The Unz Review: An Alternative Media Selection
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 BlogviewJames Thompson Archive
Gould Got It Wrong
🔊 Listen RSS
Email This Page to Someone

 Remember My Information



=>

Bookmark Toggle AllToCAdd to LibraryRemove from Library • BShow CommentNext New CommentNext New ReplyRead More
ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
AgreeDisagreeLOLTroll
These buttons register your public Agreement, Disagreement, Troll, or LOL with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used once per hour.
Ignore Commenter Follow Commenter
Search Text Case Sensitive  Exact Words  Include Comments
List of Bookmarks

It is very unlikely that even if I continue my blog for decades, it will ever have the impact of Stephen Jay Gould’s (1981) “The mis-measure of Man”. It was a best seller, cited in the academic literature over 10,000 times, and even 445 times in 2017 alone. It continues to meet an audience need.

Why was it so popular? I read it and found that it was written in a very engaging way. In my view Gould had an excellent prose style. I enjoyed his essays. His book attacked intelligence tests, which had fallen in popularity, and had come to be seen as a Bad Thing. Intelligence testing had originally been seen as a very good thing, providing opportunities to bright young children who could not afford fancy schools, but who deserved the opportunity of good quality education and employment. Intelligence tests were meritocratic, not aristocratic. You could not fool them with specific knowledge derived from private tuition. They were the great levellers. Although it is hardly relevant to their actual veracity, they were warmly received by the political Left, who saw in these assessments a vindication of the working-class talents which had been suppressed by private education.

Why was it that SJ Gould had such an impact when he argued that the tests were biased against working class and minority racial groups? Morever, how did his views ever take hold when the issue of bias in intelligence testing had just been comprehensively evaluated in Arthur Jensen’s (1980) “Bias in Mental Testing”. Jensen showed that, far from under-predicting African-American achievements, they perhaps slightly over-predicted them. I presume that Jensen’s volume was less often read, though it was written by an expert, not a polemicist. Perhaps precisely because it was written by an expert, in a restrained and far from folksy style, it had less impact on popular culture, which is what tends to determine public debates.

I leave the full explanation to others, but I think that a good prose style, no equations, few numbers and little in the way of statistical and logical arguments generally increases readership. That would be predicted by the bell curve, which makes it plain that technical books about difficult subjects are a minority interest.

Gould’s book made a number of assertions. Two that stuck in people’s minds were: that measures of brain size derived from the study of skulls of different races had been biased, and that many items on the Army tests of intelligence were culturally biased.

The debate about the ancient skulls has raged to and fro for a long time, but it seems highly probable that the measures were taken correctly

Now the redoubtable Russell Warne has taken a detailed look at what Gould said about the Army Beta test, and finds that on that topic he has been unreliable and incorrect.

https://res.mdpi.com/jintelligence/jintelligence-07-00006/article_deploy/jintelligence-07-00006.pdf?filename=&attachment=1

A number of points:

Face validity. Sure, it helps if a test item looks relevant to the job you are applying for. However, a test item may have high predictive value without seeming to. This is the famous “indifference of the indicator” dictum. If it predicts, use it. Furthermore, you cannot dismiss an item simply because you yourself can think of a way in which it might be misinterpreted, as Gould did. You need to show that such misinterpretations actually exist (and compare them with the misinterpretations which arise on items which seem fine to you).

Testees were not baffled by the use of numbers, in the sense of digits, as Gould implied. All language speakers had knowledge of digits because they had had some years of education.

Gould twists things. His reading of the instructions was that the men would be “scared shitless” whereas an officer who had actually done the testing wrote later: “It was touching to see the intense effort put into answering the questions, often by men who never before had held a pencil in their hands”. A shade different, don’t you think?

Gould claimed that “vast numbers of men” earned zero scores, and therefore, must not have been able to understand the Army Beta test instructions and/or stimuli. However, only 4% scored less than 10 in total, and only 2.6% of testees scored less than 5 points. Gould neglected to point out that the standard procedure in the Army was that the low scorers were then individually tested on the Stanford Binet to give them yet another chance to do well.

Gould reports an officer’s unfavourable view of the testing, but does not show that 13 other officers were favourable.

Gould criticised the short time limits on some subtests, saying they also were too short for his biology students, on whom he used the test (see later). Warne politely explains that short time limits on process tasks are required because otherwise they are too easy, and discriminate poorly. Short time limits are a good feature, not a bug. (This is a common misunderstanding. See Hyde on sex differences in the speed of completion of tasks).

Gould criticised the Beta test, saying that poor testing circumstances meant that it could not be considered a test of innate intelligence. He failed to tell his readers test-constructor Boring’s opinion that the tests had predictive value. Also, the test creators rarely mentioned “innate intelligence”. They simply found that test results helped them predict who would do well on the tasks the army required, which was the whole purpose of testing.

The test creators believed that different levels of education were likely to have influenced performance on the test, as did their immigrant status, but Gould cast Yerkes as dismissing that factor, when in fact he discussed it and correctly said that a correlation between years spent in the US and higher test scores showed an aculturation effect, but did not identify a cause.

Gould also downplayed the work done on establishing the validity of the Army tests. Scores on the Army Beta correlated positively with scores on other intelligence tests, including the Army Alpha (r= 0.811) and the Stanford-Binet (r= 0.727), both the “gold standard” of intelligence measurement at the time ([15], p. 634). Army Beta scores also correlated positively with external criteria, such as the number of years of schooling a recruit had (both as children and adults), commanding officers’ ratings of soldiers’ job performance, and army rank.

After all this, I would have regarded Gould has having given an unfair account of the test, and left it at that, job done. Warne, perhaps prisoner of the American work ethic, has gone further. He gave the Beta test to his students, and also pre-registered his expectations. This is excellent. Instead of getting the results and saying “I told you so” he puts his prior assumptions up for examination. If only Gould had done that.

For me the most interesting result is that the test picks up what looks like a confirmation of a secular trend. As more people get to go to college, scores go down and more resemble the average in the general population from which students are selected.

Warne says:

Given these results from our replication, it seems that Gould’s criticism of time limits and his argument that the Army Beta did not measure intelligence are without basis. Despite the short time limits for each Army Beta subtest, the results of this replication support the World War I psychologists’ belief that the Army Beta measured intelligence. We demonstrated this in the following four relevant results of our replication:

Here is my summary of those four results.

1 Gould’s Harvard students were 1.3 standard deviations brighter than Warne’s open access college. Selection was more important than the Flynn Effect. (This result makes a lot of sense).
2 Gould had administered the test properly (against Warne’s own supposition, which he could have covered up if he had not been honest enough to preregister his suppositions).
3 The Beta subtest scores correlate positively (positive manifold), and the total score correlates with two self-reported measures of scholastic attainment.
4 The Beta test is best described by a single factor.

In summary, Warne has done an excellent job in showing that Gould traduced the Army Beta test and the researchers that created it. Bluntly, Gould mis-represented the test, and misled his readers. Gould probably achieved his objective, which was to trash intelligence testing in the eyes of a generation of academics.

Warne has shown that the Beta test still works. It is a good predictor of intelligence, which correlates with current measures of scholastic attainment, shows a positive manifold and resolves into a common factor. In a standard which Gould never attempted, Warne pre-registered his prior assumptions so that the results of his experiment could be plainly seen by the reader, and so that the facts could prove him wrong.

Warne’s achievement is to have shown that Gould got it wrong.

 
Hide 243 CommentsLeave a Comment
243 Comments to "Gould Got It Wrong"
Commenters to Ignore...to FollowEndorsed Only
Trim Comments?
    []
  1. nickels says:

    Stephen J. Boas.
    If anyone ever showed evolution for the scam it is, Gould is the man.
    His contorted and convoluted attempts to save the dying theory are enough to show the whole tradition has jumped the shark.

    • Replies: @Mr. Rational
  2. res says:

    Thanks. Looks like a very interesting paper. Hard to believe this requires confirmation:

    For me the most interesting result is that the test picks up what looks like a confirmation of a secular trend. As more people get to go to college, scores go down and more resemble the average in the general population from which students are selected.

    But good evidence demonstrating it is worth having.

    I am starting to wonder if the source of much of the IQ (SAT, etc.) test hate is the “elites” discovering that they really aren’t that elite in a meritocratic sense after all. Before the tests it was easy to delude themselves that superior ability on average translated to All > All. After, not so much. Which is better, to have to give Harvard $10 million to get your not so smart child admitted or just to discredit admissions tests? Affirmative action provides good cover as well because the scores of the “special” people don’t stick out so much. And they don’t have to end up at the bottom of the class.

    Off to read the paper.

    P.S. The last time you posted about a paper from Dr. Warne it inspired a 300+ comment thread ; )
    http://www.unz.com/jthompson/fear-and-loathing-in-psychology/
    Fun fact about that article from his page below: “This article was one of the 10 most downloaded articles out of the 4,000+ articles that the American Psychological Association published in 2018. Click here to read that announcement.”
    Based on the two papers you have summarized here it looks like Dr. Warne does interesting and thorough work. Here is a link to his UVU page which includes a (partial?) list of his papers:
    https://www.uvu.edu/profpages/profiles/show/user_id/10984

    • Replies: @Wizard of Oz
  3. of course he got it wrong. deliberately, on purpose. it was never a good faith argument to begin with.

    (gould). always check. never don’t check.

    • Agree: Wally
    • Replies: @Pat Boyle
    , @m___
  4. Some fun facts.

    Wealth correlates with low IQ.

    Little Nigerian girls in England are smarter than you.

    I learned both of these fun facts right on here. Didn’t have to search around like you do with your cherry-picking.

  5. dearieme says:

    Gould was a crook, wasn’t he? Liars tell lies – that is their defining characteristic.

  6. I’ve read a few of Gould’s books too but found them not very well written. Too many suppositions – = new beginnings/ no proper thought-line. Too much confusion. I decided I would wait and see what would happen. In Germany, Gould managed to irritate and confuse the poet and essayist Hans Magnus Enzensberger, who wrote a little book, trying to sum up the scientific debate about intelligence testing etc. – Enzensberger, who wrote a book, which sold quite well worldwide about numbers/mathematics (“The Number Devil”) got stuck while writing about IQ. Title of his IQ-book: “In the Maze of Intelligence – An Idiots Guide” – “Im Irrgarten der Intelligenz – Ein Idiotenführer”.
    Interesting aside: Enzensberger, born in 1929, wrote a book about his flops, in which he did not include his book about intelligence testing etc. –

    Dr. Warne seems to do good and interesting work indeed.

    PS

    One does not really know when an argument is won. But for me, 2017 was the year I got a clear feeling, that in the battle over the IQ there was a major shift taking place. – Not least because of this blog here (and Sam Harris’ podcast with Charles Murray, and…).

  7. bob sykes says:

    In “Mismeasure of Man” Gould outright lied about the tests of skull volume. He stated that the difference between the MEANS of skull volumes for different races should be divided by the SAMPLE standard deviations. But statistcal theory requires that the divisor be thr standard error of the means. This latter is much smaller than the sample standard deviations, and the ratio is much larger.

    The researcher Gould criticizes for incorrect method actually used the correct method. Gould knew he did, because Gould himself used the same statistical procedure in his own research.

    When I first read that passage in Mismeasure, I knew Gould was a charlatan.

  8. Of course you know you are being attacked by Taleb: https://medium.com/incerto/iq-is-largely-a-pseudoscientific-swindle-f131c101ba39; it’s actually an update to an earlier article. They also believe Gould was right and that he was a great evolutionary biologist. Maybe they should stick with Math. I know of no scientist who was anything in the field who didn’t think that Gould was a con artist and had a very poor understanding of evolution. From Dawkins to Wilson they knew what Gould was about. Unfortunately, the mainstream media continues to this day to Sanctify and Fellatio him every chance they get.

    His Mismeasure of Man or as some call it the Mismeasure of Science is still being taught and used. The Communistos believe what he said is correct. Here’s a site that shows what numerous scientists thought of Gould:

    http://nathancofnas.com/what-prominent-biologists-think-of-stephen-jay-gould/

    To put it another way it’s hard to beat a Magic Jew from Harvard when all the Magic Jews of academia have made him in to this SJW “God.”

  9. res says:
    @niteranger

    I missed Taleb’s shot at Dr. Thompson. His name does not appear in the text, but a screenshot of the “Swanning About” article appears about halfway down.

    This bit from Taleb says all that needs to be said (emphasis mine):

    CURSE OF DIMENSIONALITY A flaw in the attempts to identify “intelligence” genes. You can get monogenic traits, not polygenic (note: additive monogenic used in animal breeding is NOT polygenic).

    Hint, when reality differs from theory, question the theory. I wonder how he explains skin color?

    It was fun to notice Taleb “applauding” a comment which begins:

    1. Stepehn J Gould’s Mismeasure of Man is still the best historical and critical work regarding intelligence testing.

    LOL. Even funnier in light of the paper currently being discussed here.

    One good thing about the Taleb IQ kerfuffle is it was helpful for him to out himself as an idiot.

    • Replies: @Wizard of Oz
  10. @niteranger

    Thank you for sending me the Taleb link. I did not know about it, but that is because he has blocked me! As a consequence he is able to criticise me without any prospect of a prompt reply on my part. His post has been somewhat altered, so I suppose I should look at it. However, it is mostly the same, and likely to remain so.

    I don’t know what he continues to be so angry about measures of ability.

    • Replies: @Poco
  11. Why was it so popular? I read it and found that it was written in a very engaging way. In my view Gould had an excellent prose style. I enjoyed his essays.

    I felt the same, but I realize now that I was being tricked. Gould makes his liberal readers feel good about themselves and the way they can understand the “complex” topics he is discussing, as he guides them away from the pernicious errors to which lesser intellects fall victim. But it was smoke and mirrors in typical stale-pale-male fashion. Steve Sailer nails Gould here:

    The announcement that the number of human genes is probably in the range of 30,000 rather than 100,000 – as was widely guesstimated until recently – has inspired all sorts of propaganda about how this shows that nurture is more powerful than nature. Harvard paleontologist Steven Jay Gould unburdened himself of some of his trademark sonorous imponderables in The New York Times.

    “The social meaning may finally liberate us from the simplistic and harmful idea, false for many other reasons as well, that each aspect of our being, either physical or behavioral, may be ascribed to the action of a particular gene “for” the trait in question. But the deepest ramifications will be scientific or philosophical in the largest sense.”

    https://vdare.com/articles/we-know-they-said-created-equal-but-they-didn-t-mean

    “Sonorous imponderables” is perfect! Gould didn’t believe in telling the truth or advancing scientific understanding: he believed in “What is best for the stale pale males and females to whom I related?” And because those stale pale folk are so powerful in publishing and the media, he did very well from his lies.

    • Replies: @anonymous
  12. ‘Why was it so popular?’ – I think the main reason, moreso than the style of writing (& also contrasted with Jensen’s writing), is simply that one told certain influential people what they wanted to hear at this juncture in history, and the other didn’t.

  13. We should have never let him come through the stargate.

    • LOL: Joseph Doaks
  14. Re: “Gould got it wrong…”
    No. He lied. He propagandized his lie to social political ends. He was despicable and has effectively gotten away with it, but I assume is now paying for it and his other malignant activities.

    • Replies: @res
  15. res says:

    Some thoughts/questions/comments about the paper. Criticism welcomed. Just please have an argument (and data, if appropriate).

    I found the summary of the Army Beta test in section 2 very helpful. My knowledge of the early Army testing was fairly superficial and that helped flesh it out.

    I was somewhat mystified by the concern about immigrants and numbers (originally from Gould). My understanding is that Arabic numerals are (and were 100 years ago) in use worldwide. So as long as immigrants had been educated in their homeland they should have learned numbers. Math’s invariance relative to human spoken language is one of its good features here. Concern over numbers for those with only a few years of formal education (native or immigrant) is reasonable though. Am I missing something here?

    I hadn’t realized the Gould biology course in question was on “biology as a social weapon.” At least he was open about his intentions. But more seriously, unless that course was required I’d be surprised if it got the best of the biology students (i.e. assuming an average Harvard IQ might not be appropriate).

    There was an extended discussion of “innate intelligence” on pages 9-11. I’m not sure that is a useful phrase because I think it conflates multiple concepts.
    – Genetic intelligence
    – Current intelligence unimpaired by any current environmental effect (e.g. alcohol).
    – Current intelligence
    Put another way, there can be different time constants for environmental effects on intelligence. Compare with the phrase “examinee’s developed intelligence” which is also used.

    On page 13 we have:

    The self-reported GPAs of our sample were high (M = 3.50, SD = 0.54), while the ACT scores were consistent with the university’s open enrollment mission (M = 23.1, SD = 3.8). Using the equation provided by Koenig et al. (2008, p. 157), these ACT scores are the equivalent of a mean IQ of 103.6 (SD = 2.58).

    The Koenig reference is available at http://www.iapsych.com/iqmr/koening2008.pdf
    Equation 1 is IQ = (0.685 * ACTTOTAL) + 87.760

    That IQ SD of 2.58 seems small (though it follows from the equation) and I find equating an ACT score of 36 to an IQ of 112.4 questionable. The equation can only generate IQ results within +/- 1 SD.

    For comparison, a pre 1989 ACT composite of 29 was accepted for Mensa (IQ ~130): https://www.us.mensa.org/join/testscores/qualifying-test-scores/
    Currently an ACT score of 33 is 98th percentile: https://www.manhattanreview.com/act-percentiles/
    An ACT score of 23 (per equation, IQ 103) is 69th percentile.
    Given that the ACT test taking population is selected (likely to have mean IQ > 100) I find those results inconsistent.

    Given that, I think that Koenig conversion from ACT to IQ is insufficiently accurate. Perhaps a concordance table would be better? I am not finding one for the ACT, though there are plenty for the SAT.

    Though perhaps I should defer to Dr. Warne here given that he has written a book on statistical methods: https://www.amazon.com/Statistics-Social-Sciences-General-Approach/dp/1107576970

    I liked the pre-registration (which Dr. Thompson also noted). In particular, I appreciated (and enjoyed) the frankness of the authors:

    We pre-registered our analyses because we recognized that we had a positive bias towards the Army Beta’s creators’ work and a pre-existing negative view of The Mismeasure of Man.

    I also appreciated the open science policies followed! Available at https://osf.io/y6njt/

    If only there could be more frankness about biases, use of pre-registration, and use of open science in the scientific literature as a whole.

    Page 17 equates a 1991 1390 SAT score to 30 on the ACT:

    incoming freshmen in 1991 at Harvard averaged 1390 on the SAT [49], which is the equivalent of a composite score 30 on the ACT.

    Reference 49 is a 1993 Crimson article: https://www.thecrimson.com/article/1993/5/7/report-discloses-sats-admit-rate-pa/?page=single
    (the unpublished data they describe sounds very interesting)

    Therefore the ACT conversion above ignores the 1995 SAT recentering. Based on https://supertutortv.com/psat/whats-a-good-psat-score I think 33 would be more accurate, but I can’t find a pre-1995 SAT-ACT concordance table to be sure.

    Thanks for an interesting and useful (as detailed rebuttal of one of Gould’s excesses) paper, Dr. Warne! The paper text and extensive references are also useful as a compilation of contra-Gould arguments in general.

    P.S. Worth mentioning that I believe the PDF is a version of the paper prior to peer review.

    • Replies: @europeasant
  16. res says:
    @InSearchOfTruth

    The title of Gould’s biology course says it all: “Biology as a Social Weapon.”

    Looking into that I found a 1977 book with that title: https://www.amazon.com/Biology-social-weapon-Various-Authors/dp/0808745344

    And Jensen’s review of The Mismeasure of Man (reference 4 in Warne’s paper): https://www.debunker.com/texts/jensen.html
    The title is great: The Debunking of Scientific Fossils and Straw Persons

    Does anyone have a syllabus, lecture notes, etc. for Gould’s course? I suspect that would be illuminating.

    • Replies: @Dieter Kief
  17. Perhaps the fraud thought NYRB is never read by conservatives like me, who saw through him from his first utterances on the American stage.

    A Gould quotation that explains so much:

    “I am hopeless at deductive sequencing…I never scored particularly well on so-called objective tests of intelligence because they stress logical reasoning… ”

    March 29, 1984 issue of The New York Review of Books.

  18. Re: “I am hopeless at deductive sequencing…I never scored particularly well on so-called objective tests of intelligence because they stress logical reasoning… ”
    Translation: “I am a midwit at best but good at obfuscating in favor of my chosen position. Some call it lying; I call it advocating without the facts I don’t understand anyway…”

    • Replies: @anonymous
  19. Pat Boyle says:
    @prime noticer

    There is a simple explanation for why Gould has attacked IQ so vigorously.

    Gould was a Marxist.

    Marx basically believed in the fundamental influence of economics not biology. He believed that your social and economic class determined your fate. Innate or partially innate characteristics were to be disputed. The Soviets building on this idea created the notion of the “New Soviet Man”. You were not constrained by heredity. The state could fashion new citizens of any type with the appropriate kind of society. This kind of thinking also led to Lysenko.

  20. @Pat Boyle

    Re: “The state could fashion new citizens of any type with the appropriate kind of society. This kind of thinking also led to Lysenko.”

    It also led to the killing fields of Cambodia and the destructive activities of Mao’s cultural revolution.

  21. cfhosford says:

    It should be noted that the Wikipedia entry on Gould essentially exonerates him of any errors in “The Mismeasure of Man.” (And yes, I read it when it first came out, and enjoyed his popular writing style.) While I am a prolific Wikipedia editor, I’m not versed in the scientific literature that would balance the Gould main entry, and adequately explain the critical evaluations of his book. I would urge those on this blog who know how to edit Wikipedia entries, and also have a knowledge of the science, to please update the main Stephen J. Gould Wikipedia entry with the appropriate countervailing data that details his mistakes (both political and scientific) in “The Mismeasure of Man.” The site is here: https://en.wikipedia.org/wiki/Stephen_Jay_Gould.

  22. @Pat Boyle

    For reading more about Marxism/Biology I recommend:

    Bernard Davis “Storm Over Biology” Prometheus Books, 1986.

    John Casti “Paradigms Lost” Wm Morrow, 1989

    and, shamelessly, James Graham, “Cancer Selection” Aculeus Press 1992 where I wrote:

    “… I discovered, not long ago, that the followers of that nineteenth century crackpot, Karl Marx, have considerable influence in theoretical biology, particularly in the United States and Great Britain. This is not the place to examine this bizarre situation in any detail, but to give some idea of how weirdly pervasive is the influence of that third-rate thinker in one of our major sciences I cite the following flagrant examples.

    “Stephen Jay Gould, current President of the Society for the Study of Evolution, Harvard professor and author of many popular books on evolution, seemed to think it important to proclaim in an evolutionary journal–of all places!–that he “learned his Marxism, literally at his Daddy’s knee.”

    “Two of Gould’s Harvard colleagues [Levins and Lewotin] dedicated a biology book to Karl’s fellow genius, Friedrich Engels. They did that, not in the 1930s, but in the mid- 1980s.

    “Scientists who contradict Karl’s doctrines are routinely savaged by Marxist science-polluters. Favorite targets are biologists who suggest that inheritance may have some influence on human behavior. Apparently Karl proclaimed otherwise. Fortunately for today’s heretics, the apparachiks of today’s People’s Republic of Biology don’t kill those who disagree with them. Forced to conform to the standards that prevail in the Western countries in which they and their families (Marxist Daddies included) choose to live, they lack the power of Stalin’s favorite biologist, Grigor Lysenko, who routinely arranged for the extermination of Russian scientists who disagreed with the Marxist doctrine that Mendelian genetics was an imperialist plot.”

