionately in well-to-do suburbs (e.g., Cupertino, San Marino, Saratoga, etc.).  So, the response is "what are you complaining about?"  People like "Chinese for Affirmative Action" and other liberal activists groups provide the political cover necessary to block the reasonable response of "Yes, the above may be true, but what would the numbers look like if the playing field WERE actually level?"

One quarrel I would have with you is that you seem to conflate the real problem of bashing of Asian AMERICANS with criticisms of China.  Asian Americans, IMHO, are Americans, and owe no fealty to China.  The Chinese government is a repressive, autocratic regime.  They crush internal dissent, lend alms to evil regimes like the one in Iran, make partners with other malefactors in Africa, and manipulate the world market to their own game.  Pointing this out does not amount to "bashing Asian Americans" any more than pointing out the laziness and sloth of the current governments in Greece, Italy, Spain, and now, apparently, France is "bashing European Americans."

It's my hope that as Asian Americans become more fully integrated into the US, some of the immature, provincial defensiveness that accompanies that "your criticism of the Chinese government is an attack on me" will disappear.

BlackRoseML said...

I read his work; I know Augustine's V is exceptionally high, and this V would manifest on modern psychometric tests. Easily an 800CR on the SAT. I have little idea about his M or what he would score on the Raven's Advanced Progressive Matrices. If he scored 800 V 690 M for instance, might correspond to an IQ of 138, below the threshold of Terman's study.

BlackRoseML said...

When someone retakes the same IQ test there's a lot more consistency
because reliability coefficients are so high, but when someone takes two
different IQ tests, particularly two different IQ tests that emphasize
different non-g components, there's a lot more spread. For example the
average IQ at Harvard is 143 as measured by the SAT because they were
selected by the SAT, but when tested on the WAIS or LSAT, Harvard
students regress to an Average IQ of 130.


Some say that one reason why Harvard continues to select on SAT scores is not due to admission reasons, but to maintain the perception of the school as "elite". I wonder what is the marginal impact for 10 SAT CR+M points on admission chances for a given student with SAT scores in the high 1400s/low 1500s when applying to the Ivy League. What is the impact of getting 40 more points than one's "true" score on the SAT for admission into Harvard? It is possible that a large portion of Harvard students got lucky on the SAT and were admitted on the basis of their "inflated" scores.

Yan Shen said...

 "It's my hope that as Asian Americans become more fully integrated into
the US, some of the immature, provincial defensiveness that accompanies
the "your criticism of the Chinese government is an attack on me" will
disappear."

You misunderstand me completely David on the issue of China. I used to criticize Steve Hsu on this blog constantly for not being more critical of the Chinese government. See this comment of mine from an earlier post.

http://infoproc.blogspot.com/2010/06/shanghai-expo.html

" Whoever
designed the China pavilion, I hope the relevant authorities treat
him/her better than they did renowned Chinese artist and activist Ai
Weiwei, the person behind the design of the "Bird's Nest" at the Beijing
Olympics. Unfortunately, Ai Weiwei ran afoul of Chinese officials when
he persistently reported on the shoddy construction of the buildings
involved in the 2008 Sichuan earthquake. In 2009, he was beaten by
police officials in Chengdu while attempting to testify on behalf of Tan
Zuoren, a fellow activist who also investigated the circumstances which
exacerbated the extent of the damage inflicted by the Sichuan
earthquake. The beating that Ai Weiwei took was severe enough that he
had to undergo emergency brain surgery in order to stop the internal
bleeding, though he seems to have made a recovery since then. In any
case, it's extremely alarming to see officials cracking down on decent
human beings in such a barbaric fashion.



I ask the following question. Is there any other civilized nation
that has virtually zero regard for rule of law? I can't think of any
other civilized North American, European, or East Asian nation that
exhibits such a blatant disregard for so foundational a principle. Is
this problem unique to China, or is it more characteristic of developing
nations as a whole? For a nation that has such tremendous aspirations,
it's rather disheartening to see how in certain respects China remains
incredibly backwards. You find a similar disregard for rule of law in
some of the despotic regimes of Africa.



One positive point to focus on here is that even democratic nations
such as South Korea and Taiwan were ruled by authoritarian regimes which
suppressed dissent, until as recently as the mid 1980s when they
finally reached a certain level of socioeconomic development and
eventually democratized. Francis Fukuyama seems to believe that once
China eventually reaches a certain level of socioeconomic development,
it will face tremendous internal pressure to democratize. I hope that
his reasoning turns out to be correct. For now, the Chinese government
has been legitimated mostly on the basis of the enormous advances it has
brought about for its people in terms of social and economic
development, and on the basis of a resurgent Han nationalism. I hope
that the Shanghai Expo is a harbinger of better things to come, and that
the growing pains that China is enduring are in general more the
characteristic result of a developing nation, rather than something
uniquely Chinese."

Pincher said...

Did you read his work in the medieval Latin in which he originally wrote?  If not, then you are using a translation to judge someone's verbal ability, which means you have a better idea of the translator's V than St Augustine's.

Pincher said...

Iamexpert,

(I'm replying to your comment above here because I otherwise would not have the room to reply.)

>"I'm just looking for objective criteria for identifying literary Genius. I cited both popularity and critical praise, but apparently these are both the wrong yardsticks."

I gave you a couple of standards (praise from peers; consistency of high quality work).  Both those standards could be objectified with a little thought and legwork. 

"When someone retakes the same IQ test there's a lot more consistency because reliability coefficients are so high, but when someone takes two different IQ tests, particularly two different IQ tests that emphasize different non-g components..."

Do you have any evidence Salinger took two different IQ tests?

"For example the average IQ at Harvard is 143 as measured by the SAT because they were selected by the SAT, but when tested on the WAIS or LSAT, Harvard students regress to an Average IQ of 130."

That's easier to explain.  Harvard doesn't publicly report the SAT scores of AA and legacy admits, so the average of those admitted to Harvard on the quality of their academic work is quite high (143).  But the IQ of Harvard students in a random test is much lower (130).

dwbudd said...

Yan:

A couple of quick responses.

The first is, I may have misunderstood you vis-a-vis China.  I don't have the full history of your comments about China, and I take you at your word that you are as capable of criticism of China as the next guy; more in fact.  I was judging from your comment below where verbal attacks on Chinese Americans (of which I am sympathetic with you) were mentioned alongside attacks on China (which I am less inclined to be sympathetic to).  I am not sure what "mindless China bashing" is, exactly, but it's my firm belief that, as Americans, we who are immigrants from elsewhere need to identify with our country (in this case, the US of A) and not with the country of our parents or grandparents.  That's all, really.

The second - about whether democracy (and I presume you mean a liberal democracy of the sort in the US) is "right" for China, I would say, I don't know, and I don't really care.  Ultimately, what sort of government is "right" for China is up to the Chinese people; not you or me or Francis Fukuyama (the "end of history" has not worked out quite as he thought).  The government in Beijing have achieved some incredible successes in the past 30 years or so, lifting millions and millions of people out of unspeakable poverty.  They *seem*, as you say, quite competent at dealing with their internal affairs, and face problems that we do not.  Thus, it's easy to cast stones at them.  I don't know your age, but when I was younger, the British government sent envoys to China during the negotiations to settle what would happen with the territories of Hongkong granted in perpetuity by the 1842 and 1860 treaties (Nanking and  Peking) in 1997.  The British ambassador began to lecture his Chinese counterpart about "human rights," and was quickly silenced when the Chinese envoy pulled forth copies of agreements that established various unbelievably racist laws and edicts, and then asked, politely, if the British would like to explain how to square "human rights" with their own history.

But all I meant was that pointing out what some of us think are failings of China as a nation is not "bashing" China, or by extension, Asian Americans.  I don't mean to accuse you of such knee-jerk reactionary thought; it only came across that way to me in how you presented your argument.  

Yan Shen said...

 "But all I meant was that pointing out what some of us think are failings
of China as a nation is not "bashing" China, or by extension, Asian
Americans.  I don't mean to accuse you of such knee-jerk reactionary
thought; it only came across that way to me in how you presented your
argument."

I agree that many Chinese Americans are too sensitive about criticism of the Chinese government.

Iamexpert said...

Pincher, with respect to the praise of Salinger's literary peers, I remember Jensen once stated the opinions of judges could qualify as an objective measure of an ability if there was sufficient intercorrelation. What Jensen never clarified was the opinion of which judges? The general public judges Salinger's book to be a work of genius as do the even the elites at Time magazine, but his peers look down on him. So perhaps we need subcategories of literary genius. Maybe Salinger is a genius with respect to popular literature but not with respect to the high art literary tradition. Or perhaps he's not a Genius at all, but is considered one by default because the true Geniuses won't lower themselves to compete for a mass audience, but if they chose to, they could easily dispose him in the popular imagination.

I doubt Salinger took the same test twice but I haven't investigated fully.

With respect to Harvard's mean IQ, SAScores might be inflated by selective reporting by Harvard, but the most obvious explanation for why Harvard students score much lower on non-SAT testing is regression to the mean.

Pincher said...

"Pincher, with respect to the praise of Salinger's literary peers, I remember Jensen once stated the opinions of judges could qualify as an objective measure of an ability if there was sufficient intercorrelation."

I'm just defending an objective standard of peer review for judging literary quality.  I'm not saying that Salinger's peers will be able to deduce his verbal ability on a standardized test.

Remember that the original question that began this entire discussion was the degree to which literary genius or excellence could be predicted by g. So the assumption was that the two qualities were not necessarily the same.  It was possible that one could be a great writer but not score particularly well on a verbal test beyond a certain threshold.

My argument against Salinger is that he's not a literary genius and so it's not worth discussing his IQ score.

"With respect to Harvard's mean IQ, SAT cores might be inflated by selective reporting by Harvard, but the most obvious explanation for why Harvard students score much lower on non-SAT testing is regression to the mean."

I don't think you understand regression to the mean.  You'd better look it up.

Kevin Rose said...

I don't think you get a good sense of the science of IQ from the web.
 Too many unschooled partisans abound.  Discussions are easily hijacked
and derailed. Distinctions are lost and never reclaimed.  You need to
sit down and read through several volumes of books before you get a
sense of it.

That's a fair point. The interviews and quotes on the web that I have read from the likes of Jensen, Lynn, etc, do not inspire confidence that their books will be significantly different in quality from these discussion, but I should give them a shot, you're right. I gotta make a point of getting on that :)

I notice that most of the writers you cite as examples of sloppy thinking are German, and with a few exceptions I am actually not a big fan of German thinkers in anything outside of science, where they excel. Not only that, but your examples prove my point; when I said before that in the 20th century physics became the model for thought, that was only in the Anglo world. In Germany it happened much sooner with Kant and Hegel, whose convoluted and obscure prose were products of the belief that thinking in every field must achieve the precision and mimic the style of science ( I think it was Kant who famously said this in a very explicit way for the first time). Peter Watson suggests that academic thought in America today is primarily influenced by this German movement, which goes a long way towards explaining the fondness for obscure jargon in today's academia, which is such a striking departure from the elegance and clarity that characterized Anglo-French thinking for so long. Nietzsche famously said that wherever she extends, Germany corrupts culture.