  23. m___ says:
    @prime noticer

    Indeed, in the view of general human nature, the cesspool of current society, Gould was a winner.

    One single book and consolidating the derivative of personal wealth and ego. Could as well have written the Koran, single-handedly, and glorified Muslim civilization, or Harry Potter.

    A winner is someone with disregard for the longer term, the extended context, the lack of ethics, in our current social system. Some-one who rules on crap, not unlike our current billionaire class. He wins. The up-vote done by democratic dumb-sters, the next time, my time, middle class crows.

  24. @cfhosford

    I would urge those on this blog who know how to edit Wikipedia entries, and also have a knowledge of the science, to please update the main Stephen J. Gould Wikipedia entry with the appropriate countervailing data that details his mistakes (both political and scientific) in “The Mismeasure of Man.”

    It is not as easy as it sounds. These days, Wikipedia does not want references to primary sources, so your edit will immediately be reversed if it cites Warne’s paper. The fact that the page you are editing may cite numerous other primary sources does not matter: your edit will be reversed, the previous page will remain with its own citations of primary sources intact.

    You could try citing this article as a secondary source. If that is not approved, you will have to wait until a discussion of Warne’s paper occurs in a “respectable” medium.

    • Replies: @Alfa158
  25. @res

    “Storm over Biology”, Prometheus Books 1986, by Bernard D. Davis lists a number of false conclusions by Gould. Davis is very much in favor of Arthur Jensen.

    For example: Gould: “The chimerical nature of g is the rotten core of Jensen’s edifice, and of the entire hereditarian school… Spearman’s g and its attendant claim that intelligence is a single measurable entity, provided the only theoretical justification, that hereditarian theories of IQ have ever had”. And Davis’ clear and concise answer: “This assertion is utterly false. Whether an IQ test measures mostly general intelligence or mostly a collection of independent abilities, the heritability of whatever it measures will be precisely the same. IQ’s factor structure simply does not enter the equations for calculating its heritability.” ( p.126 in “Storm over Biology” )

    I think, that solves quite a bit of the Gould problem.

    Davis mentions, that Goud was a member of the left-wing group “Science for the People”. Davis: “That group has aimed at politicizing science attacking, in particular, any aspects of genetics, that may have social implications. (…) Gould has spelled out explicitly his ideological commitment, and also its influence on his science.”
    Davis, p. 128

    I just want to add, that Davis is refreshingly clear about the distinction between scientific work and – other aspects of science. If you confuse the two, you declare intellectual war (which is something else than science altogether). That’s what Gould did: He marched into biology and social psychology and tried to conquer it for the left.

    • Replies: @res
  26. Ron Unz says:

    Many of you may already be aware of it, but here’s a very damning portrait of S.J. Gould based on personal experience, which Prof. Robert Trivers published on this website back in 2015:

    http://www.unz.com/article/vignettes-of-famous-evolutionary-biologists-large-and-small/#stephen-jay-gould

    Trivers is not only a towering figure in modern ev-bio theory, but also an ultra-hard-core leftist, probably far, far to the Left of pseudo-Marxists frauds like Gould and Lewontin. For example, I think he was the only white member of the Black Panther Party, and dedicated his most recent book to the memory of his closest friend, Huey Newton…

    • Agree: Hail
  27. res says:
    @Dieter Kief

    “Storm over Biology”, Prometheus Books 1986, by Bernard D. Davis lists a number of false conclusions by Gould. Davis is very much in favor of Arthur Jensen.

    Thanks to you and James Graham for the book recommendation(s). I actually have a copy of Storm over Biology. I just need to find it and read it. I bought it based on an earlier recommendation by JamesG: http://www.unz.com/isteve/vox-charles-murray-is-once-again-peddling-junk-science-about/

    More Gould fun in this 2011 comment by James: https://www.unz.com/pfrost/arroseur-arrose/#comment-611194
    “gouldcrap” is an amusing coinage.

    • Replies: @Dieter Kief
  28. Hail says: • Website
    @Ron Unz

    Excellent article. The Gould section is a lean 1,660 words; a few minutes’ leisurely reading and worth the time.

    Many of us theoretical biologists who knew Stephen personally thought he was something of an intellectual fraud precisely because he had a talent for coining terms that promised more than they could deliver, while claiming exactly the opposite. One example was the notion of “punctuated equilibria”—which simply asserted that rates of (morphological) evolution were not constant, but varied over time, often with periods of long stasis interspersed with periods of rapid change.

    Recently something brand new has emerged about Steve that is astonishing. In his own empirical work attacking others for biased data analysis in the service of political ideology—it is he who is guilty of the same bias in service of political ideology. What is worse—and more shocking—is that Steve’s errors are very extensive and the bias very serious. A careful reanalysis of one case shows that his target is unblemished while his own attack is biased in all the ways Gould attributes to his victim.

    • Replies: @res
  29. res says:
    @Hail

    Worth noting the Gould comment in the Lewontin section as well:

    By the way, Lewontin would lie openly and admit to doing so. Lewontin would sometimes admit, in private at least, that some of his assertions were indeed fabrications, but he said the fight was ideological and political—they lied and so would he. On other matters, such as committee work, Lewontin could be rational and useful. Much less so, it was said was Stephen Gould, who was into self-promotion, self-inflation and self-deception full time. Not only was his science hopeless but so was much of his behavior in other contexts as well. I can remember him arguing against offering a full professorship to a truly excellent Colombian biologist because we would be discriminating against a third world country by depriving it of him. A heavy silence enveloped the room. Surely there was no brain/drain problem in evolutionary biology (Colombia to the United States) comparable to the one for nursing. Why not imagine the benefit he could confer on his own country sitting on a fat sum of Harvard money, linked to one of the best Museum collections in the world, with multiple opportunities to polish off his top scientists.

    • Replies: @Dieter Kief
  30. MBlanc46 says:
    @obwandiyag

    Dang. Already used my Troll allotment for this hour.

  31. aandrews says:

    “In my view Gould had an excellent prose style. I enjoyed his essays.”

    Back in the day, I subscribed to Natural History primarily to read Gould’s essays.

    • Replies: @James Graham
  32. Alfa158 says:
    @James N. Kennett

    I wouldn’t bother . Wikipedia has a thinly veiled agenda and is also patrolled by SJW volunteers. Edit the Mismeasure of Man away from the approved catechism, and it won’t last five minutes before it gets edited back. Wiki is fine for finding out the maximum cruising speed of the Boeing 707, or Peruvian bauxite ore production for 2012, but is usually unreliable on any subject with political implications.

    • Replies: @Rogue
    , @Wally
    , @Versengetorix
  33. Hail says: • Website

    [IQ tests] were warmly received by the political Left, who saw in these assessments a vindication of the working-class talents which had been suppressed by private education.

    Herrnstein and Murray (1994) had these two graphs, illustrating this point nicely:

    The first graph reflects the U.S. b.1906-07 cohort, growing up in the 1910s and early 1920s, at which time children of families without means certainly tended to orient themselves towards work over education. I believe a majority of those of this age cohort didn’t even finish high school.

    As for the privileged few who among the b.1906-07 cohort who, by merit of family background, were able to go to college: Their in-earnest academic careers will have begun let’s say about in the late 1910s. This is right at the cusp of the first experimentation with IQ testing (Stanford-Binet first administered in 1916) and predates later IQ-proxies like the SAT (the first version of which appeared in 1926, a googling tells me; not sure of the scale of the SAT at that time, but regardless, even this SAT ver 1.0 would post-date the b.1906-07 cohort’s college entry).

    These b.1906-07 young men and women were, therefore, pre-‘Great Sort.’ Consequently, the larger bell curve (no college degree) has many times more individuals at +1 SD and above, than the meager, flattish little college-degree bell curve (which reflected social class).

    By 1990 (image 2), for the b.1966-67 cohort, there were more college graduates than non-college-graduates at +1 SD, and competitive numbers at +0.5SD to +1 SD. The “(Any) College Degree” curve’s mean had shifted up, and the “Elite College Degree” curve’s mean had shifted way up (though IQ blogger Pumpkin Person has much material showing elite college graduate IQs are overpredicted by their SAT scores, and that their true IQ mean is in the 120s not at 140, which is implied by the Herrnstein and Murray figure above).

    Such was the Big Sort.

    Gould (1941-2002; resident of Queens, New York City from birth to 1960) was himself a product of the post-‘Sort’ world; in the 1950s, even though some/much of the rest of the country was not fully ‘sorted out,’ yet, his world (NYC) was likely very much ‘sorted’ for the upcoming youth like Gould then was, i.e., Smart = College, as a given; while some may opt out, it was increasingly expected, a strong departure from those born in earlier decades.

    In this sense, it makes some more sense why Gould would have been among the anti-IQ’ists (among other possible reasons I won’t comment on) — by which I mean his anti-IQ’ism was likely a reflection of the privilege he knew of a meritocratic world, quite unlike that of the b.1900ers and b.1910ers of his parents’ generation, for whom college was a privilege of the top social elite and decidedly not the cognitive elite (see first image above). He was taking meritocracy in education for granted, because it was all he really knew.

    Certainly the pre-IQ (and pre proxy IQ) Testing world had a serious social problem in which too much talent was being ‘wasted,’ as described above. (also, e.g.: No one in my family on either side had been to college before the 1960s; some sure had the aptitude for it, but were hard at work and it was simply never an option at all.) But I’m not sure our post-Great Sort, 1990s-, 2000s-, 2010s-era system is really so much better —

    We now have a new, weird little form of social-class division distinguished by education level. In other words, many of the educated see themselves as members not “of” the nation, but members of a Brahmin-caste-like cognitive elite social class that is more important than that rabble out there called the Nation. Parallels with the arrogance of 18th-century France’s aristocracy suggest themselves.

    (I am reminded of Donald Trump who, at a rally, once said, “I love the poorly educated,” for which he was predictably criticized by the highly educated.)

    • Replies: @anonymous
    , @Logan
  34. @res

    Davis’ excellent (!) book “Storm over Biology” is so much to the point, that it hurts, almost. It is a book, which will be read two hundred years from now. For reasons I don’t understand, it is not mentioned in Bernard D. Davis wikipedia article. I should be though. As I said: Excellent.

    As an aside: Davis is Lithunian. That might be the reason, he digs Kant. Kant is the shortcut here through all kinds of Goulds confusions.
    Gould’s thoughts could be reconstructed in the same path, in which Jürgen Habermas critizises Marx, – – on the shoulders of Kant, nota bene.

    Davis’ term moralistic fallacy, which he coined in a short (!) article in Nature sums Gould’s zeigeist errors up in – – – – – – – two words!

    Incredible – but such is the beauty of thinking in the right way!

    PS

    Jordan B. Peterson is on the same track as Davis in criticizing the postmodernists, who basically trot along in the same neo-Marxian paths as Gould does. Thing is a little bit confusing because there are two kinds of postmodernists and neo-Marxists: Kantian ones and – – – the others, so to speak. All things related to the theory of the sciences can be properly solved with the help of Kant’s separation of the moral, the counting and the (aesthetically) judging spheres, which are outlined in his three critiques.

    PPS
    res – the fun fact is: When I thought of Davis’ book while reading your post, I wondered, who recommended it to me – now I’m almost sure, it was you! – So, thank you!

    PPPS

    Steve Sailer could make very good use of Davis’ Nature article about the moralistic fallacy. He is at this junction a myriad of times and could make very good use of this term of Davis: MORALISTIC FALLACY – or, as the happy few might say with a laughter: The Gould-fallacy.

  35. anonymous[845] • Disclaimer says:
    @obwandiyag

    ‘Wealth correlates with low IQ.’

    In that case I have a favour to ask. Apologies for my asking a favor unbidden but my uncle is a government administrator in England. He has recently come in to some money he cannot get access to because of a banking error and has asked me if I could help him out. All he needs is a bank account and would gladly pay to rent yours. Just send us your photographic identification, sort code, account number and pin and when he is done moving the 500000 you can have 5000.

    Many thanks in advance

    A high IQ peon.

    • Replies: @Franklin Ryckaert
  36. anonymous[845] • Disclaimer says:
    @niteranger

    ‘To put it another way it’s hard to beat a Magic Jew from Harvard when all the Magic Jews of academia have made him in to this SJW “God.”’

    Isnt it always the way?

  37. anonymous[845] • Disclaimer says:
    @Viral Architect

    ‘Gould makes his liberal readers feel good about themselves and the way they can understand the “complex” topics he is discussing, as he guides them away from the pernicious errors to which lesser intellects fall victim. But it was smoke and mirrors in typical stale-pale-male fashion’

    Stale pale male?

    How about typical semitic charlatan?

  38. anonymous[845] • Disclaimer says:
    @InSearchOfTruth

    High verbal iq goyim

    • Replies: @Wizard of Oz
  39. anonymous[845] • Disclaimer says:
    @cfhosford

    ‘I would urge those on this blog who know how to edit Wikipedia entries, and also have a knowledge of the science, to please update the main Stephen J. Gould Wikipedia entry with the appropriate countervailing data that details his mistakes (both political and scientific) in “The Mismeasure of Man.” The site is here: ‘

    Dobt bother. On certain topics of extreme importance the gatekeeping is too strong.

  40. smokey says:

    Historically, smart, hyper superior, super-brilliant elite have often been targeted for elimination in bottom up political revolutions? Has anybody done a study on that question?

    • Replies: @Logan
  41. anonymous[845] • Disclaimer says:
    @Hail

    ‘We now have a new, weird little form of social-class division distinguished by education level. In other words, many of the educated see themselves as members not “of” the nation, but members of a Brahmin-caste-like cognitive elite social class that is more important than that rabble out there called the Nation. Parallels with the arrogance of 18th-century France’s aristocracy suggest themselves.’

    Jesus wept dude theyre called jews. The French aristocracy was significantly judaised as money became the main avenue for promotion.

  42. @anonymous

    Sounds suspiciously “Nigerian” !

  43. Gould’s real motivations for his fraud has already be dealt with by Kevin MacDonald, here : Stephen Jay Gould’s Jewish Motivation : Natural selection by any other name, The Occidental Observer, May 21, 2012

    https://www.theoccidentalobserver.net/…/stephen-jay-goulds-jewis…

  44. padre says:

    I always wondered, how an mediocre person can measure the intelect of an genius!

  45. Bruno says:

    Most Gafa billionnaires have IQ above the top percentile (135, sd15) with a correlation around 0.5.

    If you’re in a 1 in 1 million average level in wealth, with a 0,5 correlation between wealth and intelligence, the average IQ will be at 1 in one hundred (135). It would have been 172 if the correlation were perfect. Knowing that Harvard college students have been tested at 125 average IQ by Pederson (when he studied creativity) or 1 in 20, despite selecting at 1 in 200 in SAT, you see how predictive is IQ.

    If you explain 25% of variability, you ve done a pretty good job. Taleb is a bullshiter with sporadic hindsights.

  46. @Ron Unz

    Thanks for reminding me. I have benefited from reading it again. How well Trivers writes! Everything is sharp, and in its place.

    • Agree: annamaria
    • Replies: @Bruno
  47. Bruno says:
    @James Thompson

    The French wiki about Trivers says he joined Black Panthers out of resentment after Harvard’s president Bok postponing his decision about giving him tenure.

    In Crimson 1979 : « He left Harvard after President Bok rejected the Biology Department’s April 1977 recommendation to grant him tenure and postponed a final decision. »

    If true, this would be a case of hard left intellectual (same for hard right) created out of resentment. It seems to be a pattern. Some failure is a good thing for activism and another source for so called « culture of critique » .

    • Replies: @Ron Unz
  48. #DrJamesWatsonIsRight

    #DrFrancesCrickWasRight

  49. @res

    I agree with your surprise at Dr T’s modest uninflammatory “For me the most interesting result,,,”. Hasn’t it been obvious to common sense for decades? I think Kingsley Amis made “more means worse” famous from 1960.

  50. Baron says:

    Michael Behe has a new book out, Darwin Devolves: The New Science About DNA that Challenges Evolution, something to look forward to.

  51. Art Deco says:

    He wrote a trade book on a subject distant from his actual expertise. If that book actually has been cited 10,000 in academic literature, dosn’t say much for the people producing the citing literature.

    • Agree: res
    • Replies: @res
  52. @res

    A pity that Taleb can’t learn from Gould to write good clear readable English. I enjoyed a book by Gould which I can’t for the moment lay my hand on or recall the title of though I recall that it said some interesting things about phenomena which exhibit limits whether in sporting prowess or evolutionary development. However The Mismeasure of Man never grabbed me although I think it may be there that he scored marks for his praise of Darwin’s reaction to slavery in Brazil. Curiously, a famous lefty broadcaster of my acquaintance agreed that, as an interview subject he was a bore.

    • Replies: @res
  53. @Ron Unz

    Like to give another try to getting up a Lysenko Prize in honour of SJ Gould awarded to those that should have known better or maybe those that surely did? I sometimes claim credit for the idea but I think it was probably Steve Sailer’s.

  54. Lyndon LaRouche died just recently. Can’t we get a write up on him? Much more interesting than this fellow I’d never heard of and got no interest in.

    • Agree: Hail
  55. What was his motive for attacking “intelligent tests?” You need to spell it out!

    • Replies: @James Graham
  56. Poco says:
    @obwandiyag

    That’s why little nigerian girls in England should go back to nigeria. They have a wakanda to build.

  57. Poco says:
    @James Thompson

    At some point he didn’t score very well on an I.Q. test. Hence the anger.

  58. res says:
    @Wizard of Oz

    A pity that Taleb can’t learn from Gould to write good clear readable English.

    Agreed, but I think part of the explanation is below.

    Part of Taleb’s shtick is to baffle people with pseudo-math (complicated looking but lacking formality and correctness).

    Gould’s shtick was to ignore the math (or get fairly basic math wrong) and captivate people with his words.

    The latter requires the skill you note, the former does not (actually requires more of the opposite, obfuscatory text as well as math).

    I found The Mismeasure of Man a good read except for the parts where I wanted to throw it against the wall because of all of the BS. Unfortunately the latter case dominated the experience. But I can see how true believers or those with non-functional BS detectors would find it compelling.

    • Replies: @Wizard of Oz
  59. res says:
    @Art Deco

    Here is the Google Scholar citation page for The Mismeasure of Man:
    https://scholar.google.com/scholar?cites=12567137434848364774

    It shows about 11,300 results. Many of those are books and I don’t see a way to separate out the books and academic papers.

    One exception to “doesn’t say much for the people producing the citing literature.” People like Dr. Warne debunking the academic validity of that trade book.

  60. Rogue says:
    @Alfa158

    Exactly the conclusion I reached several years ago.

    Totally agree.

  61. JNDillard says:

    In “The Blank Slate,” Steven Pinker slices and dices Gould’s attack on neo-Darwinian evolutionary theory.

  62. Anon[140] • Disclaimer says:

    Stephen J. Gould is a Jew, what do you expect? They’ve long taken upon themselves to be the protector of blacks and browns, who they know are just too dumb to trump up the con themselves. Jews need their alliance against whites.

    Trump loves the Jews, and the Jews are going to burn him at the stake. In addition to Rod Rosenstein and Jews who dominate the media, academia, Congress, Silicon Valley, judiciary benches and Hollywood, his personal lawyer Michael Cohen has just completely ripped him to shreds and gave the other Jews cause to impeach him — so much for attorney-client privilege, and just wait til’ his accountant Allan Weisselberg does his bit, he’ll finish Trump off.

    Moral of the story: Never trust a Jew.

  63. @Reuben Kaspate

    “What was his motive for attacking “intelligent tests?” You need to spell it out!”

    Done:

    “I am hopeless at deductive sequencing…I never scored particularly well on so-called objective tests of intelligence because they stress logical reasoning… ”

    Source: March 29, 1984 issue of The New York Review of Books

  64. I’m surprised that the Socialists/Communists in the USSofA have not introduced Lysenkoism into the college curriculum.. Maybe itz just a matter of time.

    “Lysenkoism was a neo-Lamarckian idea, claiming that in crop plants, such as wheat, environmental influences are heritable via all cells of the organism. Lysenkoism was applied to agriculture during the Stalin era with disastrous consequences”

  65. @res

    “For comparison, a pre 1989 ACT composite of 29 was accepted for Mensa (IQ ~130):
    Currently an ACT score of 33 is 98th percentile: https://www.manhattanreview.com/act-percentiles/
    An ACT score of 23 (per equation, IQ 103) is 69th percentile.
    Given that the ACT test taking population is selected (likely to have mean IQ > 100) I find those results inconsistent”

    In my state the ACT test is/was taken by all public school students. Of course the mentally challenged and dropouts are not tested. In 2010 the Chicago city average was 17 and state avg 20. Some of the selective enrollment schools avg 28. I think that the ACT has been re-normed sometime in the 1990’s. Back in 1960’s the only students who took the ACT test were students who wanted to go to college. A 21 ACT score back then would get you into the state’s flagship university. Of course that score would not get you into studies like engineering, math, physics.

    So my guess is that an ACT score of 21 equated to today’s score of around 25 or 26.

  66. Wally says:
    @obwandiyag

    So that’s why Nigeria is leading the world.

  67. Wally says:
    @Alfa158

    Spot on.
    Ron Unz has handled that as well.

    How Israel and Its Partisans Work to Censor the Internet: http://www.unz.com/article/how-israel-and-its-partisans-work-to-censor-the-internet/?highlight=wikipedia
    and an admission:
    Zionist Wikipedia Editing Course: http://www.israelnationalnews.com/News/News.aspx/139189

  68. Logan says:
    @smokey

    First thing we do is kill all the lawyers.

    • LOL: Wally
  69. Logan says:
    @Hail

    We now have a new, weird little form of social-class division distinguished by education level. In other words, many of the educated see themselves as members not “of” the nation, but members of a Brahmin-caste-like cognitive elite social class that is more important than that rabble out there called the Nation. Parallels with the arrogance of 18th-century France’s aristocracy suggest themselves.

    I quite agree. And it’s not only the level of education but where you acquire that education. One of the main sneers at Sarah Palin was that she didn’t attend an elite college. To my mind these sneers were amazingly openly class-based. She wasn’t a member of the class that deserved to rule. Very much indeed like the 18th-century ancien regime.

  70. Anon[379] • Disclaimer says:

    Gould the liar got it wrong, but his books are selling on Amazon.

    Amazon is taking out all the dissident right books.

    Amazon sells tons of Neocons books defending Nakba and Wars for Israel.
    It sells tons of ugly rap music that celebrates violence. It sells tons of Hollywood movies where Muslims are subhuman terrorists. But no more books from Counter-Currents.

    Someone should set up a new book business online called BANNED BOOKS.

  71. @niteranger

    “To put it another way it’s hard to beat a Magic Jew from Harvard when all the Magic Jews of academia have made him in to this SJW “God.”

    Jeez. I liked the rest of your comment.

    I admired and recommend elsewhere on this thread Bernard Davis’ s severe criticism of Gould. Davis had Jewish parents. https://en.wikipedia.org/wiki/Bernard_Davis_(biologist)

    Gould’s major fan boy, Niles Eldridge, on the other hand, is as Jewish as I am, i.e., not at all.

    My own negative views of him are in my book.

    https://www.amazon.com/Cancer-Selection-New-Theory-Evolution/dp/0963024205/ref=sr_1_1?keywords=cancer+selection&qid=1551314554&s=books&sr=1-1

  72. @obwandiyag

    Are they smarter than Nigerian personal trainers?

  73. Ron Unz says:
    @Bruno

    The French wiki about Trivers says he joined Black Panthers out of resentment after Harvard’s president Bok postponing his decision about giving him tenure.

    In Crimson 1979 : « He left Harvard after President Bok rejected the Biology Department’s April 1977 recommendation to grant him tenure and postponed a final decision. »

    If true, this would be a case of hard left intellectual (same for hard right) created out of resentment. It seems to be a pattern. Some failure is a good thing for activism and another source for so called « culture of critique » .

    Actually, I think it’s the other way round. From what I recall, Trivers’s great fondness for the Black Panthers and other far left racial groups was a central factor in Bok’s decision to block his academic tenure.

  74. @res

    Thanks for a good explanation of their respective presentations.

  75. @cfhosford

    Don’t even try, the SJW zealots at wikipedia will almost immediately delete anything that doesn’t rigidly conform to their holy narratives.

    If you want a balanced alternative to those evil fanatics, try the following:
    InfoGalactic

    https://infogalactic.com/info/Main_Page

    It was started by simply copying the content from wikipedia, then allowing edits from the user community without any of the miserable leftist gatekeeping.

  76. @Alfa158

    Wiki is not just edited and patrolled by SJWs. It is policed by a large group in Europe and Israel (Haifa) that is funded by Gregoryi Schwartz.

  77. Oh, come on. Gould did not “get it wrong”. He set out with a political and social axe to grind. He is not a professional of any sort, except perhaps a professional hack.

  78. @aandrews

    A perhaps not irrelevant factoid: The editor of Natural History was a boyhood pal of SJG.

  79. utu says:

    I have not read The Mismeasure of Man until yesterday. I was encouraged to read it by negative opinions about the book and its author by people who I found on many occasions to be wrong. I was not disappointed. Gould’s arguments are sound. He gets a very good account of what was in the minds of Binet (positive), Spearman (mixed), Burt (negative), Thurstone (mixed), Jensen (mixed). He fairly well explains the factor analysis but he could have done it more forcefully if he used mathematical algebraic arguments. Then he could achieve a complete démontage of this technique.

    • Replies: @res
    , @CanSpeccy
  80. When I was much youger (about 25 years ago), I was reading four of Gould’s books: “Brontosaurus” and “Wonderful Life”, and two volumes with collected essays. Back then, I liked his style and thought of him as a “scientist”. Now, that I know a little bit more about science, I have a different view. I suppose he was a solid Paleontologist, but in all other fields he wrote as an “advanced amateur” (at best), and he was never especially honest.
    He also committed the ultimate sin as a writer in poular science: he did not try to present different views / opinions / theories to people who are interested, but not experts in the field, but used his title and influence for shameless propaganda for his own (and his fellows’) ideas.
    He had also no problem to devaluate work from other scientist from a different era, like Charles Walcott. Walcott was in his 60’s and 70’s, when he collected literally ten-housands of fossiles in the early 20th century and classified endless numbers of them, just to be presented as a numb guy in Gould’s “Wonderful Life” (BTW, the theories that Gould presented for the evolution of the pre-cambrian fauna are now probably as outdated today as Wallcott’s classification from 60 years earlier).
    Gould also had his personal vendetta against Teilhard de Chardin, whom he accused of complicity in the “Piltdown-swindle”, based on some really, really lousy Sherlock Holmes-style “investigations” he made. This defamation (as laughable as it was) even made it into a stage play, and I find it most likely that the author got his “knowledge” from reading Gould.

  81. res says:
    @utu

    Wow. I guess I overestimated you. Oh well.

    • Agree: Dan Hayes
  82. World map color-coded by IQ and Income

    … using data from “IQ and the Wealth of Nations”, a 2002 book by psychologist Richard Lynn and political scientist Tatu Vanhanen.

  83. @res

    The strong Robert Trivers sentence about Gould here for me is: “Not only was his science hopeless (…).”

    As an aside: Habermas always very carefully separates the roles of the public intellectual and the scientist. He thinks this is a necessity since these are two different realms with different purposes and therefore different ways to argue / to behave.

    Now here are two interesting aspects: It wasn’t clear to me that Gould had only a very weak scientific standing altogether. So – all there is left in his case is the public intellectual in the disguise of a scientist.

    PS
    That Lewontin even admitted that he indeed did lie at times about scientific matters was new to me too.

    • Replies: @res
  84. res says:
    @Dieter Kief

    I am not familiar with Habermas (but see you also mentioned him above). I see his wiki page at https://en.wikipedia.org/wiki/J%C3%BCrgen_Habermas
    Can you recommend any of his work in particular?

    The point you (he) make(s) about “carefully separates the roles of the public intellectual and the scientist” is a good one. I would be interested in seeing a longer exposition of that.

    Thanks.

    • Replies: @Dieter Kief
  85. CanSpeccy says: • Website

    Gould’s book made a number of assertions. Two that stuck in people’s minds were: that measures of brain size derived from the study of skulls of different races had been biased …

    Still sticking to the Rushtonian nonsense about brain size as a measure of intelligence, eh.

    The problem with that is exemplified by the fact that 90% of men have a larger brain than the average woman, yet men and women have similar IQ’s.

    What that means is that if brain size has any relation to intelligence it has to be scaled to whole body mass (adipose-free) using the appropriate allometric scaling constant — something none of the Rushonians seems to have a large enough brain to accomplish.

    • Replies: @Franklin Ryckaert
  86. CanSpeccy says: • Website
    @utu

    He (Gould) gets a very good account of what was in the minds of Binet (positive).

    Yes. And here’s what Binet said about his intelligence scale, the first intelligence test:

    The scale, properly speaking, does not permit the measure of intelligence, because intellectual qualities are not superposable, and therefore cannot be measured as linear surfaces are measured.

  87. @CanSpeccy

    I cannot imagine that Philippe Rushton was not aware of the brain-to-body-mass-ratio :
    https://en.wikipedia.org/wiki/Brain-to-body_mass_ratio

    • Replies: @res
    , @CanSpeccy
  88. res says:
    @Franklin Ryckaert

    CanSpeccy likes going on about areas where he has little knowledge and apparently no inclination to check first before making glib assertions. Let’s take a look at what Rushton had to say about brain size and body mass. This looks a good starting reference: Whole Brain Size and General Mental Ability: A Review
    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2668913/

    There is disagreement about whether or not brain size should be corrected for body size when examining brain size/GMA correlations. Controlling for body size changes the question from “Is IQ correlated with absolute brain size?” to “Is IQ correlated with relative brain size?” Although these are quite different questions, evidence shows that the answer to both is yes.

    Much more there for anyone who is interested, but hopefully that makes clear the correctness of this gem from CanSpeccy:

    What that means is that if brain size has any relation to intelligence it has to be scaled to whole body mass (adipose-free) using the appropriate allometric scaling constant — something none of the Rushonians seems to have a large enough brain to accomplish.

    If one is going to hand out insults it is good to make an effort to be correct. Glass houses and all that.

    • Replies: @James Thompson
    , @CanSpeccy
  89. CanSpeccy says: • Website
    @Franklin Ryckaert

    I cannot imagine that Philippe Rushton was not aware of the brain-to-body-mass-ratio.

    Well that tells us something about your imagination, or lack thereof, but not about what Philippe Rushton was, or was not, aware of.

    • Replies: @Franklin Ryckaert
    , @res
  90. @CanSpeccy

    My “imagination” was that Rushton must have been too intelligent to have overlooked that question. And that appears to be right. See comment of res above.

  91. Intelligence tests were meritocratic, not aristocratic.

    That is wrong. There is no merit in being born more intelligent than others. And there is little or no merit in being raised intelligent, if that is possible, and pertinent to the case in hand.

    On the other hand, the word “aristocracy” means, etymologically, “rule of the best”.

    So, intelligence tests are definitely not meritocratic, and are closer to being aristocratic, if you take that word in an etymological sense.

  92. @res

    I discussed this with Rushton at a ISIR conference, I think in 2007, and he said that he though the anomaly between large sex differences in brain size but small sex differences in intelligence was that much of the larger male brain surplus dealt with visuo-spatial tasks. The implication (we did not actually cover this in detail, but probably both assumed it) was that if one were to extend intelligence tests so that they had more visuo-spatial items, and more difficult ones, then the Rushton/Ankey paradox would be resolved.
    The same argument would apply to general knowledge, especially of technology, mechanics and so on where there is probably greater male ability, though those items are dropped because they show male advantage which is interpreted as a bias in the testing.
    I think that in sex difference research, as in race difference research, one should always use the very broadest range of possible tests, so as to give every group being compared as much chance to shine as possible. That is to say, one should reject nothing on the grounds of possible bias, but put everything into the mix that increases the predictive power of testing when evaluated against real life achievements.

    • Replies: @CanSpeccy
    , @res
  93. @res

    My Habermas abstract:

    Science follows strict rules. The counting/ measuring (=objective) sciences flourish because they do follow these rules. Their purpose is to advance knowledge and avoid harm.

    The public intellectual has to speak (or write) clear, honest and non-polemic. His purpose is to influence people, not (!) to gain power. As soon, as an intellectual has become part of the functional side of society, he has seized to be an intellectual. (Between Naturalism and Religion, p. 26 – in the German edition).

    If a scientist speaks in this role of a public intellectual, he is considering what should be done with (mostly: scientific) insights. In this role, he is like anybody else, in that he does not know the future and the consequences of certain measures taken, too. – He might be only better prepared, to make his judgments. Whether this is right or not, nobody knows, because – we all don’t know the future. It has to be tried out! Decisions have to be made in the light of all arguments, that are hitherto known. All reasonable (=sound / rational) arguments count. As an idealistic rule of thump, Habermas makes this proposition: Societal questions would be decided only if all aspects of a problem which count (all arguments pro and against a proposition) could have been brought forward in the decision making process. – Sigh: Habermas knows very well, that this idea can only function as a “counterfactual ideal”. Many a Habermas critique ignores this big caveat of the counterfactuality of Habermas’ “ideal speech situation” (“Theory of Communicative Action” p. 532). This idea is not here to be fully realized; it is here to remind all proponents and opponents in a certain decision-making process what is really at stake: That a decision-making process is not only about who will win but always about the question: Was the victory sound – especially with regard to the opponents – and even with regard to those, who did not speak out at all.

    Democracy is not only about winning but – with Kant’s Moral Imperative – also about responsibility: The willingness to accept, that decision making is necessarily an open process and open means not only open to success but also open to failure. – This is a reason, by the way, why Habermas is reluctant, to condemn right away what people have done in the past. For example: Heidegger’s big mistake in Habermas’ eyes is not, that he tried to “lead” Hitler, but that he found no words of regret for his error – not even after the second world war.

    But – Heidegger’s political errors did not prevent Habermas from acknowledging, that Heidegger indeed was the major philosophical force of the first half of the twentieth century.

    PS

    Davis does make references to Kant here and there. I’ve looked it up. By and large, Davis says the same things about the role of the public intellectual than Habermas: Davis insists, that there is an objective side to the measuring (=natural) sciences, and that the wider American public and the Marxists both have made a foolish move away from this foundation of the scientific truth in establishing what Davis calls an “externalist” approach – via Marxism, but also via western attempts in the history and sociology of science.

    Natural sciences have become so complex and difficult to follow from outside, says Davis, that many a sociologist and historian of the sciences “have attempted to strengthen the similarity of their field to the natural sciences by denying that either could be truly objective.” p. 12 in “Storm Over Biology”).
    (Davis in 1986!!)

    Thus Davis holds, that the broader public needs “education in the scientific mode of thought, with its deep respect for objectivity.” And he concludes: “Without such an anchor, to a common reality, and with more emphasis on subjective values, we are more likely to drift apart. If we choose to think only with our hearts, however sincerely, we will be approaching our conflicts of interest on the basis of emotionally charged beliefs; and this can lead to what has been called thinking with our blood.” (p. 25 Storm Over Biology).
    (Now try not to think of the “coalition of the fringes” and PC and all that…)

    PS
    Davis writes of “conflicts of interest” – Habermas’ “Knowledge and Interest” – his second ground-breaking book, appeared in Englisch in 1971. – Davis might have known Habermas’ writings – or might have heard from colleagues, who knew them, or he might have been just “a clear mind” (Jackson Browne – “It takes a clear mind to make it!”). I’m curious, though – and fascinated, I have to admit, by the fact, just how much Davis got right – right away – and that he could write it down so early on – and so clearly!

    PPS

    Davis about genes: “(…) we must not confuse the normative with the empirical.” (…) If we choose otherwise and suppress human behavioral genetics for fear that the results may contradict our assumptions, the costs may be high. For a major goal in the field, long emphasized by J. D. Haldane, is to help us to adjust educational procedures to individual differences in cognitive potentials and in patterns of learning.” (p. 32, Storm Over Biology)
    (Sounds a lot like Steve Sailer, doesn’t it?)

    • Replies: @res
  94. CanSpeccy says: • Website
    @res

    James Thompson, though he lacks the courtesy to acknowledge the fact explicitly, confirms my point that any effect of brain size on intelligence must be based on scaling with body size. However, neither you, nor James Thompson, nor Philippe Rushton seem to have grasped that, as with virtually all other body scaling relationships, the scaling is almost certainly not directly proportional. You should read a bit about allometrics before ridiculing those who have actually published in the field.

    • Replies: @res
  95. CanSpeccy says: • Website
    @James Thompson

    So Philippe Rushton “thought” that the anomaly (which is not really an anomaly if one understands how functional capacities scale with body mass), of sex differences in brain size in the absence of comparable differences in intelligence “was that much of the larger male brain surplus dealt with visuo-spatial tasks.”

    Unfortunately, this doesn’t really make any sense, does it, which is perhaps why, on the only occasion that Philippe Rushton was depicted in an editorial page cartoon in the (Toronto) Globe and Mail, he was shown to have a very small head.

    The same argument would apply to general knowledge, especially of technology, mechanics and so on where there is probably greater male ability

    As noted, the argument does not make any sense. In fact it is not an argument at all, merely a vague assertion of what has to be demonstrated. Perhaps you could put it in the form of an argument susceptible to evaluation.

    • Replies: @res
    , @utu
    , @utu
  96. res says:
    @CanSpeccy

    It was kind of you to make a comment confirming my point immediately after the one where I made it. Thanks.

    P.S. Yes, I realize you probably did not read my comment first given the timestamps. Still, reading the whole thread the two comments flow together nicely.

    • Troll: CanSpeccy
    • Replies: @res
  97. res says:
    @CanSpeccy

    With such great expertise perhaps you could provide actual support for your argument rather than just making an appeal to authority? Negative bonus points for that authority being yourself.

    You might very well know more about allometric scaling than I do, but don’t assume I know nothing. You specifically calling out adipose-free mass is one good example making clear you have thought about the issue. I hope it is clear that me noticing that says the same about me.

    Two thoughts.

    1. Deciding what mass (or whether other measures are more appropriate, like surface area and height, which Rushton also discusses in that paper) to use for the scaling is non-trivial. Measuring it (how common is data on adipose-free weight?) is another issue.
    2. Across a relatively small range like height (~2x for adults) and mass (~3x, absent extreme adiposity, which you want to correct for anyway, and malnutrition) most curves can be fairly well captured by a linear approximation (especially given most of the variance occurs away from the extremes). It also helps that mass, height, and surface area have exponential relationships with each other giving three chances at a decent linear fit.

    Rushton clearly took a thoughtful look at allometric scaling as shown by the paper I linked. Whether he is correct in all his points is less certain. You did not speak to his correctness though (and if you want to discuss that, please do, hopefully with references and thoughtfulness though). You made a comment that he and his followers were too small brained to properly consider it. Which I demonstrated is laughable. So I laughed and ridiculed.

    P.S. And you calling me out for lacking courtesy is hilarious. First, physician heal thyself. Second, I think it’s clear I understand that allometric scaling does not have to be (exactly!) directly proportional.

  98. res says:
    @CanSpeccy

    which is not really an anomaly if one understands how functional capacities scale with body mass

    Please do describe how that explains the M/F difference in brain size better than other proposed explanations. With data. That would be very interesting.

    Unfortunately, this doesn’t really make any sense, does it, which is perhaps why, on the only occasion that Philippe Rushton was depicted in an editorial page cartoon in the (Toronto) Globe and Mail, he was shown to have a very small head.

    Oh, I see. Ridicule in an editorial cartoon == scientific incorrectness. Never mind. Carry on.

    In fact it is not an argument at all, merely a vague assertion of what has to be demonstrated. Perhaps you could put it in the form of an argument susceptible to evaluation.

    Pot, kettle, black. It is amazing how many comments on unz.com are so simply explained by projection.

    • Replies: @CanSpeccy
    , @CanSpeccy
  99. res says:
    @James Thompson

    Thanks! I very much appreciate the looks at the people behind the papers. I (and I suspect quite a few other readers here) don’t have access to that side of intelligence research.

    The paper I linked was published in 2009. It looks to me like Rushton had considered the spatial hypothesis further, but did not find it decisive at that point (page 707).

    Unfortunately for this hypothesis, what little information there is from the two MRI studies to date suggests that brain size is not significantly related to results on purely spatial tests (such as mental rotation) in either men or women (Wickett et al., 1994, 2000). Yet in the same studies, brain size did correlate significantly with IQ. However, one of these studies looked at only women and the other looked at only men. It would be more informative to know what happens in a combined sample of men and women, since the hypothesis that the extra brain size relates to men’s better spatial scores would predict a correlation that should appear across sexes. So far, no comparison of brain size and spatial scores has been made in a mixed-sex group.

    What do you think? I don’t think it is inconsistent with Rushton having the opinion you describe, but no confirmation by 2009.

    He then concluded the section by considering that there is a M/F difference in IQ (as we have discussed here before) and appears to consider that hypothesis favorably. From page 708:

    The male and female g factors were congruent in excess of 0.99 and found to favor males with an effect size of 0.24, equivalent to 3.63 IQ points. The male advantage was found throughout the entire distribution of scores on the SAT, in every level of family income, for every level of fathers’ and mothers’ education, and for each and every one of seven ethnic groups.

    I suspect the difference in IQ profiles (M/V/S) does have something to do with the anomaly. I see no reason to assume equal brain mass/size is required to generate an IQ point in each different area. That idea might also help explain some of the variance of the within sex data (different M/V/S profiles within male and female groups). For that matter, why assume all IQ points require equal brain size? It seems perfectly reasonable to suspect the IQ/size relationship is nonlinear. IQ is normalized by population proportions not size. Though perhaps statistical averaging makes them correspond well after all?

    I do think CanSpeccy has a useful point with bringing allometric scaling into this. I just don’t think it justifies the ridicule he has been spewing and he has utterly failed to back up his point’s practical applicability to this case which makes it unsatisfying.

    Given the thoroughness with which Rushton looked at this issue (contra CanSpeccy) I agree with Franklin Ryckaert that Rushton failing to consider allometry is unlikely. Though I don’t see any way to prove that.

    Agreed about the desirability of breadth of skills testing. It is depressing that “bias” is used to eliminate items as you describe. Prove men and women are the same by first assuming that and then rejecting anything to the contrary. Sounds real scientific to me.

    P.S. An analogy for IQ/brain size that I just thought of and want to record for future reference. Engine size (cubic inches) or mass (pounds) vs. quarter mile times. I very much doubt those are linear relationships, and imposing a normal curve on the quarter mile times (as in IQ) would further complicate the relationship.

  100. CanSpeccy says: • Website
    @res

    Oh dear. Once again I set you off on a rant.

    Why don’t you take a walk, get some fresh air, calm down and then address the issues without the repeated personal attacks, and perhaps even provide some actual data.

    Also, perhaps you might leave it to James Thompson to address the points I raised in response to his comment, or do you feel it incumbent upon you to act as his ideological bodyguard?

    • Replies: @res
  101. CanSpeccy says: • Website
    @res

    With reference to my assertion that James Thompson’s argument is:

    In fact … not an argument at all, merely a vague assertion of what has to be demonstrated. Perhaps you could put it in the form of an argument susceptible to evaluation.

    you respond with some comment about pots and kettles. That is not an argument. It’s not even a statement of anything to do with the subject.

    Take another look. Here is the argument that was made:

    Philippe Rushton “thought” that the anomaly (which is not really an anomaly if one understands how functional capacities scale with body mass), of sex differences in brain size in the absence of comparable differences in intelligence “was that much of the larger male brain surplus dealt with visuo-spatial tasks.”

    In such form, it seems nothing more than hand-waving waffle. If you can put it into a coherent argument, let’s hear it.

    • Replies: @res
  102. res says:
    @res

    CanSpeccy likes to Troll flag me when he is losing an argument. He actually has a valid point. I AM trolling him. And rather enjoying it. But some relevant thoughts.

    First, CanSpeccy does quite a bit of trolling here himself. It is difficult to interpret comments like the following in any other way.

    Still sticking to the Rushtonian nonsense about brain size as a measure of intelligence, eh.

    something none of the Rushonians seems to have a large enough brain to accomplish.

    The problem is he utterly fails to back up his comments. Love or hate (or anywhere in between) my comments here it is hard to argue I fail to back them up. Which leads us to my second point.

    I think it is unjust to Troll flag me given my (useful content) / (trolling behavior) ratio. CanSpeccy’s corresponding ratio on the other hand…

    I use the LOL flag sometimes to call out arguments which I think are laughable, but I generally back it up with real arguments elsewhere in the same thread.

    IM(not at all humble, here)O Canspeccy’s use of the Troll flag combined with an utter failure to back up his own arguments is a laughably pathetic response to his being out-argued.

    CanSpeccy, if you don’t want me to troll you then leave off with the unprovoked insults and try including an occasional substantive argument and/or reference. We have talked about “tit-for-tat” multiple times in this blog. That is not what you are doing. Despite your claims to the contrary.

    As I have observed before, you are a smart and knowledgeable man. It would be much more enjoyable to have your intellect, knowledge, humor, and rhetorical ability used to participate in interesting conversations.

  103. res says:
    @CanSpeccy

    Still no real arguments. LOL.

    I choose appealing targets as I like. If you happen to make a hypocritical comment to someone else I consider that an appealing target. I don’t like hypocrisy.

    I suppose I could restrict myself to responding to your comments to me though given that you provide SUCH a target rich environment.

    And calling me out for not addressing the issues?! I think I do need to take a walk so I can stop from laughing so hard.

  104. res says:
    @CanSpeccy

    you respond with some comment about pots and kettles. That is not an argument. It’s not even a statement of anything to do with the subject.

    It is an oblique way of saying you are being a hypocrite. Not everything I write has to be a coherent argument. How about you try addressing the parts of my comments which are?

    If you were paying attention and/or had decent reading comprehension ability you would have realized I was not affirming Dr. Thompson’s statement (he is more than capable of doing that himself, if he likes). I was criticizing yours.

    I provided an elaboration of his statement (from Rushton’s paper) in a response to him. If you actually care about coherent arguments how about you read that, then read the paper, then say something even moderately intelligent about it?

  105. res says:
    @Dieter Kief

    Thanks! (and additional thanks for giving me an opportunity to end this session here on a positive note)

    Science follows strict rules. The counting/ measuring (=objective) sciences flourish because they do follow these rules. Their purpose is to advance knowledge and avoid harm.

    The public intellectual has to speak (or write) clear, honest and non-polemic. His purpose is to influence people, not (!) to gain power. As soon, as an intellectual has become part of the functional side of society, he has seized to be an intellectual. (Between Naturalism and Religion, p. 26 – in the German edition).

    That was a very helpful synopsis. I find the “avoid harm” qualification interesting. Not least because it introduces a large loophole. Many SJWs would argue they are acting in the spirit of that directive when they try to censor science.

    I find the “influence people” vs. “gain power” dichotomy interesting. Need to think about it more. Do you have any specific references expanding on that idea?

    (BTW, your English is excellent, but there you mean “ceased” rather than “seized”, homonyms are hard enough in one’s first language, not to mention second or more)

    I like this point:

    That a decision-making process is not only about who will win but always about the question: Was the victory sound – especially with regard to the opponents – and even with regard to those, who did not speak out at all.