I have to disagree with you that there is strong evidence to support the idea that more g even beyond 2 SD corelates to genius level accomplishment in the hard sciences. It's funny you should say this, because this entire post had as its basis Feynman, an example of someone with modest g but outsized achievement in the hard sciences. I don't see where you make this distinction between the hard sciences and other areas with regard to g, or at least where you find that in Jensens remarks, since many of the examples Jensen uses are about the hard sciences. The central example, Feynman, is about the hard sciences.

It seems to me that you are trying to twist what Jensen says into your belief in the importance of g, even when the entire point of this post by Steve is to show that g is less important for genius than often conceded. I think your strongest point against my position was that I was wrong in conflating the specific abilities Jensen cites as critical for genius with the sub-factors of an IQ test, a point I already backed away from, but which does not substantially alter what I was trying to say.

I think the interesting thing about what Jensen says is that g is less important than specific abilities to genius level accomplishment, and it seems you are trying very hard to get out of admitting that he is saying this. The fact that Steve thought it was worth making a post about this shows that he thought it was an interesting departure from usual ways of thinking about this and tends to disprove your contention that what Jensen is saying is merely reiterating something that is commonly said amongst those who discuss this issue.

Yes, but how do you judge when the numbers work and when they do not?
 The answer should be obvious.  You test them.  If the numbers work
empirically, then they are valid and should be used.  If not, then you
move on.

I agree with this, but I think that the way g is usually discussed shows a marked reluctance to "move on" from a concept that is showing increasingly limited returns in explanatory power.

Pincher said...

Kevin,

You don't like the early Germans (Freud, the disciples of Marx, etc.).  So what specific examples of early-twentieth-century books in social science, that you earlier said compares favorably to today's writing on social science in many ways, did you have in mind? 

"I have to disagree with you that there is strong evidence to support the idea that more g even beyond 2 SD corelates to genius level accomplishment in the hard sciences.  It's funny you should say this, because this entire post had as its basis Feynman, an example of someone with modest g but outsized achievement in the hard sciences. I don't see where you make this distinction between the hard sciences and other areas with regard to g, or at least where you find that in Jensens remarks, since many of the examples Jensen uses are about the hard sciences. The central example, Feynman, is about the hard sciences."

It's true that the example of Feynman's reportedly low IQ of 125 sparked Jensen's comments in the interview.  But you should note that Jensen was skeptical of that figure, even though it was at the threshold (1.5 to 2 SDs above the mean) Jensen later claimed was needed for a scientist to accomplish something that could be labeled "genius."  Why would he be skeptical of a number that was still at the bottom of his range?

In any case, three important points need to be made.  

First, while you are correct that the initial thrust of the interviewer's question was about a physicist (Feynman), and while Jensen responded with examples of the IQs of other physicists, Jensen also quickly expanded his remarks about genius to other fields besides science.

Second, Jensen is basing his case on his understanding of the numbers and what he believes is genius-level accomplishment.  That's fundamentally from your anti-IQ critique.

Third, it's an interview.  It's not a scholarly paper.  As Steve Hsu points out, there's a lack of precision in some of Jensen's remarks.  He's speaking off-the-cuff.  So I don't think it's appropriate to respond with an "Aha!" to something you see in his comments and run off on a slight tangent.  Science is not decided in what might be nothing more than throwaway remarks of a scientist during an interview.

Iamexpert said...

 I'm just defending an objective standard of peer review for judging
literary quality.  I'm not saying that Salinger's peers will be able to
deduce his verbal ability on a standardized test.


But you are saying they can objectively evaluate the quality of his literary output.  I'm saying it might not be that simple.

I don't think you understand regression to the mean.  You'd better look it up.

Perhaps you're the one who doesn't understand, since you are apparently baffled by the simple idea that a population that is selected by one psychometric measure, would regress to the mean on tests that were not used to select them. 

Pincher said...

Iamexpert (continued from discussion below),

"But you are saying they can objectively evaluate the quality of his literary output.  I'm saying it might not be that simple."

Any objective measure we use will be somewhat arbitrary.  But the original standards you put forward (sales information, the English literary tastes of French audiences, and journalistic or popular rankings, etc.) were obviously flawed.

"Perhaps you're the one who doesn't understand, since you are apparently baffled by the simple idea that a population that is selected by one psychometric measure, would regress to the mean on tests that were not used to select them."

Oh, I see what you're saying.  

The problem is that you assume the SATs are an overwhelming factor in admissions, which is the only way to assume such a wicked regression to the mean on what are still two highly-correlated IQ tests.  

Even apart from the legacy and AA admits  -- which are a huge percentage of the accepted applicants -- the standardized test scores are not a primary factor in admissions for those students competing on the basis of academic excellence.  Many applicants with excellent to perfect SAT scores are turned down by Harvard, and many applicants with less than glowing, but still quite high, SAT scores are admitted.  This should tell you that the admissions committee is using them more as a threshold for entry than as a strictly competitive device for admissions.

The number of students with perfect SAT math scores who are turned down by Harvard is greater than the number of students with perfect SAT math scores who are admitted.

Pincher said...

My reply to your post is in the comments section above.  

BlackRoseML said...

Perhaps the composite SAT is more prestigious than a mere math score, and at higher levels of M, there is less correlation with g, hence perfect SAT-M scores don't guarantee an SAT-V > 650. A mere 1450, a little below Harvard average, and that applicant would need some compensating extracurriculars (or some mathematical accomplishment beyond scoring a "mere" 800 on the SAT-M) since his composite SAT score suggests he is not brainaic.

sineruse said...

re: intertest reliability,

The usual suspects will be happy to know that, for the same reason that strivers tend to underperform their credentials, striver SAT scores have higher inter-test reliability.  If you observe an increasing function of (e.g.) Talent, Effort, NumberOfRetakes, and Luck/Noise, then a higher known or estimated level of Effort (and/or Number of Retakes or other elastic factors) means lower inferred levels of inelastic factors like Talent *and Noise* for any given outcome, all else being equal.   Preparation reduces the role of chance, and of ability, wealth, retakes, etc.

The reasons why Harvard students score lower than expected on IQ tests are more complex than the prior role of luck:

1. The tests are given in large (introductory) Psychology classes, which oversamples the lower-ability students.
2. And/or given to volunteers solicited based on advertisements in the building that houses the psychology and sociology departments.  Same problem.
3. Results for in-class tests will depend on time of day.
3. If a school selects strongly for SAT, any measure that isn't extremely similar to SAT will be overestimated by SAT in the selected population.
4. SAT/IQ correlation is broken at the upper tail, especially for the uber-conscientious population that gets into Harvard. H selects for traits that destroy the correlation.

Pincher said...

The point is, even if you get a perfect score on your SATs, you aren't guaranteed a spot at Harvard or any of the other better Ivy league schools.

So Iamexpert's assumption that the nearly one standard deviation between the Harvard mean SAT score (which is essentially an IQ test showing they test at 143) and another IQ test given to Harvard undergrads (which showed the average IQ to be 130) is driven by a regression to the mean is incorrect.

That difference in the two scores is almost certainly driven by the well-known fact that colleges lie about their students' SAT scores to remain competitive in the college rankings.  Something like 40% of Harvard's students are legacies and AA admits.  Does Harvard honestly report the SAT scores of those students or do they just report the scores of the other 60% of students who compete with each other, academically, to get in.  (And keep in mind that even among that 60%, Harvard uses a holistic approach to selecting its applicants.  It doesn't focus solely on SAT scores.)

If my hypothesis is right, then the major difference in the two tests' scores is driven by selection bias in reporting.  The SAT scores are Harvard's crème de la crème.  The other IQ test wasn't so selective.  It included legacies and AA admits.  That may explain the large difference in the scores.

tractal said...

I really don't understand the reasoning in the first half of this post. If we ascribed to each score some combination of ability, practice, and test day luck, then it does follow that more diligent students at a given score will have lower ability, but why do we also expect greater inter-test reliability? If test day luck really is luck, then it seems like it amounts to a random number no matter how much you study. You seem to be saying that because "study" accounts for a greater number of points in a high studying individual, therefore everything else accounts for fewer points for that individual. If we say that a score is just a composite of luck, ability, and practice, knowing that A has 20 practice points and B has 10 practice points can't tell us anything about how lucky A or B was on the exam, except for the fact that because A will on average score 10 points higher than B, and that therefore the "luck" component of his test will be on average smaller relative to his total score. But I don't think this is what you mean, because that reasoning holds just as well for high ability groups and is almost an artifact from the way the test is scored.

Beyond that effect, though, I can't see why we would expect higher test consistency with A than B. If I know a composite score and a practice score for A and B (and the test means), I can make inferences about which of them was probably more able or luckier on exam day. But that inference would be based entirely on a comparison with the test average. Knowing that A has more study points than B doesn't seem incompatible with also saying that for both students their score will be a composite of ability + study + a totally random number. Help me out here. I know its a really minor point, but I just find the logic interesting.

BlackRoseML said...

 Admission uses multiple metrics, which reduces the contribution of luck
(vs striving) to the deflation of admitted students' SAT scores by IQ
tests.


Are those other requires significance sources of variance among the applications? For instance, GPA is only considered to cull out slackers by deterring them from applying or immediately discarding the applications of low GPA students, but among the serious applicants, they have near immaculate GPAs, thus the variance of GPA among the applicants is quite low. Furthermore, GPA and SAT scores are modestly correlated, hence academic slackers with 1550+ and > 4.0s can exist among the applicant pool, but 4.0s with lower SAT scores are quite ubiquitous, perhaps at frequency 5-8 times the number of 4.0/1550+ students. (IIRC a 1400 is the 96th percentile and a 1550+ corresponds to the 99.5 percentile; the ratio of  1400-1540 to 1550+  is about 8 to 1, but if we assume some intercorrelation between GPA and SAT scores, the ratio between 1400-1540/4.0 to 1550+/4.0 is slightly lower since a 1400 has a lower chance of getting a 4.0 compared to a 1550+ as the former needs for conscientiousness, passion, or rote memory. (I looked at the SAT applicant data from a few colleges (not Ivy or high private level), and almost no one scores 2100+ and have a GPA below 3.0.) It is likely that the deterrence effect on potential applications with a respectable, not stratospheric SAT score (low 1400s) possessing a "low" GPA, and the naturally high GPAs of 1550+ would lower the GPA variance of the Ivy League applicant pool. Due to the restricted variance of GPA of Ivy League applicants, SAT scores (in the range of 1350-1600) would be a significant factor in one's probability for admission.

Iamexpert said...

 Any objective measure we use will be somewhat arbitrary.

Which is why there might be multiple objective measures for quantifying different kinds of literary genius.  Salinger may have been a genius at writing for a popular audience but a mediocrity when writing for other writers.  It's kind of like comparing the Beatles to Mozart.  The former is for the masses, the latter is for the elites, but both are brilliant in their own genre.

The problem is that you assume the SATs are an overwhelming factor in admissions,

I think they are.  Just the fact that the average SAT scores at these schools is so high proves it.  You seem to think they are just picking the students with the best grades and that students with the best grades are the highest SAT performers without Harvard having to directly select for them.  However SAT scores only moderately correlate with grades in the college applicant population so selecting for grades is not going to drive the mean SAT score that high, especially considering there is a pretty low ceiling on how high one's grades can be.