    The rest of your comment was very interesting as well. Thanks again!

    • Replies: @Dieter Kief
  106. utu says:
    @CanSpeccy

    “much of the larger male brain surplus dealt with visuo-spatial tasks”

    It did not occur to the jokers that this statement is an admission that there is more to intelligence than one factor g. That there must be another factor responsible for the visuo-spatial tasks. Is this factor captured in IQ score? If it is, then a men and a woman with the same IQ score must have different g’s and man’s g must be lower.

    The IQism as it is is a para-science of ad hoc confabulations. They lie as they go to maintain the dogma.

    • Replies: @AaronB
    , @res
    , @CanSpeccy
  107. utu says:
    @CanSpeccy

    In fact it is not an argument at all, merely a vague assertion of what has to be demonstrated.

    This is what para-science does.

  108. AaronB says:
    @utu

    Lol, that was exactly what I thought 🙂

    I was like – you just undermined your arguments for g!

    I was gonna respond, but they have been arguing like this for so long I figured its pointless. They’re gonna do what they’re gonna do.

    Intellectually dishonest people argue like this, now one argument, now it’s exact opposite, depending on the context.

    Not that anything I or you say will change them. Still, its amusing to see.

    • Replies: @CanSpeccy
  109. res says:
    @utu

    They lie as they go to maintain the dogma.

    That would be better used as a description of your strawmen. For example:

    an admission that there is more to intelligence than one factor g

    Do you really not understand that I (and I think most of the people you mean) believe both in different abilities and an overall factor (g)? That follows pretty easily from the idea that there is variation in specific anatomical areas (e.g. the visual cortex) as well as in overall factors (e.g. neuron branching, myelination, metabolic efficiency, etc.).

    The projection. It burns.
    https://www.urbandictionary.com/define.php?term=it%20burns

    P.S. And AaronB agrees and amplifies. Perfect. LOL!

    • Replies: @AaronB
    , @AaronB
  110. AaronB says:
    @res

    Ok, so then surely you will admit that skull size does not correlate with g, but only with certain specific factors of intelligence, at least according to Thomsons anecdote. So skull size does not correlate with “general intelligence”.

    And further, you must then surely admit that although the general factor of intelligence exists, it gives a highly incomplete picture of a persons cognitive ability, and is of some, but limited, value in determining if a person will excel at a certain subject.

    And in light of this, in your subjective opinion, do you feel that the way g is commonly spoken of, or at least often spoken of, by laymen, suggests that it is a common belief among many people that g provides a nearly complete picture of a persons cognitive ability?

    And that if this is so, psychometricians have done a poor job in communicating their findings, and should expend effort towards making clear that the importance of g is vastly overrated in the popular mind, and that specific factors play a huge role in determining where a person will excel.

    • Replies: @res
  111. AaronB says:
    @res

    For instance, res, and this is a serious point that I ask you to really grapple with, I have heard that Jewish IQ breaks down to slightly below 100 on on spatial, and as high as 120 on verbal.

    What kind of a general factor of intelligence can permit such a wide discrepancy on specific abilities? Or better put, even if g is a statistical fact, doesn’t such a wide discrepancy between factors make it of very limited significance in the real world?

    I would appreciate a serious answer please.

    • Replies: @res
  112. res says:
    @AaronB

    I would be interested in the source of your numbers (it would help me think you are serious as well). They seem in the ballpark of what I have heard (we had an extended discussion about that here recently, not sure if you saw that), but a bit on the extreme side.

    The scenario I outlined above in the comment you responded to:

    That follows pretty easily from the idea that there is variation in specific anatomical areas (e.g. the visual cortex) as well as in overall factors (e.g. neuron branching, myelination, metabolic efficiency, etc.).

    allows that quite easily. Specific areas are the individual abilities and the overall factors compose g.
    Just assume Jews experienced selection on verbal to a greater degree than on spatial. That would probably result in improvements on g and verbal-specific, but stasis (or a decline) for spatial-specific. If other groups were experiencing more selection on spatial skills at the same time that would increase the differences between the groups.

    It is worth emphasizing that in this view measured spatial ability depends on both the spatial-specific ability (anatomical?) and g (overall system goodness).

    Seems pretty reasonable to me. Hopefully at some point all of the “anti-IQist” strawmanning directed at me will stop, but I’m not holding my breath at this point.

    P.S. Serious enough? Please return the favor.

    P.P.S. It is worth noting that this is an area where I may disagree with the psychometric community at large (some discussion of that in that other thread, but I would be interested in more). I think the M/V/S structure of g is likely to vary between populations. The key question to my mind is by how much?

    • Replies: @res
    , @AaronB
  113. res says:
    @res

    I think that is worth expanding on at more length. Here’s a rough cut at what I believe, Note that where I write things like M = m * g if would be more accurate to say M = f(m, g). But IMHO the former provides a simpler and more effective visualization and is easier to discuss.

    First, a definition of terms.

    g – basically what I think it has always meant conceptually, overall goodness of brain function, different groups may have different averages of this
    G – g filtered through the lens of a given group, measured g, IQ as defined by looking at a given population, group averages may differ in both direction and magnitude, this is a vector, all of the other variables are scalars (at this level of modeling)
    m – math specific ability (e.g. special anatomy improving math ability)
    M – measured math ability
    v – verbal specific ability (e.g. special anatomy improving verbal ability)
    V – measured verbal ability
    s – spatial specific ability (e.g. special anatomy improving spatial ability)
    S – measured spatial ability

    For different groups we will add numerical suffixes. So G1 would be measured G of the first group.

    Since most studies are on European whites let’s call them group 1 giving.

    G1 = first PC of M1, V1, and S1
    M1 = m1 * g1
    V1 = v1 * g1
    S1 = s1 * g1

    And further define all of those variables (except G1, that is more complex) as 1.0 since this is the reference group.

    Then let’s call Ashkenzai Jews group 2 and spitball some numbers.

    m2 = 1.0, v2 = 1.1, s2 = 0.9, g2 = 1.1

    I’m not going to do all the math here, but that gives numbers in the ballpark of observational reality. It would also result in G2 being different from G1 (I should check by how much, the math is easy to do in R).

    Hopefully this gives a more concrete feel for what I am trying to say. Any thoughts, anyone?

  114. res says:
    @AaronB

    Ok, so then surely you will admit that skull size does not correlate with g, but only with certain specific factors of intelligence, at least according to Thomsons anecdote.

    That’s an interesting point. I think it has significant truth in it (much brain functioning “goodness” is not size related, e.g. electrolytes and action potential propagation velocity, metabolic efficiency) but is not completely true. Two reasons
    – Changes which increase the number of brain cells uniformly could increase both g and brain size.
    – Changes which increase brain cell volume uniformly (e.g. increased myelination) could increase both g and brain size.

    So no, not only don’t I admit that. I disagree with you.

    And in light of this, in your subjective opinion, do you feel that the way g is commonly spoken of, or at least often spoken of, by laymen, suggests that it is a common belief among many people that g provides a nearly complete picture of a persons cognitive ability?

    Given that I disagreed with the first part, I don’t think I really need to answer this. But what the heck.

    I honestly don’t know. This is one of those cases where I think most laymen manage to hold two contradictory ideas in their mind at the same time. One is significant buy-in to what you say (what I would refer to as the classic IQist strawman). The other is a realization that that picture is incomplete.

    I really don’t have a good sense of what other people think deep down inside. Externally there are different themes that manifest.
    – IQ says quite a lot about a person. A common notion of a smart person and rough agreement over who qualifies.
    – Reasonable and/or resentful blowback about that resulting in an overemphasis on specific skills.
    – Quite sensible observation that the best people in a variety of intellectual areas tend to be different people. The best mathematicians are not the best novelists. Though both likely have fairly high IQ.

    Arguably the first two of those correspond to the “IQist” and “anti-IQist” ideas we see mentioned here.

    And that if this is so, psychometricians have done a poor job in communicating their findings, and should expend effort towards making clear that the importance of g is vastly overrated in the popular mind, and that specific factors play a huge role in determining where a person will excel.

    Communication is a two way street. And it is especially hard to communicate complex ideas. Is it the psychometrician’s fault when he (or she) is asked for a simple explanation and gives one? I would say the media and educational institutions are much more to blame here than psychometricians.

    I also disagree with your assessment of the relative magnitudes of the importances of g and the specific factors. It is important to remember that different forms of screening have different purposes. The US Army AFQT provides a good example for discussion. There is a minimum threshold (commonly stated as ~85 IQ IIRC) then various subtests which rely on specific abilities for fine placement.

    I think restriction of range is what makes you so sure you are right here and so vehement about it. If you only look at people in the IQ range of 110-125 (call that competent college students) the specific abilities matter a good bit (although arguably the difference between the extremes of that range might account for more variance). Basically that initial screen (the >125 IQ group is more off at the elite colleges, not perfect sorting of course) uses up much of the predictive power of IQ thus making specific skills more important in that context. Also notice that given that sorting the best people at specifics will probably be the more unbalanced ones. For example, the 110/130 M/V and 130/110 M/V people rather than the higher IQ 125/125 person.

    I hope how you can see how the sorting in our educational system and the resulting restriction of range can so easily result in perceptual errors.

    P.S. Thanks for engaging nicely and making for an interesting conversation!

    • Replies: @AaronB
  115. CanSpeccy says: • Website
    @utu

    Yes, the idea that male brains are better than female brains because larger, the extra mass resulting in greater skill at visuospatial tasks despite no difference in IQ, indicates either that skill at visuospatial tasks is not reflected in IQ score, or that women are better than men in some other respect that is also reflected in IQ score.

    A full explanation for the size difference between male and female brains despite similarity in IQ thus requires evidence of the area(s) in which female brains, despite their inferior size, are superior to male brains. But we will, I suspect, wait in vain to learn here in what way the female brain, though smaller, is necessarily superior to that of the male. (Maybe it can be inferred from the fact that women seem less inclined than men to engage in debates about IQ.)

    But in any case, the notion that male and female brains differ in relative capacities despite the same IQ, confirms that IQ provides not a measure of some pure essence of intelligence, but only the summation of scores on tests for different types of mental ability that are not, as Binet put it, “superposable”. Likewise the result of a decathlon: it does not indicate that winner is necessarily the best at running, jumping or throwing, only that they achieved the best overall score.

  116. CanSpeccy says: • Website
    @AaronB

    the strange thing about psychology is its seeming isolation from mainstream biology. Thus, for example, in discussing brain size and intelligence, there is little apparent interest in why brains differ in size. There is just the simplistic inference that if there is a difference in head circumference then, hey, there must be a difference in brainpower.

    Yet, not only was Einstein’s brain of normal dimensions, so also was Carl Gauss’s. Indeed some of the deep thinkers at Goettingen were so surprised by the very normal dimensions of Gauss’s brain when observed at autopsy that they inferred he could not have been such a great mathematician as had generally been supposed.

    However, according to this paper, in “slowly dividing neurons and adipocytes, cell volume scales with body size.” So apparently larger brains have larger neurons, which, as with computer chips, must mean slower processing speeds.

    • Replies: @res
    , @AaronB
  117. res says:
    @CanSpeccy

    So apparently larger brains have larger neurons, which, as with computer chips, must mean slower processing speeds.

    Because Dennard scaling applies to neurons as well. Or not. There may be a size effect, but I doubt it is anything like that for computers. Do you have any real evidence for the neuron size/speed connection or are you just speculating?

    https://en.wikipedia.org/wiki/Dennard_scaling

    • Replies: @CanSpeccy
  118. Habermas is a theorist. – As head of one of the departments of the Max Planck Institute with some dozen employees (if that many) he failed and soon resigned. He made the practical experience, that he is unable to handle even a fairly small amount of power.

    One of Habermas’ greater insights is about Freud. It’s a two-word thing, as so many great insights are: The scientistic self-misunderstanding of Freud (ok – maybe three words). He says Freud was wrong when he thought his main focus was on nature. Freud’s main focus was on perception/ deception (cf. Knowledge and Interest, p. 300 -332 in the German edition).

    We can’t escape this since we all have a subconscious. But we have ways to work around this fundamental problem of us as human beings. And one solution in human societies – from pretty early on – is to functionally differentiate roles in order to make sure, that different aspects (=perceptions) of a problem arise (=exist!).

    His focus is on power here: The person that decides executes power and is therefore focused on how to do things, whereas the person that reflects has not to carry the burden to make things work (ends meet…) and therefore has a structurally different perspective on our affairs. A look from outside. As I said: That’s structural – and it’s a structural benefit of societies to allow members to act in these different roles, because this approach increases the number of aspects, which then can be discussed.

    As an aside: The Frankfurt School is a lot about rationalization (in Max Weber’s sense) and that has a lot to do with secularisation. Habermas (and Adorno, too) thinks quite a lot about this subject. The Frankfurters all agreed, that this phenomenon is very important. Maybe Erich Fromm found the best solution, but that’s now an aside in my aside: Fromm simply said, what counts is the clear mind, and not the frame of reference – i. a. words: There is room for religion in the modern world and in the realm of rational discourse (=rationality). Fromm was expelled from the Institute from Horkheimer and Adorno, whereas Habermas moved away from Horkheimer and Adorno’s (and Freud’s!) ideas about religion and moved – the longer the more – in the direction of Fromm’s position.

    As a philosopher of language, Habermas has come to the following – in my eyes: important – conclusion: We all, as speakers in a social setting, don’t really know what we’re talking about (Truth and Justification, p. 40 & p.246ff. in the German edition) – or in other words: We can never be absolutely sure what we want to say. And that’s deeply rooted in the nature of language itself and therefor unescapable: Language and reality are so intertwined, that there is no way whatsoever, to precisely disentangle them (Habermas refers here not least to the late Wittgenstein).

    Simple solution: We better interact! And we better be careful, to interact in a civilized way. Truth, rightness (=correctness) and beauty rest on reasonable interaction more than on anything else – with the exception of God’s grace, I might add, on this sunny Sunday morning.
    (See – that’s still Kant’s way to rationalize (or, maybe better: to translate) a Christian thought, namely the one about God: That we can’t know, what God is. Kant said: And not only that, we don’t know, what reality is, either. All we can know is how we try to describe/ analyze/ understand reality (= our means, our methods, our laws, our traditions….).

    PS
    seized ceased – I make such mistakes every now and then. They are accompanied by a certain identifiable feeling when I make them. I make them anyway. I’m glad you don’t mind! – This error has at least two deep bodily roots. One is my eyes’ blind spots are way over average in size. The other one is: I almost choked as a kid on a coin, which a guest in my family’s restaurant handed over to me because I was so cute (a little over two years old – curly blond hair…). So for me, speaking/writing and choking are close relatives. What means: I at times prefer to march on rather than go back and correct, because I have the impression then, that I might run out of the air to breathe if I went back… Still crazy after all these years… –

  119. Since correlation between brain size and intelligence is roughly 0.3 pointless to mention outliers because the correlation already tells you there will be many (both better and worse than predicted).

    Worth looking at the work of Genc about intelligent brains.
    http://www.unz.com/jthompson/the-well-tempered-clavichord/

    Assume you have seen this:
    http://www.unz.com/jthompson/intelligent-brains/

    The hypothesis that men are actually 4 points higher in IQ
    http://www.unz.com/jthompson/men-4-points-ahead/

    • Replies: @CanSpeccy
    , @res
  120. AaronB says:
    @res

    Unfortunately I don’t remember where I got those numbers. Its been years since I did serious reading in the literature, but I remember being astonished at the enormous discrepancy between sub-factors in Jewish IQ.

    Since my figures are in the ballpark of what you remember, it seems we are approximately on the same page here.

    Res, you didn’t address my point. I can readily see how evolution might select for specific abilities. However, the general tendency is to downplay the importance of specific factors in favor of g, which is spoken of as if it is the key determinant in a persons ability to excel in any field.

    However, with such wide discrepancies between specific factors, the general factor obviously cannot account for a large fraction of ones total ability in any field, even if it is a statistical fact.

    In other words, g simply isn’t that important if there can be a 20 point difference in specific abilities.

    That would probably result in improvements on g and verbal-specific, but stasis (or a decline) for spatial-specific

    You are actually positing a situation where g goes up, but spatial declines (because of sub-factor decline)! Does this not further reduce the importance of g?

    And how many laymen who discuss IQ would recognize such a thing is even possible when all they hear is the importance of g, so that they practically equate it with intelligence?

    A general factor which does not account for much of the total outcome does not deserve any special emphasis.

    Now of course, res, you may recognize the relatively low contribution g makes to total outcome, and the relatively high contribution specific factors make (of in fact you do), but this is not how most people think. For which you are not responsible, of course.

    • Replies: @res
  121. AaronB says:
    @res

    – Changes which increase the number of brain cells uniformly could increase both g and brain size.
    – Changes which increase brain cell volume uniformly (e.g. increased myelination) could increase both g and brain size.

    So no, not only don’t I admit that. I disagree with you.

    Anything is possible. But the anecdote Thompson provided, which was meant to support the claim that brain size correlates with intelligence, at best only supports the notion that bran size correlates with a specific sub-factor.

    As for restriction of range – OK, but then it comes down to what we see as important and our impressions of how this issue is generally discussed. My impression is that a 110v/130m split – a 20 point difference – seriously downgrades the importance of g, and that this isn’t reflected in the way IQ is generally discussed, particularly among HBDers – who often treat g as equivalence to intelligence.

    For instance, one hears much about how the Ravens Matrices is the purest measure of intelligence because it is the most g-loaded. This is a nonsensical statement on light of our conversation here.

    As for determining a minimum threshold of general intelligence then testing sub-factors, its the other way around. G is derived from testing sub-factors, so it does not add any useful iformation.

  122. CanSpeccy says: • Website
    @James Thompson

    Since correlation between brain size and intelligence is roughly 0.3 pointless …

    I was once at Waterloo Station in London on race day at Ascot, and in fact, I was on the platform of the Ascot train. A number of the men boarding the train were in top hat and tails, obviously headed for the races. What was noticeable about most of these fellows was how large they were, with commensurately big heads*. Yet if you look at your average old-fashioned, i.e., actually English, cab driver in London, the fellow offering a ride to “Eefrow Airpor!” he’ll likely be quite scrawny and generally small in all dimensions.

    So what you can say, pretty certainly, is that whole-body mass and cranial capacity have a connection with social class, which in turn means that variation in cranial capacity pretty certainly has an environmental component. Not that I’m saying that the son of a cabbie raised in the home of duke will necessarily be the size of a typical duke,** but he may be quite a bit less scrawny than if raised in a traditional English home in the East end of London (if there still are any after the Bliar/May’s ethnic cleansing). So I think that trying to make anything of a trivial relationship of cranial capacity on IQ (r-squared of less than 10%), is probably a waste of time.

    ———
    * As I recall from my very early days, which means back in what to virtually everyone else is a historical time, common folk where I lived often used the term “bighead” as a term of abuse. I have no idea about the actual etymology of the term, but I have long supposed that it reflected recognition of the generally large dimensions of upper class males, including their heads, compared with the laboring class.

    ** Several generations ago, when English schools were still filled with English kids, I competed in an athletic event at Eton College, and I can confirm that the English upper class male was then, generally speaking, much larger than the youth of the hoi polloi. What’s more some of them were very fit and ran damn fast. Today of course Eton, like all the top English schools, is filled with ethnic Pashtuns, Hindus, Africans, etc. so goodness knows how large are the heads of students there today.

  123. res says:
    @James Thompson

    Since correlation between brain size and intelligence is roughly 0.3 pointless to mention outliers because the correlation already tells you there will be many (both better and worse than predicted).

    I wonder if GWAS will eventually be able to extract separate factors for brain size and speed/efficiency. It would be interesting to see what the R^2 for a model of IQ with those two factors (either genotypic or phenotypic) would be.

    Thanks for the additional links.

    P.S. Regarding the neuron size vs. speed discussion above I think it is worth noting that increased myelination increases both cell volume and propagation speed.

  124. AaronB says:
    @CanSpeccy

    I think there is a bias towards “linear thinking” at play here, which is common among HBDers. Bigger = better.

    It’s a primitive way of thinking.

  125. CanSpeccy says: • Website
    @res

    or are you just speculating?

    Yes. But I believe that the now defunct British Post Office Electrical Engineers’ Journal had an article that considered the question of brain size versus processing capacity and concluded that the effect on capacity of an increase in neuron number would be offset by a reduction in the speed of each operation. So they concluded that there is an optimal brain size, about the average size of a human brain.

    • Replies: @res
  126. res says:
    @CanSpeccy

    Yes. But I believe that the now defunct British Post Office Electrical Engineers’ Journal had an article that considered the question of brain size versus processing capacity and concluded that the effect on capacity of an increase in neuron number would be offset by a reduction in the speed of each operation. So they concluded that there is an optimal brain size, about the average size of a human brain.

    This is exactly the sort of connection and conversation I look for here (and illustrates why I value smart, knowledgeable people so highly). Thanks! A reference would be very helpful though. I would very much like to see that argument fleshed out. What kind of assumptions did they make and what kind of factors were taken into account?

    A similar effect occurs in electronic circuits where increased chip size can cause longer interconnection length (increasing delay which decreases maximum clock speed). Much effort is expended to manage that as well as dealing with clock distribution issues: https://en.wikipedia.org/wiki/Clock_signal#Distribution
    The brain has a great advantage over electronic circuits in being three dimensional though.

    Regarding optimal brain size, the prevailing view (I think) is that the driving factor there is female hip width. I find it unlikely those two effects would produce the exact same optimum. Though I do find it plausible that a decreasing returns on brain size effect could reduce selection pressure in favor of intelligence enough to influence optimal hip width.

    A decreasing returns on brain size effect would also influence the optimum trade off for metabolic cost vs. brain size.

    I think the parsimonious explanation here is a sub-linear increase in intelligence with brain size. The “optimum” would be imposed by other pressures (hip width and metabolic cost) so no need to posit the intelligence vs. brain size relationship turning negative. Of course the interesting question here is “how sublinear?” Hard to tease that out in the presence of both allometry and differences in non-size based brain functioning. Also unclear is how non-linear the size-intelligence relationship is. Presumably the second derivative is negative, but by how much?

    What do you think?

    P.S. Also see my comment about increased myelination increasing both cell volume and processing speed.

    • Replies: @CanSpeccy
    , @James Thompson
  127. res says:
    @AaronB

    Now of course, res, you may recognize the relatively low contribution g makes to total outcome, and the relatively high contribution specific factors make (of in fact you do),

    Um, no. Across the population as a whole g makes a large difference. Linda Gottfredson’s work shows this clearly.

    You can observe that within specific selected subgroups (often selected roughly by g ; ) specific abilities matter more. You may have noticed I gave an extended example illustrating just that. But that in no way contradicts my previous paragraph.

    As I have observed before with you, we are clearly in agree to disagree territory here. I gave a detailed exposition of my views and all I am getting in return is sniping at details and attempts to put words in my mouth.

    This sentence of yours summarizes the issue.

    However, with such wide discrepancies between specific factors, the general factor obviously cannot account for a large fraction of ones total ability in any field, even if it is a statistical fact.

    Wow. It is your choice to ignore reality. Good luck with that.

    • Replies: @AaronB
  128. AaronB says:
    @res

    Right, it seems like we agree on the facts, but disagree on interpretation.

    To me, a 20 point discrepancy on sub-factors renders g far less important than commonly acknowledged, and renders ridiculous the notion that g is more or less equivalent to intelligence.

    You disagree with this. And that’s perfectly OK.

    I think the important thing is to present these facts to the average educated layman, and let him decide for himself.

    I believe that when the average educated person realizes that IQists speak of g as practically equal to intelligence, but sub-factors exhibit the levels of discrepancy that they do, they will lose faith in the judgement of IQists.

    Perhaps I’m wrong. The important thing is to present the hard facts.

    You can observe that within specific selected subgroups (often selected roughly by g ; ) specific abilities matter more

    There is no logical reason to limit this to within subgroups. IQ differences between subgroups are modest, typically between 5-15 points.

    Since sub-factor discrepancies can be as much as 20 points, it is clear it can compensate for discrepancies in group g in a large number of cases.

    Furthermore, it’s worth pointing out that Jewish g cannot be higher than their lowest sub-factor score, spatial, which in this case is below 100. Although g is, as far as I am aware not quantified, it would appear the “general factor of intelligence” for Jews is quite low, and without excelling in certain sub-factors, Jews would be a low IQ group based on g alone.

    So Jewish excellence seems driven entirely by sub-factor performance, and not general intelligence.

    Nevertheless, g is the most important factor 🙂

    Or maybe, the whole science of IQ is incoherent 🙂

    Good luck, Res.

    • Replies: @res
    , @James Thompson
  129. res says:
    @AaronB

    Since sub-factor discrepancies can be as much as 20 points, it is clear it can compensate for discrepancies in group g in a large number of cases.

    Don’t underestimate things here. Individuals can have even bigger differences than that. Why not just quote the biggest number you can find?

    P.S. A large portion of the point of statistics is to try to quantify the kind of handwaving that appears in the last part of your statement.

    Also, regarding

    Furthermore, it’s worth pointing out that Jewish g cannot be higher than their lowest sub-factor score, spatial, which in this case is below 100.

    Thanks for making clear you did not read and understand my comment 117 about how I see things. I won’t waste my time next time.

    • Replies: @AaronB
    , @utu
  130. AaronB says:
    @res

    This conversation ended exactly where I expected it to. My points are lucid and exceedingly simple. You were completely unable to grapple with any of them, and just became steadily more insulting and agitated, as expected 🙂

    Ah well, as I told utu, you people are not capable of being intellectually honest, and nothing we do can change that.

    • LOL: res
    • Replies: @res
    , @utu
  131. res says:
    @AaronB

    I ask anyone inclined to take comment 134 at face value to go back and read comments 115-118. Then decide for yourself who is being intellectually honest here.

  132. CanSpeccy says: • Website
    @res

    What do you think?

    I think you are thinking about it in the right way, which is to say in evolutionary terms. But I think that approach requires some hard and careful thinking, particularly if you go beyond the trivial question of brain size and function to the question of what the brain has to do to provide an adaptive edge. From that perspective, one begins to see that intelligence, however that term is understood, is not, except at a basic level, so important. But to say more than that, I’d have to write an essay, which I don’t have time for now, so I’ll leave that task to you.

    • Replies: @res
  133. res says:
    @CanSpeccy

    Thanks for responding. I think it is too hard to do much analysis here without more data regarding the selection pressures involved (e.g. intelligence, hip width, metabolic cost). As you note, understanding the adaptive edge provided by different levels (and types) of intelligence is difficult. Though perhaps comparisons with animals all along the evolutionary hierarchy would prove helpful. How does intelligence help them at different levels?

    I suspect if we are to get answers here they will come from first finding the genetic contributors to various traits and then attempting to understand how their SNP frequencies (both individually and the relationships) have varied across time both within and between populations. We should also be able to look at where various intelligence related genes appeared (and how they changed) across the evolutionary timeline.

    • Replies: @CanSpeccy
  134. utu says:
    @res

    “did not read and understand my comment 117 about how I see things”

    I read the #117 comment but I decided not to comment it to spare you an embarrassment however since you insist that it was an important comment of yours I have no choice but to address it.

    You stated

    G1 = first PC of M1, V1, and S1

    which means that G1 is a linear combination of M1, V1 and S1 and G2 and G3 also are linear combinations of the same three test results. Conversely M1, V1 and S1 each are a linear combination of G1, G2 and G3, so for example one could write: M1=a*G1+b*G2+c*G3, where a, b and c are constant coefficient. These coefficients when variable M1, V1, S1 are normalized are correlations or in other words loading. For example a=Cor(M1,G1).

    Then you proceed to write

    M1 = m1 * g1
    V1 = v1 * g1
    S1 = s1 * g1

    where the asterisk ‘*’ may or may not denote a multiplications. In you introductory remark you have stated

    Note that where I write things like M = m * g if would be more accurate to say M = f(m, g). But IMHO the former provides a simpler and more effective visualization and is easier to discuss.

    so it is hard to figure out what do you mean. Clearly you have abandoned the certainty of the math of the principal component analysis and entered a territory of fantasy and confusion. You should stick to empirical facts and mathematical certitudes.

    • LOL: res
    • Replies: @res
  135. res says:
    @utu

    The * is indeed multiplication. Which is standard enough that you really should not need to ask: https://www.dictionary.com/browse/multiplication-sign

    The qualification about M = f(m, g) was an acknowledgement that reality is more complex, but a simple multiplication is easier to reason about. And serves as a useful first order approximation.

    As for the rest. WTF? Do you think that added anything? How about discussing the implications of what I wrote rather than trying to nitpick?

    As far as:

    to spare you an embarrassment

    The projection. It burns.

    P.S. Did you not get that in

    which means that G1 is a linear combination of M1, V1 and S1 and G2 and G3 also are linear combinations of the same three test results.

    G2 and G3 are for completely separate groups ? Therefore your conclusion is wrong.

  136. @res

    Agree the brain size/intelligence link is sub-linear in homo sapiens, because the species is already optimized for intelligence, and might go further if required. Hip width an important limiting factor, so C-section would be required (already a majority in Brazil, to my great surprise, but not for intelligence boosting reasons).

    The supposed limit due to dendrites might not be such a limit if the Genc finding of fewer connections holds up.

  137. @AaronB

    Plomin and Deary regard the general factor as accounting for 40% of the variance. Then there are the group factors, 3 or even 4, but I think 3 more easy to justify, (say another 40 % for those) and then specific skills.

    In my view these figures vary according to the analyses used, and also the purpose of the factorization. In understanding the variety of human skills, an approach which favours factors is understandable. In summarizing results from different people tested on different assessments, then using a more all-embracing general factor is more useful.

    • Replies: @AaronB
  138. utu says:
    @AaronB

    Ah well, as I told utu, you people are not capable of being intellectually honest, and nothing we do can change that.

    Many of them believe they are honest like for instance our friend ‘res’ who is very earnest to the point of obsequious sycophancy with respect of those whom he consider his intellectual superiors within the IQist community. The IQists are self-recruited from among the people who have a predilection for social order sorting into inferiors and superiors and when they are really earnest, like ‘res’, they submit to their superiors willingly and will be more than happy to render their services to their superiors which you can sometimes observe in the sycophantic behavior of ‘res’. I believe that in the case of ‘res’ it is possible to rescue him because his basic core of personality is still healthy. However the top echelon of IQists can’t be salvaged, so the rehabilitation of them will not be possible and thus it should not be undertaken.

    I am disappointed with the attitude of the mainstream psychologists who opted for an easy way out by assuming the position of moral superiority and disgust with the deplorable IQists instead of engaging them and challenging them. Our friend ‘res’ has not been sufficiently challenged and this lead to inconsequential ‘merry creativity’ on his part which can be exemplified by his comment #117.

    Gould’s book was a good start at the démontage of IQism but it fell short. Additional chapters should have been written. One by a philosopher good at epistemology and ontology and another one on the ethical dimension of science and finally the third one by a mathematician that would deconstruct Jensen’s use of factor analysis.

    Gould’s graphic explication of factor analysis’s was insufficient and he missed several very important points. Most imprtantly, he did not question the intelligence tests and their structure as if they were somewhat objective or God given on the Mount Sinai or by Angel Moroni in Palmyra, NY. Binet himself, I think, would have never claimed such attribute of objectivity of his tests. Let me elaborate on this.

    If we had an access to raw data from all questions that constitute each test used in the battery of tests and performed principal component analysis on the huge covariance matrix we could identify a question that when removed from the data set increases the ratio between the first and the second components the most and also a question that reduces the ratio between the first and the second components the most. So we can restructure the tests to make the first component more dominant or less dominant if we continue this process. By removing a subset of questions we can achieve a parity between the first and the second components. Then no tricks of factor analysis could be puled off to dismiss the second factor.

    • LOL: res
    • Replies: @AaronB
  139. AaronB says:
    @James Thompson

    In summarizing results from different people tested on different assessments, then using a more all-embracing general factor is more useful.

    But the general factor only accounts for a little more than a third of variance. So if you want to know how two people will do on any specific task, g will tell you very little in most cases, unless the discrepancy is so large as to outweigh the remaining two thirds of variance. And that certainly does not apply to IQ differences among groups, although it may between individuals, although rarely.

    So g can only be important in rare, extreme cases.

    Now, I must say that this little more than a third figure for g is an astonishing figure. This surely isn’t reflected the way HBD people and IQists talk about it.

    Judging by how they talk, most of them I bet would say g accounts for like 95% of variance.

    So I merely propose that the facts become more widely known.

    I think if people understood that g accounts for a minority of variance they would stop thinking silly things like g is practically equivalent to intelligence and that the Raven Matrices offer the “purest” and “best” measure of intelligence because it is the most “g-loaded”. (Not to mention it is the test most affected by the Flynn Effect, but that’s another ridiculous story)

    In fact, the importance of g-loaded tests or activities cannot survive the sensational revelation that g accounts for a little over a third of variance.

    G-loaded would come to be synonymous with tasks that require so little specific intelligence as to be relatively unimpressive.

    But since there is widespread ignorance of the facts, silliness and stupidity reign – as perhaps they always do in human affairs.

    • Replies: @res
  140. res says:

    There is much talk about intellectual honesty (and IMO little walk) in this thread from both utu and AaronB. I thought I would include a passage of mine from comment 118 which I think exemplifies intellectual honesty. I challenge either of them to find a similar example in their comments above. Or an example of intellectual dishonesty in mine.

    Ok, so then surely you will admit that skull size does not correlate with g, but only with certain specific factors of intelligence, at least according to Thomsons anecdote.

    That’s an interesting point. I think it has significant truth in it (much brain functioning “goodness” is not size related, e.g. electrolytes and action potential propagation velocity, metabolic efficiency) but is not completely true. Two reasons
    – Changes which increase the number of brain cells uniformly could increase both g and brain size.
    – Changes which increase brain cell volume uniformly (e.g. increased myelination) could increase both g and brain size.

    So no, not only don’t I admit that. I disagree with you.

    First, I agree there is some validity to his point then call out the specific areas where I agree. Then I give my specific reasons for disagreement.

    P.S. And as a bonus, here is an example of intellectual dishonesty from utu. From comment 83 above:

    I was not disappointed. Gould’s arguments are sound.

    From the previous comment.

    Gould’s graphic explication of factor analysis’s was insufficient and he missed several very important points.

    Which is it, utu?

    P.P.S. Utu, if you haven’t noticed the cases where I question and disagree with the experts then you haven’t been paying attention. I think my comment 117 above where I make a specific proposal counter to the idea that g (there G, measured g) has the same profile across populations serves as an example. Though I do think g as defined in that comment has the same meaning across populations (roughly: general brain functioning “goodness”) and corresponds to what psychometricians really mean when they talk about g. IMO the disagreement occurs because of the necessity of measuring g. Which is much harder in the framework I propose. In a single population g and G are effectively the same (e.g. g1 is the scalar magnitude of the vector G1).

  141. res says:
    @AaronB

    But the general factor only accounts for a little more than a third of variance. So if you want to know how two people will do on any specific task, g will tell you very little in most cases, unless the discrepancy is so large as to outweigh the remaining two thirds of variance. And that certainly does not apply to IQ differences among groups, although it may between individuals, although rarely.

    That reasoning does not hold up for individuals. For two individuals (with the same racial background) with different g’s the expectation of their specific abilities will be in line with their g (unless the two individuals already were selected for g, as in the example I gave above). The wildly divergent special abilities case is the uncommon one. (statistics, again)

    Again, I will note that selecting first for g changes things greatly. The positive correlation of specific abilities present in the general population decreases (or even turns negative!) after g-based selection.

    For groups you have a better point. Those statistical differences in mean special abilities between groups matter. There is a reason we see massive under or over representation for some groups in various fields. Even beyond the g differences.

    If you actually engaged with my arguments you would have noticed my comment 117 attempts to address the between group case. I think if you actually understood what I am saying (and were more intellectually honest) you would be more sympathetic to that comment.

    Also, rounding 40% to “a little more than a third” and then implicitly to a third with “remaining two thirds” is a great example of intellectual dishonesty. 40% and 60% are actually shorter to write.

    So g can only be important in rare, extreme cases.

    I am astonished how proudly you display your ignorance of statistics.

    But since there is widespread ignorance of the facts, silliness and stupidity reign – as perhaps they always do in human affairs.

    The projection. It burns.

    • Replies: @AaronB
  142. AaronB says:
    @utu

    I agree that a “multi-front” approach should be adopted, because IQism can be challenged on many levels.

    Even in this thread I conceded many assumptions I do not agree with – for instance, that g is in fact a meaningful statistical artifact, when your mathematical arguments demonstrate it’s a trivial and artificially constructed one.

    But I wanted to focus on one simple, clear, and lucid point.

    I am intrigued by Goulds book now – for so long its been held up by HBDers as the model of frivolous science that I ignored it, but of course if HBDers scorn it it probably has merit.

    Res really does seem to think he’s honest, which is curious, but his behavior is more reflective of a desperate desire to “save the appearances”. Unfortunately IQism is probably tied to ones whole belief structure, and questioning it might induce panic in people who recognize it will undermine “truths” they depend on for a feeling of stability.

    But perhaps, as you say, res has a healthy core and can recover his senses.

    Yes, IQism seems to satisfy a desire for rigid hierarchy, and lends itself easily to elites wishing to justify the status quo. As I mentioned somewhere else, IQism is the “divine right” of our age.

    • LOL: res
    • Replies: @res
  143. AaronB says:
    @res

    That reasoning does not hold up for individuals. For two individuals (with the same racial background) with different g’s the expectation of their specific abilities will be in line with their g (unless the two individuals already were selected for g, as in the example I gave above). The wildly divergent special abilities case is the uncommon one. (statistics, again)

    You are assuming that we are dealing with statistically average specimens of the same racial group.

    You are basically saying that since statistically it is more likely we are dealing with average specimens, the one with the higher g will do better on any sub-factor, since we can assume both have the average sub-factor profile of their group, and g is the only variable.

    You are holding sub-factors constant and g as the only variable.

    Yes, if “all else is equal”, then of course boosting one factor will give an advantage.

    However, if we are to assume that both individuals are within the average range, as is statistically likely, then the g for both individuals must also be within the average range, not just the sub-factors.

    The average IQ score range is about 20 points for whites, I believe, between 90-110. That’s a 20 point range for average.

    So we have two individuals who are average – statistically the most likely to be encountered – and both whose g and sub-factor scores fall within the average range for their racial group.

    We know that one has a higher g. We don’t know where within the average range their sub-factor scores fall.

    Now, g only accounts for a bit more than 1/3 of the variance in their performance on any sub-factor – but the average range for sub-factors is quite wide, 20 points. Clearly, the g discrepancy between the two would have to be quite wide to allow us to come to any conclusions about their sub-factor performance, which I admitted earlier.

    However, there is a problem. As an individuals g rises, in order to still remain within the range of an average specimen within his racial group, his sub-factor scores must fall. This would skew his profile towards being very unusual for his group, rendering him no longer a statistically average specimen for his group.

    Clearly then, there is a very limited range of g discrepancies possible between two “statistically average specimens” of the same racial group, so a wide discrepancy is not possible in our example.

    Again, I will note that selecting first for g changes things greatly. The positive correlation of specific abilities present in the general population decreases (or even turns negative!) after g-based selection.

    G is derived from the correlation between sub-factors. Why would a group selected on the basis of g exhibit a lesser or a negative correlation.

    The only way this is possible, is if g actually contributes very little to the outcome of performance on any specific factors it, leaving room for wild variations after g is accounted for.

    Which rather tends to support my point.

    For groups you have a better point. Those statistical differences in mean special abilities between groups matter. There is a reason we see massive under or over representation for some groups in various fields. Even beyond the g differences.

    I am pleased that you accept this. And we agreed that we don’t differ on facts with regards to this – only on emphasis and interpretation.

    To me this massively downgraded the importance of g, to you it does not. To me this is not reflected in the way most IQists discuss this issue.

    But perhaps we don’t really disagree so much free, which is great.

    So g can only be important in rare, extreme cases.

    I am astonished how proudly you display your ignorance of statistics.

    Lol.

    • Replies: @res
  144. res says:
    @AaronB

    Are you trying to gaslight me? The degree to which your comments match the bullet points in this article (more detail there) is interesting: https://www.psychologytoday.com/us/blog/here-there-and-everywhere/201701/11-warning-signs-gaslighting

    1. They tell blatant lies.
    2. They deny they ever said something, even though you have proof.
    3. They use what is near and dear to you as ammunition.
    4. They wear you down over time.
    5. Their actions do not match their words.
    6. They throw in positive reinforcement to confuse you.
    7. They know confusion weakens people.
    8. They project.
    9. They try to align people against you.
    10. They tell you or others that you are crazy.
    11. They tell you everyone else is a liar.

    • Replies: @AaronB
  145. AaronB says:
    @res

    Lol, I was actually going to write a comment yesterday about how the whole IQ thing may be an attempt to gaslight us – it is so full of holes and logical discrepancies.

    However, gaslighters often accuse others of gaslighting – its one of the tactics of gaslighting.

    So now I’m not sure I was wrong…

    • Replies: @res
  146. res says:
    @AaronB

    Since I am still not sure I think I will engage in an exercise. Going forward I will note the bullet points I see represented in your comments. Let’s start with that one. In this case it’s pretty much the same thing you accuse me of so is not convincing, but hopefully the aggregate over time will give a better indication.

    1. They tell blatant lies.
    8. They project.

    And since it’s not specifically in the list of bullet points I will note that your response is a perfect example of its own point.

  147. @res

    Thanks. Comment No. 122 is the answer to this post of yours. I made just another mistake. Sigh!

    • Replies: @res
  148. CanSpeccy says: • Website
    @res

    … understanding the adaptive edge provided by different levels (and types) of intelligence is difficult. Though perhaps comparisons with animals all along the evolutionary hierarchy would prove helpful.

    Yes, but in one respect, human intelligence is very different from that of all other creatures. That I think is what you need to focus on: In what way is human intelligence unique and in what way is it, or has it been, adaptive.

    In thinking about that one needs to think about how the neurological attributes of the organism determine whether it thrives or dies. To do that successfully, one needs to forget all about g, IQ tests and the so-called cognitive capacities that IQ tests measure. Instead one must think like a naturalist, like Darwin, in fact, and look at behaviour not in terms of SAT tests and employment at Google, FB or Tesla, but in terms of representation in succeeding generations. In particular, you need to be a naturalist of humanity as humanity existed in the past, not in the crazy and biologically dysfunctional way that people of the Western world live today (unlike those cheerfully multiplying low-IQ Africans, so held in contempt by certain people here at Unz.com).

    I suspect if we are to get answers here they will come from first finding the genetic contributors to various traits and then attempting to understand how their SNP frequencies (both individually and the relationships) have varied across time both within and between populations.

    That sounds like the way to gather data for boring articles in boring journals, but not a way to understand what created the human mind, how it was shaped, and how it serves, or fails to serve, the inhabitants of the world we live in now.

    • Replies: @res
  149. res says:
    @AaronB

    So much wrong there it is hard to know where to begin. And I learned a long time ago that entering into a long discussion like this with AaronB is useless. But let’s take care of some of the obvious BS and hopefully readers will realize the rest is not to be trusted either.

    The average IQ score range is about 20 points for whites, I believe, between 90-110. That’s a 20 point range for average.

    “average IQ score range” is not exceptionally rigorous or meaningful. +/- 1 SD (85-115) includes 68% of people. 50% of people are within about 2/3 SD. Perhaps that is where you got your range.

    You are assuming that we are dealing with statistically average specimens of the same racial group.

    Nope. Your statistical ignorance again shines brightly. What I am observing is that if you pick two random people differing in both g and specific abilities then the overall specific abilities will vary in a way that the specific abilities will on average vary with g.

    Now, g only accounts for a bit more than 1/3 of the variance in their performance on any sub-factor

    Nope. And again note the intellectual dishonesty of “a bit more than 1/3” when the more honest “about 40%” requires less space to write.

    g accounts for about 40% of the variance of the entire suite of tests. How much it corresponds with each subtest varies quite a bit. Explained variance of 40% is a correlation of 0.63. Some subtests correlate much more highly with g than that. Here is a look at different subtests in terms of variance explained by g: https://www.quora.com/IQ-Testing-How-does-each-subtest-in-the-WAIS-IV-correlate-with-the-Raven-test

    However, there is a problem. As an individuals g rises, in order to still remain within the range of an average specimen within his racial group, his sub-factor scores must fall. This would skew his profile towards being very unusual for his group, rendering him no longer a statistically average specimen for his group.

    Now that was some LOL-worthy “reasoning.”

    G is derived from the correlation between sub-factors. Why would a group selected on the basis of g exhibit a lesser or a negative correlation.

    Restriction of range results in that effect. Which I have mentioned repeatedly. It appears your lack of statistical knowledge prevents you understanding this, but let’s try again with a concrete example: https://www.statisticshowto.datasciencecentral.com/restricted-range/

    A negative correlation can appear if the initial correlation is a bit lower (e.g. 0.4). It is a fairly pathological case.

    Your tendency to follow up an incorrect premise like that with a strong conclusion illustrates just how sloppy your “reasoning” is. Then there is your glib assertion “The only way this is possible.”

    The only way this is possible, is if g actually contributes very little to the outcome of performance on any specific factors it, leaving room for wild variations after g is accounted for.

    I am curious how many times I have to demonstrate your factual incorrectness (not just “emphasis and interpretation”) before it becomes clear you are building castles of sand on a non-existent foundation.

    And regarding your LOL at the end, do you honestly not realize how apt my characterization of “how proudly you display your ignorance of statistics” is? My current comment makes that even more clear.

    • Replies: @AaronB
  150. res says:
    @Dieter Kief

    Thanks. I made the same mistake with comment 144 to utu. It is an easy mistake to make when responding to the most recent comment.

  151. res says:
    @CanSpeccy

    Yes, but in one respect, human intelligence is very different from that of all other creatures. That I think is what you need to focus on: In what way is human intelligence unique and in what way is it, or has it been, adaptive.

    I strongly agree. I do think primates are also worthy of comparison though because I think social ability is an important component of human intelligence and its adaptive importance. But humans add another level.

    In thinking about that one needs to think about how the neurological attributes of the organism determine whether it thrives or dies.

    Yes. One area which I think shows this exceptionally well is spatial ability. I think anyone who has traveled off trail in complex terrain with dense foliage can appreciate its importance.

    That sounds like the way to gather data for boring articles in boring journals, but not a way to understand what created the human mind, how it was shaped, and how it serves, or fails to serve, the inhabitants of the world we live in now.

    Sometimes you have to assemble a large collection of “stamps” to understand the world. I think Darwin’s finch data collecting would seem quite boring to anyone unfamiliar with where it led.

    One area I would call out for particular attention is systems biology. I think the way it connects genetics to low level physiology to high level physiology offers a great deal of potential for answering questions of the sort which interest you. But I think a great deal of boring legwork and boring articles in boring journals will be necessary to reach the payoff.