Even apart from the legacy and AA admits  -- which are a huge percentage of the accepted applicants


The average SAT scores of legacy and AA admits would have to be absurdly low to drag Harvard's overall IQ from 143 to 130.  The data does not confirm this.  First of all AA admits are a pretty small percentage of the total student body, and they probably average around IQ 135 on the SAT.  Meanwhile legacy admits score virtually the same as non-legacy admits.  It's a myth that legacies are mediocre.  Just the fact that they are legacies suggests they come from higher IQ families than non-legacies, and while children from high IQ families regress to the mean, it's the brightest of these children who get accepted at Harvard.

Pincher said...

"Which is why there might be multiple objective measures for quantifying different kinds of literary genius.  Salinger may have been a genius at writing for a popular audience but a mediocrity when writing for other writers.  It's kind of like comparing the Beatles to Mozart.  The former is for the masses, the latter is for the elites, but both are brilliant in their own genre."

Writing a symphony is galaxies more difficult than writing a three-minute ditty.  If you're going down this route, why not just call Richard Bach of Jonathan Livingston Seagull or Eric Carle of The Very Hungry Caterpillar literary geniuses?

I don't mind multiple objective measures to help narrow the field down, but one still must discriminate between difficult tasks performed well and easy tasks that are routinely performed by a wide range of cognitive abilities.  If your standard doesn't do that, then it's of no use in helping you select geniuses in a particular field.

"I think they are.  Just the fact that the average SAT scores at these schools is so high proves it."

No, it doesn't.  If you're on the admissions committee and you select a student with a 1540 combined V & M on his SAT score, but you're turning down four other students with the exact same score, then you're not solely discriminating on the basis of an IQ test and you therefore have no reason to expect such a severe regression to the mean on the next test they take. 

In fact, you probably shouldn't expect to see much of a regression to the mean at all because you are also looking to cross-reference that score with several other factors.

"The average SAT scores of legacy and AA admits would have to be absurdly low to drag Harvard's overall IQ from 143 to 130."

I would guess approximately 120.  The rest of the difference between the scores of the two IQ tests could be explained by a slight regression to the mean (in the manner you argue) and a selection bias towards the least intelligent students at Harvard joining the test that showed them to have a 130 IQ.

"First of all AA admits are a pretty small percentage of the total student body, and they probably average around IQ 135 on the SAT."

I'd like to see where you get your information.  Approximately 2100 freshman applicants are offered admissions to Harvard every year and 1600 of them decide to matriculate.

12% of those students are African-Americans (200 students), which given the upper range of their SAT scores and the many choices (Yale, Stanford, Princeton, etc.) smart black kids have in choosing to go to school, we can guess their SAT test scores average well below 1300.  Another 12% are Hispanic students (200 students).  Another 2% are Native-American and Pacific islanders (35 students).  I have less confidence in predicting the test scores of Hispanics and Native-Americans/Pacific Islanders, but I think we can pretty sure they are significantly lower than average for Harvard, even if higher than the scores of African-Americans at the school. 

According to Andrew Ferguson's Crazy U, a Harvard Dean admitted that 30% to 35% of Harvard's freshmen class would be legacies (480 students).  Since we can be pretty certain that there is very little overlap between legacies and AA admittees, that would mean approximately half of Harvard's 1600 incoming students are either legacy or AA admittees.

"Meanwhile legacy admits score virtually the same as non-legacy admits."

Then you must not believe in genetic regression to the mean.

" It's a myth that legacies are mediocre."

There's substantial space between "mediocre" and excellent enough to get into Harvard in an open and fair academic competition. 

Iamexpert said...

 I don't mind multiple objective measures to help narrow the field down,
but one still must discriminate between difficult tasks performed well
and easy tasks that are routinely performed by a wide range of cognitive
abilities.  If your standard doesn't do that, then it's of no use in
helping you select geniuses in a particular field.


Here's my point:  Salinger might not be able to write high art literature anywhere near as skillfully as the giants in that field, but the geniuses of high art literature probably are not able to write a book that appeals to the masses as successfully as Salinger did.  Salinger created a work of enormous cultural relevance that very few people have the literary talent to rival, even if it's a type of literary talent the hoity toity don't respect.

In fact, you probably shouldn't expect to see much of a regression to
the mean at all because you are also looking to cross-reference that
score with several other factors, all of which are correlated with high
IQ.


I agree that even if Harvard did not use SATs in their selection process, the average SAT score at Harvard would still be high, but nowhere near the IQ equivalent of 143.  Probably more like 120.  So that's probably the mean that Harvard students are regressing to, as opposed to the U.S. mean of 100.  However because we're talking about fairly homogenous populations, the correlations between different IQ tests becomes much smaller, so the degree of regression (to the higher mean) becomes much greater. 

I'd like to see where you get your information.

In the book "The Bell Curve" they report that black Harvard students only scored 95 points lower on the old SAT than white Harvard students.  It also reported that in 1990, the average student admitted to Harvard scored 697 on the
verbal SAT and 718 on the math section and legacies scored
674 on verbal and 695 on math.

Then you must not believe in genetic regression to the mean.

Well the parents of legacies are probably genetically smarter than their classmates at Harvard student (since they're typically rich an powerful) so that negates some of the regression in their kids.  And even though their average kid has regressed to the mean, it is their brightest kids that become legacies.

BlackRoseML said...

I think the legacies are in the top decile of intelligence; people have a reasonable chance of succeeding at that level of intelligence (and possessing self-confidence, ambition, and conscientiousness) with especially the unique connections and opportunities that Harvard affords.

Pincher said...

You seem stuck on Saint Augustine.  I'm not sure why.  You can't know someone's V from a translation.  Ancient theological insights couched in a literary language you are not familiar with might seem impressive only because you don't know the times and the language the author wrote in.  If Augustine uses a cliché of that period that seems verbally fresh to your modern ear, because you never heard it before, does that make him a literary genius?