    • Replies: @CanSpeccy
  152. AaronB says:
    @res

    What I am observing is that if you pick two random people differing in both g and specific abilities then the overall specific abilities will vary in a way that the specific abilities will on average vary with g.

    Granted, each sub-factor test has a different g loading.

    But the specific factor contribution to the total score is in an independent variable.

    So say 2 people from the same racial group get two different scores on a sub test. Say Block Design.

    49 percent of the variance is accounted for by g in this case. 51 percent is accounted by for by each individuals specific factor ability.

    So if all we know about two individuals from the same racial group is that they differ on g, that isn’t enough information to say that the one with the higher g will score better on Block Design, unless the difference in g is so great that it cancels out the potential 51 percent accounted for by the specific factor.

    That would be an atypical case, since very high or very low g are extremes.

    So g alone is only useful when differences are extreme, both with individuals and groups.

    A negative correlation can appear if the initial correlation is a bit lower (e.g. 0.4)

    The only way this is possible, is if g actually contributes very little to the outcome of performance…

    Isn’t this saying exactly the same thing?

    • Replies: @res
    , @AaronB
  153. res says:
    @AaronB

    Granted, each sub-factor test has a different g loading.

    And 4 of the subtests have a variance explained by g in the range of 59-69%.

    So say 2 people from the same racial group get two different scores on a sub test. Say Block Design.

    So if all we know about two individuals from the same racial group is that they differ on g, that isn’t enough information to say that the one with the higher g will score better on Block Design

    Better to think of it as two normal curves for the special ability. Offset by an amount determined by the difference in g. That difference influences the probability of which person has the higher score. I hope that analogy provides a useful way to think about the excerpt you quoted–which on rereading was a bit cryptic.

    Per the table above Block Design correlates 0.7 with g (49% variance explained). A good comparison which might help the Americans here with intuition is to compare with the SAT V-M correlation of 0.71 per Table A.4 in https://www.ets.org/Media/Research/pdf/RR-99-02-Dorans.pdf
    BTW, that table is a nice quick reference for ACT and SAT subtest/composite correlations in general.

    Isn’t this saying exactly the same thing?

    Not really. Unless you are only looking at a subpopulation which has already been selected by g. Which we have discussed multiple times now as a special case. And note that a 0.4 correlation is equivalent to variance explained of 16%. Much smaller than the cases we are discussing.

    • Replies: @AaronB
  154. AaronB says:
    @AaronB

    So g alone is only useful when differences are extreme, both with individuals and groups

    On sub-factors where g accounts for 69 percent of variance, differences in g alone can be meaningful at a much lower threshold. (At least within racial groups with the same g)

    But for IQ as a whole, where it is .4, the threshold is much higher, not to mention for certain specific sub-factors.

    Spatial tests seem to be some of the most g-loaded, second only to verbal. However, Jews do poorly on spatial and high on verbal….where does this leave their g?

    Clearly, their verbal ability is so high that even a modest contribution to the total score results in an outside impact.

    • Replies: @res
  155. CanSpeccy says: • Website
    @res

    I do think primates are also worthy of comparison though because I think social ability is an important component of human intelligence

    True, but human intelligence is totally different from that of other primates in one enormously important respect, a difference reflected in the vast difference in cortical mass between humans and chimps, etc.

    One area which I think shows this exceptionally well is spatial ability. I think anyone who has traveled off trail in complex terrain with dense foliage can appreciate its importance.

    Yes. Several indigenous groups excel at path finding, the Innuit for one, and the Australian Aborigines, despite their abysmally low IQ.

    Sometimes you have to assemble a large collection of “stamps” to understand the world.

    Well, who knows what will come out of new areas of knowledge and certainly the molecular info will tell us a lot about something. But one should not be blinded by science, or what is excitingly novel anyhow, and thus miss what may have been blindingly obvious for centuries, and which I believe was generally recognized as obvious before the rise of psychology as a “scientific” discipline.

    • Replies: @res
  156. Factorize says:

    res, on vocabulary.com I have scored millions of points with their vocab builder game and I have likely hit the ceiling of usable everyday vocabulary, yet when I tried a block design test once I had no clue. I can hardly ever remember solving jigsaw puzzles or playing with tinkertoys. How would you interpret this? Vocabulary is highly g loaded on your chart, while block design is somewhat less so. I am inclined to believe that all of my vocab practice has removed the g loading from this subtest, while my near absence of block design experience has largely made this subtest a poor measure of g for me as well. I would think that a test such as this in which it could be observed that performance improved as testing progressed would largely invalidate the results. Spearman noted similar observations in one of his earlier papers. In particular, he noted that the village school children did not appear to have any familiarity with one of the tests that headminstered (musical notes?).

    • Replies: @res
  157. AaronB says:
    @res

    Better to think of it as two normal curves for the special ability. Offset by an amount determined by the difference in g. That difference influences the probability of which person has the higher score.

    But g only accounts for 49 percent of variance between two individuals in our example. That means the increase in g must be capable of offsetting the remaining 51 percent that isn’t g.

    So even if our two individuals lie on a normal distribution curve for special ability, we have no idea where within the normal range they lie.

    One can have a much stronger specific factor than the other, yet both be within the statistical normal range.

    Assume one has higher g but much weaker specific factor, the other has lower g but much higher specific factor, yet both fall within the statistical normal – are not extreme cases.

    So the difference in g must be large enough to offset the potential contribution of the specific factor to be predictive, which automatically makes it an extreme case.

    • Replies: @res
  158. res says:
    @CanSpeccy

    True, but human intelligence is totally different from that of other primates in one enormously important respect, a difference reflected in the vast difference in cortical mass between humans and chimps, etc.

    The raw brain size difference is huge, and clearly important. But I’m not sure whether I would rate that or the speech capability higher. Do we know much about the development of speech in humans? It seems hard to judge given the lack of tangible artifacts before writing.

    I think the abstract of this paper has some useful information: https://www.ncbi.nlm.nih.gov/pubmed/22230639

    Evolution of the brain and intelligence in primates.

    Abstract

    Primates are, on average, more intelligent than other mammals, with great apes and finally humans on top. They generally have larger brains and cortices, and because of higher relative cortex volume and neuron packing density (NPD), they have much more cortical neurons than other mammalian taxa with the same brain size. Likewise, information processing capacity is generally higher in primates due to short interneuronal distance and high axonal conduction velocity. Across primate taxa, differences in intelligence correlate best with differences in number of cortical neurons and synapses plus information processing speed. The human brain stands out by having a large cortical volume with relatively high NPD, high conduction velocity, and high cortical parcellation. All aspects of human intelligence are present at least in rudimentary form in nonhuman primates or some mammals or vertebrates except syntactical language. The latter can be regarded as a very potent “intelligence amplifier.”

    I should probably go read the paper itself (it’s actually a book chapter). The paper has DOI 10.1016/B978-0-444-53860-4.00020-9 and is at libgen. I included the conclusion after the MORE. I think it’s worth a look, but is a bit long.

    Yes. Several indigenous groups excel at path finding, the Innuit for one, and the Australian Aborigines, despite their abysmally low IQ.

    Is there good data on the M/V/S balance of those groups? It would be interesting to see how well the M/V/S framework I outlined above captures their ability profiles.

    Well, who knows what will come out of new areas of knowledge and certainly the molecular info will tell us a lot about something. But one should not be blinded by science, or what is excitingly novel anyhow, and thus miss what may have been blindingly obvious for centuries, and which I believe was generally recognized as obvious before the rise of psychology as a “scientific” discipline.

    I strongly agree with this. I think we are actually regressing in much of our understanding of group differences. This ties in with one of my current projects. Collecting pre-PC books with an emphasis on history, and old versions of well known bowdlerized books (with the bowdlerization often being well known because it was so proudly done).

    [MORE]

    Conclusion of the primate paper above. Not block quoted for readability. Some acronyms used below:
    ABS – Absolute brain size (ABS, cm3 or g), note allometry discussion following page 418
    NPD – neuron packing density
    IPC – Information processing capacity

    General conclusions

    Primates turn out to be, at least on average, more intelligent than other mammals, with great apes and ultimately humans leading. Primates generally have larger brains than other mammals of the same body size. However, there is only a moderate fit between differences in ABS and intelligence. This is predominantly due to considerable variation in the brain–body relationship, that is, some primates have much larger and others much smaller brains and cortices than expected, and there is additional variation in NPD independent of brain and cortex size. Cortex volume and packing density determine the number of cortical neurons, which largely determines memory capacity as one decisive factor for intelligence. Primates in general have much more cortical neurons than other mammals of the same brain size. The second important factor for intelligence is information processing speed, which is generally high in primates due to their relatively high NPD and resulting short interneuronal distance and high axon conduction velocity. The third factor is extensive parcellation of the cortex according to the rule of intense local and sparse global connectivity. Thus, IPC rather than ABS yield the best correlate with intelligence. The human brain combines a large cortex with a relatively high NPD, high conduction velocity, and high parcellation, which together result in the highest IPC and intelligence among animals.

    Cetaceans and the elephants have larger to much larger brains and cortices than even
    humans, but less cortical neurons due to the fact that their NPD is much lower, interneuronal distance much larger, and axonal conduction velocity lower. Finally, cortex parcellation seems to
    be poorly developed. This appears to strongly impair IPC and could explain, why whales, dolphins, and elephants are not nearly as intelligent as one would predict on the basis of brain size. The opposite seems to happen in corvid birds (and presumably parrots) with very small
    brains, but high NPD and processing speed, which could explain why these animals reveal an intelligence comparable to monkeys and even great apes with much larger brains.

    In most respects, the human brain fits the general trends found in primates and mammals. With one remarkable exception, we find no aspects of human intelligence that is not present at least in
    rudimentary form in nonhuman primates and at least in some mammals or vertebrates. The exception appears to be syntactical language, which presumably has evolved about 100,000years ago, and the neuroanatomical correlate is the formation of the Broca speech center. Syntactical language can be regarded as a very potent “intelligence amplifier,” as was later the case with
    the invention of scripture and eventually computers.

    • Replies: @CanSpeccy
  159. res says:
    @AaronB

    Spatial tests seem to be some of the most g-loaded, second only to verbal. However, Jews do poorly on spatial and high on verbal….where does this leave their g?

    I have a feeling that high g loading decreases a bit if you start looking across groups.

    As for where that leaves their g, I think that depends on how you define g. If you base it on IQ tests normalized for white Europeans and without a large spatial component I think that probably overstates their g (though I can come up with counterarguments).

    It would be interesting to try to understand how the M/V/S profiles and resultant g and G’s for the different groups relate, but unfortunately that topic is much too toxic in the Current Year.

    One thought. Does the high g loading of spatial ability imply something about its relative evolutionary importance? To me that implies relative uniformity of the spatial-specific anatomy between individuals. Is that a sign of greater or lesser selection pressure on that anatomy?

  160. res says:
    @AaronB

    But g only accounts for 49 percent of variance between two individuals in our example.

    I’m not sure that follows. That 49% of variance explained is averaged over the entire group. A good way to think about this is to look at two variable scatterplots. The point density is instructive. Any two individuals might have their difference in a specific ability underestimated/on target/ overestimated by their g difference.

    There is much uncertainty here (which is pretty much your point), but we can draw probabilitistic conclusions based on the statistical data.

    I don’t think words are the right way to address these questions. Or at least consider that a balance of words and statistical reasoning is better.

    • Replies: @AaronB
  161. res says:
    @Factorize

    Hard to say. Practice can make a big difference in facility, even if not in the underlying raw ability. I think in some cases heuristics can substitute for intelligence in some subtests. The question though is when does the ability to develop and use those heuristics == intelligence? (my answer tends to be that developing and using heuristics differ greatly in difficulty, which is a big part of the reason teaching helps, and of the reason the people who expect students to come up with everything from first principles are misguided IMO)

    I think your speculation is reasonable. How does your family do on comparable tasks? Both for better and worse. Especially if they have made a different level of effort with respect to the different tasks than you. As an example, I have a family member who was quite good at math, but never really pushed that. I suspect our respective test results overstate the raw ability difference between us. But by less than a naive observer looking at us perform would.

    In any case, I think I would be more interested in the opinion of someone with extensive background in test development and administration than I would be in my own. Perhaps Dr. Thompson has something to say here? Or pointers to past posts?

  162. AaronB says:
    @res

    Thanks for the discussion, res. I understand a little better where you are coming from, but I still think it’s impossible to describe something that only accounts for 16 percent variance as very important, and must say that just isn’t reflected in the way this issue is commonly discussed.

    I would like it if these facts were more widely known, and people can decide for themselves.

    But it seems we disagree on assessment, judgement, interpretation, etc. And we discussed only one aspect of the IQ question.

    There are so many things one can discuss regarding IQ – for instance, above the subject of practice and how it affects IQ was raised. It seems clear that a child raised in a culture that encourages constant determined abstract problem solving – not giving up easily, being highly motivated, persistence, – will eventually arrive at his first IQ test in a better prepared state.

    This makes IQ as a measure of innate ability highly questionable.

    Then there is personal motivation, which is malleable and highly contingent, and can’t be abstracted from the total score to leave a residue of pure talent.

    Then there’s utu’s mathematical demonstrations that g is trivial and an artificial construct.

    So much ground to cover.

    In the meantime, if IQists would only downgrade their claims and be more modest, it would be a good start.

    • Replies: @res
  163. CanSpeccy says: • Website
    @res

    The raw brain size difference is huge, and clearly important. But I’m not sure whether I would rate that or the speech capability higher.

    Are you sure the two (brain size and speech) are unconnected?

    As for human intelligence, I wonder if, in a few decades, it will be of much interest to anyone, when it will have been greatly surpassed by AI. It would be like caring about the physical strength of the workforce after the replacement of muscle power by the power of machines.

    The fact is, intelligence research has been a bust. There have been no breakthroughs, no new understanding of the mind. The major advances in psychology have occurred in the areas of behavioral economics, and neurophysiology.

    • Replies: @res
  164. res says:
    @AaronB

    but I still think it’s impossible to describe something that only accounts for 16 percent variance as very important

    16??? What happened to g explaining 40% of the variance? The point I was trying to make there was that the case I saw where the correlation turned negative was a much lower correlation (0.4, 16% of variance) than that between g and the average subtest (0.63, 40% of variance).

    Then there’s utu’s mathematical demonstrations that g is trivial and an artificial construct.

    Pro tip. Utu is not as good at math as he thinks he is. And he has a tendency to start from questionable premises. Pages of math are useless if they make meaningfully incorrect assumptions. Be cautious about taking his math at face value if you do not understand it in detail. Shalizi’s version is probably better, but he still makes the fundamental mistake of conflating simple existence of a principal component with existence of a PC explaining 40% of variance.

    In the meantime, if IQists would only downgrade their claims and be more modest, it would be a good start.

    Perhaps. Rigorously defining “IQist” and debunking (and I mean debunking in a fashion which survives counterargument) specific claims would also be a good start.

    As far as the facts being available. They are pretty much there in the papers. How about you learn more and inform people in a more rigorous fashion?

    I will note the one thing we seem to meaningfully agree on is the need to go beyond g when discussing group differences in subtest profiles. Except I see that as needing to improve a quite good model for some incremental gain. While you seem to think it is an excuse to throw out 100 years of psychometrics baby with the bathwater.

    Round and round we go. Although parts of the conversation are interesting, the overall experience is extremely tedious.

    • Replies: @AaronB
    , @utu
  165. AaronB says:
    @res

    Sorry, 40%.

    I trust my little utu. I think he’s good. Even though I don’t really know the math he uses, reading his comments do not show any serious logical discrepancies, while reading IQist comments show tons. So he has surface credibility with me at least, while IQists lack even that.

    It would be nice to rigorously define IQist and collect all the multi-level arguments against them, but that would probably be best on my own blog and not on a comments section.

    On this thread, I took up one single, and simple, line of argument, and wanted to follow it to the end.

    • Replies: @res
  166. res says:
    @CanSpeccy

    Are you sure the two (brain size and speech) are unconnected?

    Absolutely not. Almost certainly a connection. I think the paper I linked talks about this (what they call “syntactical language”), but I have not gotten to it yet. I do not think they are the same phenomenon though.

    As for human intelligence, I wonder if, in a few decades, it will be of much interest to anyone, when it will have been greatly surpassed by AI.

    Agreed. If the AI singularity really happens much will change. Though there is a possibility trying to understand human intelligence will become even more important as part of trying to give those AIs the full spectrum of human abilities. Interesting times.

    The fact is, intelligence research has been a bust. There have been no breakthroughs, no new understanding of the mind. The major advances in psychology have occurred in the areas of behavioral economics, and neurophysiology.

    I would be interested to hear someone in the field respond to this. What kind of understanding of the mind are you looking for?

    • Replies: @CanSpeccy
  167. utu says:
    @res

    Shalizi’s version is probably better, but he still makes the fundamental mistake of conflating simple existence of a principal component with existence of a PC explaining 40% of variance.

    I am glad you acknowledge Shalizi who exposed the concept of g as a mere rhetorical device that its existence supposed to prove something and give some solid theoretical foundations to Spearman and Jensen mumbo jumbo while the existence of g is a mathematical inevitability so g existence can’t signify anything. You are too smart not to acknowledge Shalizi yet you are still in a denial so you turned to Kabbalah and found a special meaning in the number 40. Because of the magic of the number 40 you think you can ignore Shalizi and persists in your obscurantism.

    Getting 40% of variance explained is not a big deal. Keep in mind that all PC’s must explain in total 100%. So for example in case of three variables like in your #117 comment (the pinnacle of your merry creativity) your G1 can’t explain less variance than 33%. 33% is not that far from 40%, right? So perhaps 40 is not as magical as you wanted us to think in your last resort argument.

    When the variables (individual tests in the battery of test) are strongly correlated with each other then the dominat component will be able to explain larger fraction of variance. If you want to boost artificially the variance explained by the first PC you can append to the battery of tests an additional test that is strongly correlated with others tests. For instance if you append Raven Matrix test to other intelligence tests your g will be ‘stronger’. One can play many games. And the field of playing games is even wider in Factorial Analysis where pretty much anything goes once you allow for oblique rotations.

    BTW, in Kabbalah 40 suppose to signify:

    The number 40 represents transition or change; the concept of renewal; a new beginning. The number 40 has the power to lift a spiritual state. Consider:

    When a person becomes ritually impure, he must immerse in a ritual bath, a mikveh. The Talmud tells us that a mikveh must be filled with 40 se’ahs (a measure of water). Immersion in a mikveh is the consummate Jewish symbol of spiritual renewal.

    Listen res, wake up. Take a ritual bath and cleanse yourself from the mumbo jumbo and the filth of IQism.

    • LOL: res
  168. res says:
    @AaronB

    reading his comments do not show any serious logical discrepancies

    You mean like the one I pointed out in comment 144?

    P.S. And as a bonus, here is an example of intellectual dishonesty from utu. From comment 83 above:

    I was not disappointed. Gould’s arguments are sound.

    From the previous comment.

    Gould’s graphic explication of factor analysis’s was insufficient and he missed several very important points.

    Which is it, utu?

    Regarding your statement:

    while reading IQist comments show tons

    If there are “tons” then it should be easy to show a dozen. How about you show five? Even one fully quoted and referenced would be a good start. Put up or shut up with your vague negative comments.

    We all get to choose who we believe. Choose wisely. Especially if you don’t possess the ability to check.

    P.S. I have been lax about continuing my gaslighting bullet points exercise. Here are the ones I think best fit that comment of yours.

    1. They tell blatant lies.
    5. Their actions do not match their words.
    11. They tell you everyone else is a liar.

    • Replies: @utu
    , @AaronB
  169. utu says:
    @res

    “intellectual dishonesty” – Really? You see a contradiction between “sound” and “insufficient” ?

    • Replies: @res
  170. CanSpeccy says: • Website
    @res

    Though there is a possibility trying to understand human intelligence will become even more important as part of trying to give those AIs the full spectrum of human abilities.

    So far, it’s been the computer geeks who’ve guided the thinking of the psychologists. Remember the genius Harvard school of psychology led by B.F. Skinner and others who denied the reality of consciousness and human intentionality? But then the development of missile guidance computers (among other developments in cybernetics) required systems able to model the external world in order to attain a target location or state, i.e., these were goal-seeking mechanisms. The absurdity of Skinnerism could thus no longer be maintained.

    there is a possibility trying to understand human intelligence will become even more important as part of trying to give those AIs the full spectrum of human abilities

    It was artificial intelligence research that demonstrated the information processing power of neural networks. This was more than the psychologists or the neurologists were able to do. I suspect that cybernetics will continue to lead in unraveling the workings of the mind. However, the findings of neurologists and physiologists will surely continue to provide inspiration to those engaged in that effort, just as the development of neural networks in silico was inspired by knowledge of brain anatomy and physiology.

    What I expect is that as AI systems increase in power and versatility, they will be increasingly modular (like the brain), with different modules processing different types of sensory input, with other modules relating the processed inputs to past inputs (memory) to establish the current state of the system in relation to a hierarchy of goals, and yet other modules formulating outputs via artificial limbs, rockets, radio signals or whatever, to advance the system along its goal sequence.

    With different AI models of the same basic type, the g factor will emerge, reflecting differences among systems in chip speed, architecture, etc. Thus g will be seen as no more mysterious than a consequence of the properties of the substrate from which the system has been constructed and the architecture of the basic system components.

  171. AaronB says:
    @res

    Your example of utu’s logical inconsistency shows you to be a rigid binary thinker, with all the strengths and weaknesses of that mode of thought.

    Utu general assessment of Goulds arguments are that they are sound, however, that does not preclude him from pointing out certain weaknesses in them. He never said they were perfect.

    IQist logical inconsistencies?

    1) G is more or less equivalent to intelligence.

    G explains only 40% of variance.

    2) Anyone who does well in one area is likely to do well in another, because there is a general factor of intelligence underlying all mental abilities.

    Jews excel at verbal, but do dismally at spatial.

    3) Both verbal and spatial tests are highly g-loaded.

    Jews do great at verbal, but terrible at spatial, leaving their g where exactly….

    4) IQ measures innate ability.

    Any test necessarily measures motivation+innate ability, and no method exists to disentangle the two. (And I would argue it is conceptually impossible).

    5) Training or practice cannot increase IQ.

    The Flynn Effect affects mostly the Raven Matrices, which has risen as societies become more familiar with abstract problem solving and technology, which is precisely the kind of thinking the Ravens tests for.

    6) G is derived from the correlation between tests.

    Scores on the Ravens Matrices have been increasing for decades, but not on any of the other tests, meaning g should be constantly recalculated downward as true correlations weaken. Instead, the Ravens is continuously “recalibrated” in order to “save the appearances”.

    7) IQ exhibits a near perfect correlation with national wealth.

    Asian countries with higher IQs remain significantly less wealthy than Europe and America as a whole.

    8) Jews dominate in America because of IQ, which is the key determinant to nearly all group outcomes.

    Mean IQ and population size shows that whites produce far more intelligent people in absolute terms.

    9) Asians are naturally suited to STEM.

    Modern Asian culture places tremendous pressure on succeeding in areas that lead to wealth, industrial, and military power, making it impossible to know how much of this is motivation. Asians frequently hate these fields, but are driven by tiger parents to adopt them anyways. The role of tiger parenting is completely ignored by IQists. Classical Asian culture was literary and had an anti-technological bias.

    I can only give you one piece of advice, res.

    Listen res, wake up. Take a ritual bath and cleanse yourself from the mumbo jumbo and the filth of IQism.

    • Agree: utu
    • Replies: @res
  172. res says:
    @utu

    Interesting that you ignored the rest of my quote: “missed several very important points.” And yes, that contradicts “sound.”

    More intellectual dishonesty (partial quote, and the weaker part). Thanks for the confirmation. I love self refuting comments.

    • Replies: @utu
  173. res says:
    @AaronB

    References to where “IQists” have actually said those things would help. I think most of your points are outright wrong (e.g. Asian’s math abilities do make them suited to STEM, though math is certainly only one part) or overblown. That you think those vaguely stated “inconsistencies” are more important than the examples of utu’s intellectual dishonesty I have given (explicitly, with supporting quotes from him) then that says a great deal.

    Not going to engage one by one with your points. You have already amply demonstrated you are not a sincere interlocuter and that engaging with you in depth is a waste of my time.

    What’s fun is that the way you and utu are supporting each other in lockstep (regardless of content) means that whenever I score a point on either of you it impacts both.

    • Replies: @AaronB
  174. utu says:
    @res

    You are not arguing in a good faith. You must be angry.

    Gould arguments were sound though not complete but good enough to demolish the g-ism of Spearman and Jensen, however Gould could have made even stronger arguments like the one I outlined in the comment #142.

    • Replies: @res
  175. AaronB says:
    @res

    But you completely missed my point, res.

    How much of the Asian math ability is the result of training – a culture of tiger parenting that encourage a focus on these fields and incredible persistence and effort, from a very young age.

    We already know from the Flynn Effect that IQ is quite responsive to training, familiarity with technology, and environment – especially and particularly when it comes to STEM (only STEM-like Ravens experiences Flynn, not verbal).

    Plus, Asian math advantage is negligible, like 3 points. That accounts for the massive and overwhelming concentration of Asians in STEM?

    I believe Jews have a higher math score, and show nowhere near that kind of concentration in STEM.

    But you cannot answer these points, and I don’t blame you for not trying.

    Also interesting that was the one point you chose.

    Not going to engage one by one with your points

    Of course you won’t 🙂

    IQists never do.

    You have already amply demonstrated you are not a sincere interlocuter and that engaging with you in depth is a waste of my time.

    Obviously 🙂

    • LOL: res
    • Replies: @res
  176. res says:
    @utu

    Rebuttal by ad hominem. Well done, utu. I think that qualifies as more intellectual dishonesty.

    I am engaging with exactly what you wrote using specific (and complete!) quotes to do so. Far more in good faith than you.

    We must have different definitions of “missed several very important points” and “sound.”

    Fortunately all of our statements are recorded above so others can judge for themselves.

    Oops, almost forgot. The projection (bad faith). It burns.

    • Replies: @utu
  177. utu says:
    @res

    Are you really serious? Read my comments #83 and #142 on Gould again. Do you honestly believe that there is inconsistency or contradiction between them? I have elaborated it for your benefit in comment #178. Would anybody in a good faith accuse me of intellectual dishonesty on the basis of these two comments?

    Your accusation can only be explained by bad-faith, anger or stupidity. I did not impute the latter because I know you are not stupid while anger can temporarily lead to a bad-faith even in a just man.

    I have noticed you like to invoke various formal fallacies and impute Freudian projections as your favorite eristic devices of the last resort. Remember that sometimes a cigar is just a cigar. Not everything negative said about you or your behavior is a projection and even if it is, it still can be accurate. Have you read Schopenhauer’s Eristic Dialectic as a young man? Perhaps too young.

    • Replies: @res
  178. res says:
    @AaronB

    I think everyone here is well aware of how good you are at spewing vague BS. But at least try to get the facts right on the rare occasion you cite one.

    Plus, Asian math advantage is negligible, like 3 points. That accounts for the massive and overwhelming concentration of Asians in STEM?

    Asian – White mean difference on the SAT Math is about 60 points (612 – 553): https://www.insidehighered.com/news/2017/09/27/scores-new-sat-show-large-gaps-race-and-ethnicity

    Per this page the full group SD is 107 (would be larger within races, so this understates Cohen’s d).: https://nces.ed.gov/programs/digest/d17/tables/dt17_226.40.asp
    So a minimum gap of 0.55 SD. And don’t forget that the low ceiling of the SAT Math decreases the apparent gap. Per this link 2% of Asians score 800 on the SAT Math: https://collegereadiness.collegeboard.org/pdf/sat-percentile-ranks-gender-race-ethnicity.pdf

    Where was your 3 points number from? Or was it just a lie?

    If you knew anything about statistics, which it is clear you don’t, you would realize that gap implies a large representational advantage at the upper tail. Emil’s tail calculator: http://emilkirkegaard.dk/understanding_statistics/?app=tail_effects
    indicates a representational advantage of 3.2x at >2 SD and 5.3x at >3 SD.

    I believe Jews have a higher math score, and show nowhere near that kind of concentration in STEM.

    Than Asians? Presenting data would be helpful. When I was in college Jews seemed quite well represented in STEM, but since then with the rise of Asians in the US and the Jewish advantage in verbal they seem to have concentrated more in fields which value verbal more. Even now I’m pretty sure Jews are quite well represented in STEM. Remember that there about 4x as many Asians as Jews in the US (5.6% vs. 1.4%).

    But you cannot answer these points, and I don’t blame you for not trying.

    Oops.

    How many times do I have to present clear proof that you are wrong on particular points (here the magnitude of the Asian-White gap) before you acknowledge it? This is a big part of what I mean when I say you are a dishonest interlocutor.

    P.S. The best way for you to “win” this argument would be to give a comment containing accurate facts with references and a well reasoned argument. I would probably keel over from the shock.

    • Replies: @AaronB
    , @AaronB
  179. AaronB says:
    @res

    The 3 point roughly math advantage of Asians I remember from my reading in the literature back in the day, probably from Murray.

    Strangely, a quick Google search did not unearth a detailed breakdown of Asian IQ, and I suspect you had the same problem as you posted only SAT links.

    However , I found this –

    https://www.google.com/url?sa=t&source=web&rct=j&url=http://www.uvm.edu/provost/20%2520-%2520Ztatny%2520reading.pdf&ved=2ahUKEwjNhsCDsOzgAhUEVN8KHaElAmYQFjAIegQIChAB&usg=AOvVaw0iSPWEZkLbURIQn_o2YW6X&cshid=1551836424988

    I could not copy and paste, but on page 301 right column it says the Asian advantage on math 3.3 points, and higher on spatial. And the article goes into an extended discussion of how the Asian skew is tilted towards spatial.

    It has been common knowledge for some time now that the Asian advantage is especially high on spatial, small to modest in math, and negative to zero in verbal.

    The discussion about Asians has always focused on their spatial.

    Interestingly, James Flynn believes Asians have no IQ advantage over whites based on his research, which I did not know!

    Like me, he basically thinks they do well because they are overachiever.

    You reveal an interesting pattern – you ignore many of my points and try and attack only the ones where you think, usually mistakenly, you have an advantage. You conspicuously ignore the significance of motivation and a long standing culture of persistence and problem solving – obviously, because IQ research methods simply cannot account for these factors, and they ate potentially devastating to the whole IQ project.

    No one likes to admit that their field simply lacks the capacity to even address potentially devastating challenges, which would mean it is basically a null field, or as canspeccy says, bust, as a serious field.

    An honest man would address all points.

    Of course, we have been told that Jewish IQ is 115 based on highly atypical samples, and I’ve seen reports that Poland and scores as high as 106.

    IQ researchers must all be taken with a grain of salt.

    • Replies: @res
    , @res
    , @res
  180. AaronB says:

    I admire your bulldog tenacity, res – but you have picked a losing side. You cannot win.

    I propose we do another 40 comments, and when that mystical number is reached, you start anew as an honest man.

    • LOL: res
  181. AaronB says:
    @res

    If you knew anything about statistics, which it is clear you don’t, you would realize that gap implies a large representational advantage at the upper tail.

    Oh, and I am not talking about the upper tail.

    At mid level state schools and even community colleges, where the less bright Asians congregate, they all concentrate in STEM just as much as their elite brethren do.

    To understand why, you have to know Asian history. Asians were traumatized by a West whose edge lay in technology. Clearly perceiving this, they understood their path out of humiliating subordination and towards self-determination lay in rejecting their literary properties and embracing this new learning, repugnant as it was to their traditional culture. Further, they wanted to vindicate themselves in the eyes of the West, and in their own eyes, that they were not inferior.

    These psychological complexes have not relaxed their hold on them, and most Asians, especially Chinese, still feel the sting of humiliation from their encounter with the West.

    This is a powerful motivation to excel in technology.

    • Agree: utu
    • Replies: @res
  182. res says:
    @AaronB

    small to modest in math

    I think my SAT Math does a good job of substantiating a larger difference than that in favor of Asians.

  183. res says:
    @AaronB

    Oh, and I am not talking about the upper tail.

    Who exactly do you think is going into STEM? The people near the mean?

    Those Asians at the mid level schools majoring in STEM are there because places like Harvard don’t admit them. Not because they are weak at math.

    Your argumentation is a fascinating look at the workings of a high V low M mind.

    • Replies: @AaronB
  184. res says:
    @AaronB

    An honest man would address all points.

    As you have done with mine above. LOL.

  185. res says:
    @utu

    Would anybody in a good faith accuse me of intellectual dishonesty on the basis of these two comments?

    Yes. And have you forgotten your intellectually dishonest incomplete quote leaving out the most persuasive part of my argument? Comment 173 to be clear.

    Do you have any real arguments or is it ad hominems all the way down?

    And you let AaronB’s completely unsubstantiated intellectual dishonesty statement in comment 112 slide by without comment. Is that more intellectual honesty on your part?

    • Replies: @utu
  186. res says:
    @AaronB

    I followed your reference:

    However , I found this –
    https://www.google.com/url?sa=t&source=web&rct=j&url=http://www.uvm.edu/provost/20%2520-%2520Ztatny%2520reading.pdf&ved=2ahUKEwjNhsCDsOzgAhUEVN8KHaElAmYQFjAIegQIChAB&usg=AOvVaw0iSPWEZkLbURIQn_o2YW6X&cshid=1551836424988

    I could not copy and paste, but on page 301 right column it says the Asian advantage on math 3.3 points, and higher on spatial. And the article goes into an extended discussion of how the Asian skew is tilted towards spatial.

    There are better versions out there. Here is a plaintext version: https://publicism.info/psychology/curve/15.html

    And a complete OCRed PDF is available on libgen.

    Here is the relevant paragraph from page 301.

    Vernon’s overall appraisal was that the mean Asian-American IQ is about 97 on verbal tests and about 110 on visuospatial tests.81 Lynn’s 1987 review of the IQ literature on East Asians found a median verbal IQ of 98 and a median visuospatial IQ of 106.82 As of 1993, for Asian-American students who reported that English was the first language they learned (alone or with another language), the Asian-American SAT mean was .21 standard deviation above the national mean on the verbal test and .43 standard deviation above the national mean on the math test. Converted to an IQ metric, this amounts to a 3.3 point elevation of mathematical scores over verbal scores for the high IQ Asian-American population that takes the SAT83

    It turns out you misread that. Read that last sentence again. More carefully this time. They were comparing Asian verbal and math scores. Your 3.3 number comes from (0.43 math – 0.21 verbal) * 15

    I will also note that they used an SAT Math score comparison just as I did. The difference from 1993 was 0.43 SD. The 2017 difference was 0.55 SD as I documented above. Using the standard SD of 15 for IQ that gives a gap of 6.45 for the Bell Curve data you cited and 8.25 for the more recent data I cited.

    You are not very good at facts and evidence. Are you at least going to own this error of yours?

    P.S. Below is reference 83 (pp. 724-725) for those who like to follow things back to the original source. In this case it does not give an original source though, just commentary.

    [MORE]

    83. The SAT data actually provide even more of a hint about genetic ortgins for the test-score pattern, though a speculative one. The College Board reports scores for persons whose first language learned is English and for those whose first language is “English and another.” It is plausible to assume that Asian students whose only “first language” was English contain a disproportionate number of children of mixed parentage, usually Asian and white, compared to those in whose homes both English and an Asian language were spoken from birth. With that hypothesis in mind, consider that the discrepancy between the Verbal and Math SATs was (in IQ points) only 1.7 points for the “English only” Asians and 5.3 points for the “English and another” first-language Asians. Nongenetic explanations are available. For example, one may hypothesize that although English and another language were both “first languages,” English wasn’t learned as well in those homes; hence the Verbal scores for the “English and another” homes were lower. But then one must also explain why the Math scores of the “English and another” Asians were twenty-one SAT points higher than the “English-only” homes. Here one could hypothesize that the “Englishonly” Asians were second- and third-generation Americans, more assimilated, and therefore didn’t study math as hard as their less assimilated friends (although somehow they did quite well in the Verbal test). But while alternative hypotheses are avaitable, the consistency with a genetic explanation suggests that it would be instructive to examine the scores of children of full and mixed Asian parentage.

    • Replies: @AaronB
  187. utu says:
    @res

    And you let AaronB’s completely unsubstantiated intellectual dishonesty statement in comment 112 slide by without comment. Is that more intellectual honesty on your part?

    Now I get it. You apparently could not get over my check mate comment #110 and its follow up by AaronB with the comment #112. Your idol J. Philippe Rushton has been owned by me. For a little sycophant like you it must have been an unpardonable offense. Whether Rushton was intellectually dishonest as suggested by AaronB or just not too smart or both (there is no other possibility) I won’t be bothered finding out. Rushton’s reputation if he ever had it is beyond repair anyway.

    This ‘unpardonable offense’ lead you to concoct the false antinomy based on two adjectives ‘sound’ and ‘insufficient’ so you could accuse me of intellectual dishonesty. A pretty simple psychological mechanism, isn’t it? Perhaps IQist are right that human mind is exceptionally simple that it can be explained with just one number. Or rather the IQists make the world in the image of their simple minds.

    P.S. Read my comments #83 and #142 on Gould again. Take your time. Calm down and think it over. You do not need to apologize.

    • LOL: res
  188. Jay says:

    Gould was an evolutionary theoretician of the verbal sort, and most of all, a polemicist. He was not a scientist of any note. His empirical work (on Bahamian land snails) was poorly done. His life’s focus was entirely on trying to shape opinion, both of evolutionary biologists and of the public.

  189. Jay says:
    @niteranger

    Your assessment of Gould and of the reactions of actually excellent evolutionary biologists to him could not be more correct. In a seminar I taught, it was easy to pick apart one of the few empirical papers that Gould ever published. Students were shocked by how poor the paper was.

    • Replies: @Dieter Kief
  190. AaronB says:
    @res

    Res, immediately before that it says Asians have a verbal IQ of about 97 or 98. So s 3.3 point Asian difference between math and verbal puts them actually a little under a 3 point advantage over the white average of 100.

    I interestingly, it says Asians have a verbal SAT of about .21 above the national mean – presumably the national mean is depressed by blacks, etc, and Asians do not have a .21 advantage over the white mean in verbal. Presumably whites also have an advantage of .21 over the national mean, scoring slightly above Asians in verbal.

    And they have a math advantage of .43 over the national mean which is depressed by blacks, etc – which suggests their advantage over the white mean is much lower.

    I find it odd that it isn’t easy to find Asian breakdown on google!

    I have also noticed Asian IQ inflation recently. In the Murray days, it was pegged at around 103 to 105 at the high end. Japan would usually be like 103 or 104.

    These days countries like HK and SG get silly numbers like 108 despite economically underperforming Europe (HK) and having negligible intellectual output, and Japan gets like 107.

    What is clearly happening is that the Asian academic dominance requires a “saving the appearances” project if IQ is to be saved, while as usual, IQists ignore the economic and intellectual underperformance of Asians, because they cannot address it.

    It is like when IQists pegged Jews at 115 based on small atypical samples from elite Jewish day schools even though the researcher collecting the data noted the sample was atypical, in order to “save the appearances”.

    • Replies: @res
  191. AaronB says:
    @res

    I’m not talking about state flagship schools which are often just behind the elite schools.

    I’m talking mid to low level schools. I know Asians from this segment, many of them hate STEM and are not especially good at it. They compensate by studying ferociously hard – Asian culture believe a hard work is more important than innate talent, unlike IQists. They are forced into STEM by their parents.

    I also know elite Asians who hate STEM but were forced into it.

    • Replies: @res
  192. res says:
    @AaronB

    So you are honestly not going to simply admit you misinterpreted that passage. I showed exactly how you made your error giving the incorrect 3.3 number. Consider this Exhibit A of AaronB intellectual dishonesty (and after you were the one who accused others of that in comment 112, the first mention of intellectual dishonesty in this thread, what projection, shame).

    Regarding your attempt to talk around the issue further. In comment 182 I gave a reference: https://www.insidehighered.com/news/2017/09/27/scores-new-sat-show-large-gaps-race-and-ethnicity
    showing the 2017 Asian – White SAT Math gap was 0.55 SD. That’s the best evidence presented in this thread and it clearly shows you are wrong about the magnitude of the gap. It is clear you don’t understand numbers so your attempts to use them to create handwaving arguments is laughable.

    Rest assured we will be hearing about this exchange again the next time you attempt to accuse someone of intellectual dishonesty.

    P.S. My accusations of your intellectual dishonesty and projection both validated in a single comment. Nicely done if I do say so myself.

    • Replies: @AaronB
  193. res says:
    @AaronB

    Do you keep those goalposts on wheels? You should. It makes them easier to move.

    Your argument takes one of the favorite forms of the statistical illiterate. I know of a few outliers and those disprove the conclusions drawn from statistics for the whole group.

    Unfortunately, arguments like that are only convincing to other statistical illiterates. Well, with an exception for true believers who find them convenient.

    You should stick to fields that are strictly subjective. There your arguments will be lauded (you are pretty good with the rhetoric).

    • Replies: @AaronB
  194. AaronB says:
    @res

    Vernon’s overall appraisal was that the mean Asian-American IQ is about 97 on verbal tests and….. Lynn’s 1987 review of the IQ literature on East Asians found a median verbal IQ of 98

    added to –

    …Converted to an IQ metric, this amounts to a 3.3 point elevation of mathematical scores over verbal scores for the high IQ Asian-American population that takes the SAT83

    Equals =

    Asians have a 3.3 higher math score than their verbal, which is 98 at best.

    Asian math = 101.3

    It is crystal clear. Asian verbal is low, and only 3.3 points separate their verbal from their math, meaning their math cannot be very high.

    Can it be you suffer from Dunning-Krueger syndrome?

    A pity.

    As for my other “terrible” math calculations, if Asians have a verbal IQ of 98, but score .21 above the national mean on SAT verbal, then either whites who have a a verbal of 100 score even higher above the national mean, or SAT and IQ don’t correlate well at all.

    Of course, SAT and IQ don’t correlate all that well, but I was willing to concede they do for the sake of the argument.

    And if Asians score .43 above the national mean on math, but only 3.3 IQ points separate their math from their verbal and their verbal is lower than whites….

    Must I connect the dots?

    It may be I overrated your intelligence.

    • LOL: res
    • Replies: @res
  195. AaronB says:
    @res

    First of all, I find it weird that of the 9 points I raised, we’re getting bogged down on the Asian thing. Are you Asian, res?

    Ok, Asians with a math IQ of 101.3 almost all flock to STEM. That is why community colleges and mid-level state schools are filled with Asians doing STEM.

    Not because of tiger moms, not because of psychological, cultural, or historical reasons. It has nothing to do with their humiliation by the technologically advanced West.

    And the 2,500 plus history of Asian culture being obsessively literary and having an ant-tech bias betray nothing about Asian proclivities. It is a mere fluke, and the past 50 years betray the truth.

    OK, res.

    • Replies: @res
  196. res says:
    @AaronB

    A weak attempt at ex post facto justification. You specifically quoted the 3.3 number in comment 183:

    on page 301 right column it says the Asian advantage on math 3.3 points

    Here is an interesting article on intellectual dishonesty: https://www.shanesnow.com/articles/intellectual-dishonesty#articles/intellectual-dishonesty-toc

    It is kind of fun to read with the current discussion in mind.

    P.S. And to be clear. If someone actually sees intellectual dishonesty in my comments please point to a well documented (quote with full context and link) example. I make mistakes sometimes and I also try to make an honest effort to improve. But there is a world of difference between vague accusations of intellectual dishonesty and well documented examples (as in Exhibit A for AaronB). That he just keeps doubling down on it only makes it serve better as an example.

    P.P.S. And his use of ad hominems in comment 198 is breathtaking. Perhaps 12 year olds find that style of argument persuasive. What was that statement I used to like? Ad hominems, the best way ever to say to someone arguing with you: “you win.”

    • Replies: @AaronB
  197. res says:
    @AaronB

    And now changing the subject. This is textbook. Such fun.

    And no, I am not Asian. More projection given the way you base your opinions on which group you belong to. And since I like evidence: http://www.unz.com/akarlin/short-history-of-3rd-millennium/#comment-1729073

    perhaps with the exception of Jews (I am Jewish, btw)

    I’ll be spending some time further exploring Exhibit A of AaronB’s intellectual dishonesty (and am expecting Exhibits B, C, etc. to show up if he keeps doubling down on it). I figure it will be very useful in the future so is worth some effort. Unless AaronB is smart enough to never again use the phrase “intellectual dishonesty” to refer to someone else in this blog.

    One other principle Exhibit A illustrates very nicely is the importance of choosing one’s battles wisely. One of AaronB’s primary tactics is to use vague arguments and keep shifting them whenever something specific enough to be countered makes an appearance. His comment 198 illustrates an attempt at this technique.

    Unfortunately that does not work when one makes specific well documented errors. Which just demonstrates AaronB’s extreme foolishness at choosing this battle by doubling down on his error.

    On the other hand, I recognize good ground for battle when I see it ; ) We might be here a while if AaronB continues to fail to understand this.

  198. AaronB says:
    @res

    I am honesty mystified on what you think is my intellectual dishonesty here – but in case your point is that my initial claim was that the article “stated” there was a 3.3 IQ point difference, when in fact it only “deduced” it from SAT scores, and it referred not to Asia advantage over whites but internal Asian IQ structure, I freely and happily concede that I did not present that aspect of it accurately.

    However, this in no way alters the substance of my point, which remains, obviously, valid. So in my haste I did not perfectly represent the exact way they reached their conclusions.

    Another example of your rigid, narrow thinking. This thinking has its strengths, but is severely limited.

    • Replies: @res
    , @utu
    , @utu
  199. res says:
    @AaronB

    He finally saw the light. But Exhibit A of AaronB’s intellectual dishonesty is recorded here forever.

    And to be clear, the original mistake is quite trivial. On the other hand the deceitful attempts to justify it without taking responsibility for the error is textbook intellectual dishonesty. Then there is the failure to acknowledge the superior (more recent and directly comparing Asians and whites) 2017 SAT Math evidence I presented. The trivial error matters because it illustrates how intellectually dishonest AaronB is. His extreme reluctance to admit error even in such an obvious and well documented case as I presented in Exhibit A of AaronB’s intellectual dishonesty makes clear that one should never expect him to be intellectually honest in a case where even a little bit of uncertainty or wiggle room is available.

    And concluding with an ad hominem. Well done, AaronB.

    P.S. And as far as your mystification. I have been very clear. That looks like a great example of:
    1. They tell blatant lies.
    And the whole discussion seems like
    2. They deny they ever said something, even though you have proof.
    to me.

    • Replies: @AaronB
  200. AaronB says:
    @res

    You’re losing it, res 🙂

    Calm down.

    • Replies: @res
  201. utu says:
    @AaronB

    You have been found out. Your IHQ (intellectual honesty quotient) is very low and you can do nothing about it except for blaming you father and mother for giving you the intellectual dishonesty genes. You will burn in the eternal hell. Not even purgatory for you. (NB: The first line of Luther’s attack on the Catholic Church was the belief in purgatory which was solidified with Calvin’s predestination doctrine which forever tainted the Anglo-Saxon mode of thinking. The IQism is its more recent manifestation.)