"I think the legacies are in the top decile of intelligence; people have a reasonable chance of succeeding at that level of intelligence (and possessing self-confidence, ambition, and conscientiousness) with especially the unique connections and opportunities from a Harvard education that they can avail."

I brought up the legacies to point out that they are not strictly selected on the basis of their academic record, and so they skew the statistics somewhat.  A legacy with a 125 IQ is plenty smart and will likely graduate from school and succeed in life.  But he most likely could not had entered Harvard without the legacy boost. 

BlackRoseML said...

 There are over 1.4 million college freshman who enrolled in U.S. institutions in 2008.
 Some of those (less than 10%) are foreign students, but since Harvard
accepts foreign students, we'll assume that's a wash.  Approximately 30%
of Americans that age don't go to college, so there is also that to
factor in.


Let's assume 2 million freshman-age students of which 1.4 million go
to college of which approximately 2,000 get accepted to Harvard.  That
means one out of every 1,000 college-age students, or about one out of
every 700 freshman students, can go to Harvard.  If that's accurate, we
should expect an IQ at Harvard that tracks pretty close to 3 SDs above
the mean (i.e., 145).


Those are 4-year institutions, and I thought only about 50% of freshman-aged students go to four-year institutions.

Pincher said...

"Those are 4-year institutions, and I thought only about 50% of freshman-aged students go to four-year institutions."

If the freshman-aged students belong to the same age group as the incoming Harvard students, it doesn't matter what kind of learning institution they go to or whether they even go to an institution of higher learning.  IQ is a relative measure across the entire population of an age demographic, not just against the population of college students.So if the frequency of 18-year-olds who are accepted to Harvard is one out of a thousand, and we agree that selection to elite colleges sorts out strictly according to IQ, then it really doesn't matter what the other 18-year-olds are doing.  They could go a state school; they could go to a two-year school; they could skip college altogether; they could go to prison.  It doesn't matter.  You still need to include them in the bell curve.

Pincher said...

Oh, I see what you're saying.  If my numbers are incorrect, I've erred on the conservative side and Harvard students are even rarer than I assumed.  So either way, no big deal.

BlackRoseML said...

 Pincher, I really like you. I do agree my last response was caviling about the 2 million number when I think it is 2.8 although I am currently too lazy to search the numbers.





However, you said: "which would make the average IQ close to 2.5 SDs above the mean."





No, that is the cut-off IQ, not the average IQ. The average IQ would be higher in your model.

Pincher said...

Correct again.

I'm not trying to nail down an exact figure.  I'm trying to show how we might make a ballpark estimate.  The reason I don't worry about the exact figure is that I'm always assuming the stats I use are off by some large degree.

So I prefer to do the exercise more as an illustration of how we might arrive at a figure than I do worrying about the exact figure I arrive at.  There's just too guesswork in all these assumptions.  The statistics I'm using give the illusion of precision and accuracy that just isn't there.

sineruse said...

The logic was that higher contribution from one set of factors (the elastic ones such as preparation, retakes, hours spent on homework) will lower the estimate of the overall contribution from the inelastic factors (luck, talent, good genetics, parents' education) and this reduction will be distributed in some fashion among the inelastic factors.  For example, in a statistical model there will be some negative effect on all of the Inelastic coefficients.

Luck working the same way for everyone doesn't matter because we aren't considering everyone's level of luck on the day of the test, but only the level for those who passed a high cutoff.  

Imagine that Harvard's admission procedure is to select people above the 99.Xth percentile on SAT, or Physical Beauty, or Luck, or on more than one of those traits (presumed independent of each other).  Then, in a population consisting of Slackers and Strivers, if Strivers are twice as likely to meet the SAT requirement, then the admitted Slackers distribution will be about 33% smarties, 33% beauties, 33% luckies while the admitted Strivers are about 25% smarties, 25% future underperformers (high SAT due to striving not ability), 25% beauties and 25% luckies.  You can see that the average level of both Beauty and Luck is lower in the Striver population.  If Harvard academic honors are effectively a test of being at the "smarties" level of ability, Slackers will gain this honor about 33.3/25 = 1.33 times as often per capita as Slackers, or a higher level of disparity at higher cutoffs such as valedictorian.

I gave a similar example in a discussion with a Yan Shen-ish character on the CC message boards (high V, low M, ABC grad student "fabrizio") that worked out in slow motion the admissions implications of disregarding Striver indicators: affirmative action for those with inflated credentials relative to ability.

http://talk.collegeconfidential.com/13591902-post419.html

And for the actual valedictorian data:

http://talk.collegeconfidential.com/13725735-post10.html

BlackRoseML said...

This does not detract from your argument, but I would like to point out that HS GPA has a low ceiling for conscientiousness when measured among a high-g population (SAT > 1400). I would presume that HS GPA doesn't account for much variance among the Harvard applicant pool since high g people can achieve a stellar HS GPA by only applying a minimum of effort; thus, their achievement is not a signal of g-independent conscientiousness. Since GPA cannot be used to discriminate among candidates, in order to select for conscientiousness independent of g, Harvard would need some other indicator to circumvent this ceiling such as extracurricular activities.  

Iamexpert said...

Yan any group or individual that is disproportionately successful is going to get relentlessly bashed, especially in the Internet age where any anonymous poster can say anything about anyone without being held accountable. This is because success (money, power, status, and prestige etc) is largely a zero sum games so those who have a lot of it (whether they deserve it or not) devalue everyone else which creates a backlash.

Having said that I haven't noticed much East Asian bashing and I read HBD blogs all the time. On Steve Sailer's blog I've seen a ton Jewish bashing particularly by anti-neocon types, a moderate amount of East Indian bashing (by those in the computer business) but the only times I've seen East Asian bashing is by those who want to argue than whites contributed more to civilization (and these people are not really bashing, they're just debating).



I have seen a lot of East Asian bashing on Half Sigma's blog however, along with a ton of East Indian, Muslim and black bashing. Half Sigma is Jewish so he does not allow anti-semetic comments, so other groups pick up the slack.

BlackRoseML said...

 
"The SAT scores and school grades/recommendations Harvard uses both
heavily reflect verbal and math skills.  So grades are probably not
really adding any new information about g that the SAT does not
provide."

Another independent confirmation of high g in an applicant makes
it less likely you are picking them on the basis of an outlier test, and
thus makes a stiff regression to the mean on other IQ tests less
likely. Since nearly every successful Harvard applicant is likely to
have several independent confirmations of their high g in their application packet, you have no reason to assume their high SAT scores are outliers.


I don't think high school grades are that g-loaded in a high g population, nor are they are measure of conscientiousness due to the low ceiling of grades. I think many "mere" 1400s can get a > 3.85 HS unweighted HSGPA with a moderate amount of effort. A better confirmation of "g" is whether that GPA was attained with a minimum of effort, since high g people need to apply less effort to receive desirable results, although that would not measure conscientiousness.

Pincher said...

Black Rose,

"I don't think high school grades are that g-loaded in a high g population, nor are they are measure of conscientiousness due to the low ceiling of grades."Your point might be valid if the level of classes by each applicant weren't also taken into consideration by the admissions committee when comparing GPA.  Even an applicant who wants to major in English is likely to have taken AP Calculus, three or four years of a foreign language, and four years of science courses (including at least one AP course) before a school like Harvard even gives his application a more thorough vetting.What's more, there are now the SAT subject tests to think about.  I read somewhere that Harvard requires at least three -- although this Harvard Crimson article says that requirement has been lowered to two.But that's the minimum.  You should note that many applicants to Harvard are competitive and submit more than the minimum.  The young lady in the article above who is surnamed Wang, for example, is submitting four subject tests: Biology, Math II, Chinese, and Chemistry.  Each of those tests would serve as an independent correlation of her g quite apart from whatever her score is on the SAT.  They would also confirm that the science and math courses she took in school were of the top quality and that she was not "given" her good grades in an uncompetitive environment, as is done at some high schools.

That is what makes Iamexpert's opinion about a potential stiff regression to the mean from the SAT test on any second IQ test Harvard students took unlikely.  There are just too many independent confirmations of gnowadays for that to happen.

BlackRoseML said...

 You answered my objection adequately. Although the SAT subject tests probably aren't as g-loaded as the SAT, but as achievement tests, they more g-loaded than grades and it also requires conscientiousness to master the concrete subject material. Your response gave me some appreciation of the value of the SAT subject tests as psychometric instruments, and help me understand why Charles Murray suggested that the SAT should be abolished.

Makes me wonder if the Harvard admissions committee (explicitly or subconsciously) realizes that the requiring the SAT subject tests is a way to prevent admitting people who merely got lucky on the SAT.

BlackRoseML said...

 Well “independent” is the key word.  SAT scores correlate with g because
the verbal and math skills they measure correlate with g.  School
grades also measure verbal and math skills so they correlate with g for
the same reason, so they’re not really an independent confirmation of g.


I think the word "independent" in this context means "independence" from measurement errors of the SAT as these errors render the SAT an imperfect measure of "g" as is all IQ test. School grades are primarily used as a measure of conscientiousness by most colleges, but it is obvious to me that they don't correlate to "g" or conscientiousness in a high ability population. As Pincher pointed out, the SAT subject tests act as an independent measure of "g". Requiring high scores on the SAT subject tests would ablate the advantageous effects of luck on the SAT, but the SAT also measures non-g academic abilities (which are not measurement errors that can be ascribe to random error or luck), thus high SAT scorers not only possess high g, but also these non-g abilities too. The SAT subject tests are likely to capture non-g abilities too. It is likely that there will be some regression (probably not on the magnitude of 1 SD) in a population pool selected by their SAT scores since the SAT captures non-g abilities when this population takes another g-loaded test that doesn't reward the non-g abilities tested by the SAT.

Iamexpert said...

Well to me, independent measures of g are measures of g that only correlate with each other because they all measure g. No other reason. The SAT, grades, SAT subject tests and even letters of recommendation all correlate with each other, but not just because they all measure g, but also because they all measure verbal and math skills. Thus these are not independent measures of g. Truly independent measures of g would be a test like the Raven Progressive Matrices (which has no verbal or math component) or chronometric testing or a card sorting test if executive function and perhaps even indicators of musical talent or chess ratings. That's not to say that grades and SAT subject tests add no independent information; they are independent measures of verbal and math skill and thus increase the odds that Harvard is selecting the most verbally and mathematically gifted youth, and since verbal and math skills are very g loaded, getting a more robust measure of these traits increases the g loading of Harvard's selection process. It just doesn't do so anywhere nearly as efficiently as a truly independent highly g loaded measure like the Raven or chronometrics would. And so because the Raven is so independent of the SAT, people with extreme SAT scores regress precipitously to the mean on this test. Pincher makes a good point that Harvard sudents should regress less than high SAT performers who are not accepted by Harvard, and the reason Pincher's point is valid (though perhaps overstated) is that Harvard students have more g than non-Harvard SAT champions because Harvard students have demonstrated their verbal and mathematical giftedness in more and more diverse ways than just the SAT (grades in many subjects, SAT subject tests, letters of recommendation), however because they were not highly selected for intellectual giftedness in any non-verbal or non-mathematical domain, they too would regress (to a lesser degree) especially on an IQ test devoid of verbal or mathematical content.

Measurement error is probably a less significant factor because long standardized tests like the SAT tend to have near-perfect reliability. That's not to deny people can just get lucky on the SAT, but the luck is probably more than they just happen to be gifted in the particular parts of intelligence that the test measures, rather than that the test measured these parts incorrectly.

Iamexpert said...

Spearman's law of diminishing returns is greatly exaggerated. Most of the studies documenting it seem highly flawed.

Pincher said...

My response is above.

Pincher said...

Iamexpert and Black Rose (discussion continued from below),

"Well to me, independent measures of g are measures of g that only correlate with each other because they all measure g. No other reason. The SAT, grades, SAT subject tests and even letters of recommendation all correlate with each other, but not just because they all measure g, but also because they all measure verbal and math skills. "

I disagree.  One's success at lower levels of math is obviously not assured at higher levels of math.  In fact, the thinning takes place rapidly the higher you go.  Some conscientious and moderately intelligent students are able to do quite well through algebra, but then hit their limits when studying pre-calc.  This will show up in both grades and various test scores.

Similarly, anyone who has studied a foreign language at any depth knows that it has only a modest correlation with English ability.  I know many people who scored very well on the SAT verbal who struggle to learn foreign languages.  You should not assume a high overlap between the two.

So the more subjects you test, the more likely you are to thin according to g.

Look at the basic info from these two potential students's applications:

Student #1:

SAT (math/reading/writing): 780/700/700 (99%/95%/96%)
GPA: 3.94 (unweighted) B in AP Calculus
Math II subject test: 760 (81%)
French subject test: 680 (66%)

Student #2:

SAT (math/reading/writing): 800/680/690 (99%/93%/95%)
GPA: 3.94 (unweighted) B in 3rd-year Latin
Math II subject test: 800 (90%)
French subject test: 780 (89%)
Latin subject test: 750 (86%)
Physics subject test: 760 (82%)

You should note that there is essentially no difference between student #1 and student #2 when looking at their GPAs and SAT scores.  In fact, student #1 has a slight advantage in his total SAT score (2180 to 2170) over student #2.  But the are essentially identical.

Where the two students begin to more clearly separate from one another is in the subject tests, and I contend we should assume that separation represents a cognitive gap between the two.  Student #2 has a solid lead in Math II, an enormous lead in French, and two additional tests where he has scored very well.

Pincher said...

See reply above.

Iamexpert said...

Many of these qualities (voice tone, body language, facial
expressions, etc.) are like singing, dancing, or playing a sport.  They
are natural physical abilities that may require some training, but they
aren't mental tasks that correlate with g.  The fact that Oprah is good at them doesn't tell us anything about her intellectual abilities.


I think being a good talk show host is like being the writer, director, and lead actor of your own movie every single day and doing it all largely extemporaneously, and making it so amusing and interesting that many millions of people hang on your every word, day after day, week, after week, year after year, in the face of extreme competition.  I wouldn't compare it to singing and dancing, both of which require physical aptitudes like vocal range and motor coordination; by contrast virtually everyone has the physical ability to make goofy facial expressions or alter their tone of voice, however having the creativity, timing, and rhythm to rapidly improvise these actions in a way that entertains millions might depend heavily on lateral thinking.     




As to changing the culture, who cares?  We are talking about
geniuses, not influential people.  They are not the same.  When someone
calls Gandhi a genius, I don't know what the hell they are talking
about, but he was obviously quite influential.



Influence is a necessary but not sufficient criterion for genius.  Remember Jensen and others use the term genius to describe socially significant creativity sometimes even substituting the term "eminence" for genius. 

Iamexpert said...

Your point about SAT subject tests is well taken.  But you have to remember that even if Harvard did a perfect job screening for g (which it doesn't), IQ tests don't perfectly measure g, so people with an extreme level of g inevitably regress to the mean on IQ tests simply because IQ tests include non-g variance, and the shorter the IQ test, the higher the standard error.  I think that's probably one main reason Harvard students averaged "only" a 130 IQ on the abbreviated Wechsler.  The other main reason (as you hinted at) is that it's probably social science students participating in psychology experiments, and these typically average about 10 IQ points lower than STEM students.

One interesting thing about the Wechsler testing is that some Harvard students scored below IQ 100.

In the book "The g Factor" Jensen reports that Berkeley students have an average IQ of 120.  This is probably another example of regression combined with selective sampling producing unspectacular IQ scores for impressive schools.
 

Pincher said...

Iamexpert,

Since it was a part of our discussion, I thought you might like to read this: Did Obama have lower SAT scores than George W. Bush?

"Only 450 students applied to transfer to Columbia in 1981 and sixty-seven were admitted, according to the Columbia Spectator, compared to 650 applicants just four years before."
"If Obama’s SAT scores were near the average of the transfer students entering Columbia in the fall of 1981, he would have scored significantly lower than George W. Bush, whose combined math and verbal scores were 1206 out of a possible 1600 points (as revealed by the New Yorker in 1999)."
"In his autobiography, Dreams from My Father, Barack Obama describes himself as an unfocused high school student whose mother scolded him for being a "loafer" (142). He describes his attitude toward his studies at Occidental as “indifferent” (146), calling himself a “bum” who abused drugs (138) and who was notorious for partying all weekend (165)." 
"That has raised questions about how Obama earned a place at Columbia in 1981, paving the way to Harvard Law and beyond. Indeed, like much else in his biography, Obama seems to have fictionalized the process through which he gained admission to Columbia. Obama writes in Dreams: "[W]hen I heard about a transfer program that Occidental had arranged with Columbia University, I’d been quick to apply"(172). "


Interesting stuff that puts some parameters around Obama's academic accomplishments prior to Harvard.

Blog Archive

Labels