    AaronB’s intellectual dishonesty makes clear that one should never expect him to be intellectually honest

    Note the adverb ‘never’ which with its opposite ‘always’ is a favorite general quantifier signifying the IQists’ outlook and thinking. They are scared of the world and are terrified of the flux it is in. They would like to put everything in one big freezer and keep it there: “I am become Death, the freezer of worlds.”

    • Agree: AaronB
    • Replies: @res
    , @AaronB
  202. res says:
    @AaronB

    Stop with the gaslighting attempts. Your attempt to present yourself as a font of calm and rationality is hilarious.

    • Replies: @AaronB
  203. res says:
    @utu

    At least that was genuinely funny.

    But while we are on the topic of intellectual honesty I will just note the contrast between your excerpt

    AaronB’s intellectual dishonesty makes clear that one should never expect him to be intellectually honest

    and the complete quote from my comment 203

    His extreme reluctance to admit error even in such an obvious and well documented case as I presented in Exhibit A of AaronB’s intellectual dishonesty makes clear that one should never expect him to be intellectually honest in a case where even a little bit of uncertainty or wiggle room is available.

    I followed never with a qualifier. Not very intellectually honest of you to omit that. Not to mention the way omitting the part ending with “Exhibit A” causes your excerpt to be a misrepresentation of what I wrote.

    I think that is a good enough example to qualify as Exhibit A of utu’s intellectual dishonesty. Thanks for being so gracious as to provide it.

  204. AaronB says:
    @utu

    I would love to calculate my g for intellectual dishonesty – it would be derived from the intellectual dishonesty I display in all the various subjects I discuss.

    However, it would only explain 40% of the variance between mine and res’s intellectual dishonesty, so may not be so informative.

    However, since I’m intellectually dishonest, ill just go around claiming my IHQ g is an accurate measure of my intellectual dishonesty, and all will be well 🙂

  205. AaronB says:
    @res

    You are so triggered 🙂

    • LOL: res
  206. CanSpeccy says: • Website

    It is amazing how Binet’s simple test for mental retardation has resulted in a supposed science of intelligence measurement, which asserts that (a) intelligence can be measured on a single linear index, and (b) that the essence of intelligence, named g, can be deduced mathematically from the correlations observed among tests of different cognitive capacities.

    In fact, intelligence is clearly multi-faceted, and neurological research shows that the facets of intelligence depend on a multitude of specialized brain modules. Moreover, a large number of genes regulate brain development, which means with near certainty that the development of each brain module is in some degree independent of the development of other brain modules. This fact indicates potential for rapid evolutionary adaptation of specific cognitive capacities in response to selective pressures leading to what is observed, namely, substantial variation among individuals in relative cognitive capacities.

    Thus intelligence cannot be one thing, but a collection of abilities or intelligences that vary independently among individuals. This claim the IQ “scientists” counter by triumphantly pointing to the g factor, a measure of the correlation among test scores on tests for different cognitive capacities. However, what these scientists rarely acknowledge is that the correlations observed among scores on tests of different cognitive capacities are low, explaining on average less than 10% of the variation in one cognitive ability by variation in any other cognitive ability. Thus, the observed values of g, far from proving the existence of some mystical intelligence factor measured by an IQ test, prove that intelligence is very far from being a unitary property of mind. In fact, an IQ score is no more than a summation of scores on tests for a variety of aptitudes, a summary that may conceal genius or obscure mental deficits, by the conflation of results of multiple tests.

    As for what g represents, the answer must, at least in part, be the condition of the common substrate of intellectual activity; chiefly, that is, brain cells with their membranes, their energy systems, their secretory and electro-chemical properties, etc., upon which all brain modules depend. Thus, for example, a general impairment of cellular energy metabolism, due say to favism, by which Pythagoras was apparently afflicted, will cause a correlated decline in all mental faculties. The g-factor, is thus not a measure of the essence of intelligence, but potentially, merely an index of the consumption of beans.

    • Replies: @AaronB
  207. AaronB says:
    @CanSpeccy

    The g-factor, is thus not a measure of the essence of intelligence, but potentially, merely an index of the consumption of beans.

    By far the most intelligent thing I have heard about g in a while.

    • Replies: @CanSpeccy
  208. utu says:
    @AaronB

    Few things about SAT V and M.

    (1) Panta Rhei: Whatever they try to measure keeps changing.

    https://en.wikipedia.org/wiki/Math–verbal_achievement_gap
    Around 1990 the national average on the math portion of the SAT began its slow but steady ascension over the national average for the verbal portion. It took only one decade for the math average to eclipse the verbal average, continuing to widen since that point. The difference is sizable and significant. The average SAT test-taker today produces a math score that is 13 points higher than their verbal score. However, this was not always the case. In fact, in the 1970s the relationship was precisely the opposite when national verbal scores routinely trumped the national math average by similar margins.

    (2) Ability to write Chinese makes difference on math score.

    https://www.researchgate.net/publication/225681364_Writing_Chinese_and_mathematics_achievement_A_study_with_Chinese-American_undergraduates
    “writers and non-writers of Chinese characters from the Chinese-American group in the United States were selected to participate in this study”

    “Overall, the Chinese-American group scored higher on SAT-Math (M = 638, SD = 108) than SAT-Verbal (M = 548, SD = 111). The results do not support the alternative interpretation that Chinese_writers scored high on SAT-Math may be due to their high verbal ability as measured by SAT-Verbal.”

    “Writing Chinese was significantly correlated with SAT-Math but not correlated with SAT-Verbal.”

    “For the Chinese-American sample, the gender difference on SAT-Math was not significant (see Table 3). Females scored slightly higher on SAT-Math than did males.”

    (3) SAT-V and SAT-M are correlated.

    The correlation between V and M is r=0.6 according to Cohen, R at al. (1992). Psychological testing and assessment: An introduction to tests and measurement. Mountain View, CA: Mayfield.. and here

    https://www.statcrunch.com/5.0/viewreport.php?reportid=21828

    on small sample N=162 it was r=0.68. This means that V and M explain about 40% (res’s Kabbalah number again) of each other’s variance. The SAT=M+V or a weighted sum SAT=xM+(1-x)V where 0<x<1. For some value of x SAT would be the first principal component of the battery of two tests (V,M). This principal component will not be able to explain much more than 40% of variance in the data of the battery (V,M). At the same time it is claimed that SAT is a proxy for IQ or some g with correlations from 0.7 to 0.8 which means that 35%-50% of variance in SAT remains unexplained by IQ or g. One can pose a question what explains the remainder of the variance which is quite significant. The other question is whether the claimed correlations are valid.

    (4) Tweaking correlations to make them bigger.

    Spearman came up with two methods to boost correlations by correcting the effects of (1) limited range and (2) attenuation. Corrections are as valid as the assumptions behind them. Limited range correction requires making assumptions on data behavior outside of empirical accessibility. In an extreme case one can imagine appending to data results of SAT and IQ tests taken by illiterate subjects. This would boost the correlation.

    The attenuation correction formula by Spearman is very enthusiastic as it is capable to make correlation greater than one. So everything is possible for a motivated researcher. Correlations that are broadcasted by the IQists are the corrected one usually often w/o disclosure that they were corrected or only in the small print. For instance in the famous Sottish study of IQ stability between childhood and the old age it is claimed that correlation is about 0.8 however before correction for the attenuation the childhood IQ could explain only about 40% (again res's Kabbalah number) of IQ variance in the old age. I suspect but I could not verify it that various corrections might be responsible for large heritability results from the twin studies. I hoped that Dr. J. Joseph (http://www.jayjoseph.net/publications) who is on the warpath against the IQ twin studies looked into the boost of correlations but I did not find it in his book.

    • Replies: @AaronB
    , @res
    , @res
  209. @Jay

    I assume, that Bernard B Davis in his here at times quoted book Storm Over Biology (1986) meant that: At least he wrote it down explicitly: Gould was a poor scientist: “Moreover, while he is admired as a clear writer, in the sense of effective communication, he is not a clear, in the deeper sense of analyzing ideas sharply and with logical rigor, as we have the right to expect of a disciplined scientist.” (p. 129)

  210. AaronB says:
    @utu

    Fascinating stuff.

    1) But surely the relative performance on V and M could not have reversed themselves in the population, much less rather suddenly in a short period of time? The IQists inform us that intelligence is innate, and is not responsive to environmental changes, training, or anything like that (in developed societies with adequate nutrition).

    1)Ability to write Chinese = maybe a proxy for less Americanization, and more Chinese acculturation. Similarly, Chinese who speak worse English, I have observed, are thinner and dress better than Chinese who speak better English. But it couuld also be something about how our minds process the Chinese language, of course.

    (another minor IQist discrepancy – females scored slightly higher on math? Aren’t women supposed to be worse at math? This is trivial, though.)

    2)Correcting the correlations = “Saving The Appearances”. Similar to recalibrating the Raven Matrices. Indeed, a motivated researcher can find anything.

    3) So only 35%-50% of SAT variance unexplained by IQ or g. As I thought, SAT and IQ do not correlate well. However, IQists inform us that this level of explanation is actually an extremely high corelation between IQ and SATs, so I suppose we have to trust them. They seem for some reason extremely happy with below 50% variance explained and think of it as a normal, average person might think of 100% explained variance.

    4) 40% – hmmm, that mystical number again. Very interesting.

    5) So even the stability of IQ into old age had to be tweaked. Not surprising. Funny what you find when you look.

  211. CanSpeccy says: • Website
    @AaronB

    Re: Pythagoras and beans

    It’s not just beans. Alcohol will surely have an impact on g, while nicotine is known to have a “remarkably large effect” on IQ. I guess that’s why I’ve not felt quite normal for the last 50 years, ever since I gave up a three-pack-a-day cigarette habit. And according to the late Linus Pauling, a spoonful of aspartate can raise your IQ by 25 points, suggesting that if we wished, we could all be geniuses, more or less.

    • Replies: @AaronB
    , @res
  212. res says:
    @utu

    I like that your first reference is a wiki page which leads with a box describing multiple issues with it.

    Regarding

    In fact, in the 1970s the relationship was precisely the opposite when national verbal scores routinely trumped the national math average by similar margins.

    Be careful about trusting current data there. They have run those old scores through the wringer of one or more conversions to attempt to equate them with today’s scores.

    Let’s look at how the unconverted scores appeared before the 1995 SAT recentering. From my favorite reference for pre-1995 recentering SAT score data, Dorans 2004: The Recentering of SAT® Scales and Its Effects on Score Distributions and Score Interpretations
    https://www.ets.org/Media/Research/pdf/RR-02-04-Dorans.pdf

    The real decline in SAT scores did not start until after the Wilks report was issued. Shortly after the Wilks report, from about 1963 until 1980, both SAT V and SAT M means dropped noticeably from about 475 for SAT V to around 425, and from about 500 to 470 for SAT M. Now the difference in SAT V and SAT M mean scores was close to 45 points. By 1990, the SAT M mean had increased to near 475, while the SAT V mean remained around 425, a 50-point difference.

    Except for the famous score decline of the mid-1960s to late-1970s, SAT mean scores have been remarkably stable. Prior to this dramatic decline, mean SAT V scores for all test-takers on the 1941 scale ranged in the 470s from 1951-52 to 1965-66, while mean SAT M scores during that same time ranged from 490 to 502. From 1980 until 1995, mean scores on the 1941 scale for the College Bound Senior Cohort have ranged from 422 to 431 for SAT V and 466 to 482 for SAT M. Outside of the period of the decline studied by the score decline panel and reported in On Further Examination, SAT means have been remarkably stable.

    The decline halted by 1980. By then there was a definite need to realign the verbal and math scales. The SAT V and SAT M averages were 50 points apart. And there was a clear need to repopulate the top end of the score scale, especially for SAT V.

    So during the 1970s SAT math scores were actually higher than verbal at the time (contrary to your source). A rather different perspective there. I am quite interested in which perspective more accurately reflects reality. Does anyone have a thoughtful take on that? Preferably with some supporting evidence?

    Worth mentioning that the whole point of having separate math and verbal SAT subtests is to try to measure two different abilities as much as possible.

    But please, don’t let me interfere with your poo flinging.

  213. AaronB says:
    @CanSpeccy

    There are other nicotine delivery sources. Try Swedish Snus. Its harmless. Or smoke a pipe. I’m thinking of taking up pipe smoking myself. Then I will intellectually crush res like the bug he is, even more so than now 🙂 Watch out, res!

    Alcohol probably has an ambiguous effect on mental acuity, expanding the imagination and range of ideas, while maybe slightly dulling sharpness.

    • Replies: @res
  214. res says:
    @AaronB

    It’s good to have pleasurable fantasies. Maybe you can do that crushing in your dreams tonight. Since it is certainly not happening here ; )

  215. AaronB says:
    @res

    Asians are notorious for putting glutamate in everything they eat, especially the Japanese.

    We have discovered the source of their IQ advantage!

    I myself love umami and seek out foods with high glutamate content, and often add glutamate to my foods, which no doubt accounts for my intellectual superiority to res.

    Res, do you eat lots of glutamate? I bet you don’t, and that explains your performance here.

    But the remedy is simple, res. Glutamate and tobacco will do the trick.

    • Replies: @res
    , @CanSpeccy
  216. res says:
    @AaronB

    Who exactly is triggered here?

  217. CanSpeccy says:
    @res

    Did you mean aspartate or glutamate?

    Ha! Right. Now I see why this aspartate stuff I’ve been taking by the bucket-load isn’t working for me. I’ll switch to glutamate. Thanks for the link.

    Although aspartate is also a neurotransmitter and seems to have some relation to IQ. Thus from a journal with the puzzling (to me) title of Neurolmage ( a Journal of Brain Function), the following article:

    N-acetylaspartate concentration in corpus callosum is positively correlated with intelligence in adolescents.

  218. res says:
    @utu

    Let’s take a closer look at point 3.

    (3) SAT-V and SAT-M are correlated.

    The correlation between V and M is r=0.6 according to Cohen, R at al. (1992). Psychological testing and assessment: An introduction to tests and measurement. Mountain View, CA: Mayfield.. and here

    https://www.statcrunch.com/5.0/viewreport.php?reportid=21828

    on small sample N=162 it was r=0.68. This means that V and M explain about 40% (res’s Kabbalah number again) of each other’s variance. The SAT=M+V or a weighted sum SAT=xM+(1-x)V where 0<x<1. For some value of x SAT would be the first principal component of the battery of two tests (V,M). This principal component will not be able to explain much more than 40% of variance in the data of the battery (V,M). At the same time it is claimed that SAT is a proxy for IQ or some g with correlations from 0.7 to 0.8 which means that 35%-50% of variance in SAT remains unexplained by IQ or g. One can pose a question what explains the remainder of the variance which is quite significant. The other question is whether the claimed correlations are valid.

    Regarding this example of utu’s pseudo-math. Rather than handwaving pseudo-math let’s take a look at some better data (N= 103,525 from the College Board rather than N=162 from wherever). In comment 157 I cited this source of SAT and ACT correlations: https://www.ets.org/Media/Research/pdf/RR-99-02-Dorans.pdf

    There we see the following correlations:
    Test SATM SATV SATV+M
    SATM 1.0 0.71 0.93
    SATV 0.71 1.0 0.92
    SATV+M 0.93 0.92 1.0

    Focusing on this assertion: “This principal component will not be able to explain much more than 40% of variance in the data of the battery (V,M)” And where exactly did that 40% number come from? Proof by assertion? If you did want to assess the ability of SATV OR SATVM to predict both then you would have something like each one explaining 100% of its own variance and 50% of the other (0.71^2). So about 75% of the total variance. But the combined predictor turns out to be even better.

    With only two variables the first PC will be on the axis of V+M. The second PC will be on the axis of V-M.

    So from the correlations above we have SATV+M explaining 85% of the SATV variance and 86% of the SATM variance. Rather more than “40%.”

    Oops. Still trust utu’s math, AaronB?

    P.S. Now the question becomes: is utu intellectually honest enough to admit his error?

    • Replies: @AaronB
    , @utu
  219. CanSpeccy says: • Website
    @AaronB

    Nicotine, tobacco, they’re not good. They kill the emotions. But it’s an interesting question whether the Asian IQ advantage versus Europeans reflects the relative nutritional qualities of soy sauce versus ketchup.

    • Replies: @AaronB
  220. AaronB says:
    @res

    I don’t know math, but I do know simple logic. You are so confused, res.

    If you did want to assess the ability of SATV OR SATVM to predict both then you would have something like each one explaining 100% of its own variance and 50% of the other (0.71^2). So about 75% of the total variance.

    We are trying to find a common factor underlying both V and M.

    That V or M explain 100% of their own variance does not help us find the common factor shared by both of them.

    For that, only how much one can explain the variance of the other is relevant. Because only that can help us find the common factor.

    So lets say V alone can explain 75% of variance involving (V+M), this is at least partly tautaological, as (V+M) contains V. Of course V can explain 100% of its own variance!

    But precisely for that reason, it is useless in helping us find the common factor.

    Step back for a moment, and think, res.

    • LOL: res
    • Replies: @res
  221. AaronB says:
    @CanSpeccy

    Nicotine kills the emotions? Interesting. I would be interested to read about that, if you have any links.

    Many great artists and writers used to smoke pipes, cigars, and cigarettes, so I’m surprised.

    • Replies: @CanSpeccy
  222. res says:
    @AaronB

    Since you are obviously too stupid (and/or too intellectually dishonest) to understand my original presentation, let’s try again. In the SAT that common factor is the combined score. Let’s take another look at those variance explained numbers I computed from the correlations.

    So from the correlations above we have SATV+M explaining 85% of the SATV variance and 86% of the SATM variance.

    So we have a common factor (SAT M+V) explaining 85% or more of the variance of each of its component subtests.

    It is entertaining to watch high V low M people try to argue about math. I expect you would find it embarrassing if you were less shameless.

    • Replies: @AaronB
    , @AaronB
  223. AaronB says:
    @res

    Lol, no.

    The ability of one factor to explain the total variance (two factors combined) – obviously, since half of the total variance is that one factor, it will be higher.

    That’s tautological. The added explanatory power merely reflects the fact that each factor explains itself.

    Meaningless when trying to find a common factor.

    V=G+V and M=G+M

    The factor common to both is G.

    Obviously, V will explain more of the variance in (V+M) than in just M alone, because both parts of V are included in (V+M), and only one part of V is included in just M.

    Since we want to find out the one part of V that is shared by M, calculations that depend on both parts of V do not suit our purposes.

    Its exceedingly simple res.

    Now stop playing games and get serious 🙂

    • LOL: res
    • Replies: @res
    , @utu
  224. AaronB says:
    @res

    It is entertaining to watch high V low M people try to argue about math

    And anyways, res, since V is significantly more g-loaded than M, and high SAT verbal is significantly harder to obtain than high SAT math, the Gods of IQ decree that you genuflect before me, your g- loaded superior.

    I am your elder in g. Show some deference.

    • Replies: @res
  225. res says:
    @AaronB

    You are so proud of your poo flinging. Fascinating.

  226. res says:
    @AaronB

    You might consider that some people are high V and higher M.

  227. CanSpeccy says: • Website
    @AaronB

    Many great artists and writers used to smoke pipes, cigars, and cigarettes, so I’m surprised.

    Probably because, being hyper-emotional, they found tobacco a useful tranquilizer.

    The tranquilizing effect of nicotine is touched on indirectly in the movie “Heavens Above” (starring Peter Sellers), based on a idea of Malcolm Muggeridge’s, in which an Anglican parish priest sets out to follow in the footsteps of our Lord, aided with massive financial aid from the heiress to a pharmaceutical company fortune.

    Naturally, the result is the exact opposite of heaven on earth. But the thing is, in the context of the present discussion, that the pharmaceutical company’s profits were based on a product named with Muggeridge’s love of paradox and reflective of his training in biology: Tranquilax, “the Three-in-one Tranquilizer, Stimulant and Laxative.” That is, in fact, a fair summary of the chief effects of nicotine. Further, as I wrote somewhere, the stimulus, i.e., the increased heart rate, blood sugar level, etc. (i.e., flight or fight response) acts as a negative feedback on emotional arousal. As for the laxative effect, that is quite well known.

    • Replies: @AaronB
  228. utu says:
    @res

    If Var(V)=Var(M)=1 and if Cor(V,M)=r and if we define Z=V+M then Var(Z)=2+2r and Cov(V,Z)=Cov(M,Z)=1+r. Then Cor(V,Z)=Cor(M,Z)= sqrt((1+r)/2). This means that even for r=0 correlations, Cor(V,Z)=Cor(M,Z)=1/sqrt(2)≈0.71, are large.

    Note that Dorans in Table A.4 for Cor(V,Z) and Cor(M,Z) gets 0.92 and 0.93 while the numbers should be equal if indeed his SAT I V+M equals to (SAT I Verbal)+(SAT I Math). This must be a numerical rounding off error and and also an indication that Doran is a mindless number cruncher. And you, res, did not catch it?

    If Cor(V,M)=0.6 as in Cohen et al.

    “The high correlation (r≈ 0.60) between SAT-Verbal and SAT-Math (Cohen, Swerdilik, & Smith, 1992)”
    https://www.researchgate.net/publication/225681364_Writing_Chinese_and_mathematics_achievement_A_study_with_Chinese-American_undergraduates

    or if Cor(V,M)=0.68 as in

    https://www.statcrunch.com/5.0/viewreport.php?reportid=21828

    then Cor(V,Z)=Cor(M,Z)=0.894 or 0.916. Which means that Z explains about 80% – 84% of variance of V and of variance of M. Is this contradictory with the following statement?

    This principal component will not be able to explain much more than 40% of variance in the data of the battery (V,M).

    Yes, absolutely. I took one too many mental shortcuts.

    Let’s calculate the variance explained of a 2×2 correlation matrix [1,r][r,1]. The eigenvalues are l1=1+r and l2=1-r. The variance explained by l1 is l1/(l1+l2)=(1+r)/2 and by l2 is l2/(l1+l2)=(1-r)/2. These variances are for the eigenvectors V+M and V-M (which are orthogonal), respectively. Note that for a correlation r=0 (which is the case of the lowest possible variance explained in a 2×2 correlation matrix) the variance explained is 50%. You can’t do worse than 50%. In 4×4 matrix you can’t do worse than 25%.

    Thank you, res. If you ever need a recommendation letter for a job of a proof reader I will be more than happy to provide it.

    • LOL: res
    • Replies: @res
  229. AaronB says:
    @CanSpeccy

    Ah, I see what you mean. Nicotine definitely is a tranquilizer, and I suppose that isn’t good for everybody.

  230. res says:
    @utu

    So you make a mistake resulting in a greater than a factor of 2 error and then you try to wave it away as proof reading. LOL!

    This will serve as Exhibit U (for utu, since it represents him so well) of utu’s intellectual dishonesty.

    And I am only beginning to realize how apt “hand-waving pseudo-math” is as a description of your “contributions” here. Though to your credit, taking a quick look at your math in that comment it looked OK to me.

    BTW, to be clear, the lack of ability to detect even such a gross error as that is a big part of why I disparage utu’s hand-waving pseudo-math. He makes careless assumptions then manipulates equations in a cavalier fashion without any effort towards rigor or checking for correctness. And his “one (at least!) too many mental shortcuts” introduce far too much opportunity for difficult to detect errors to creep in.

    And why do you prefer your r = 0.6 and r = 0.68 numbers over the much larger N (~3 orders of magnitude!) and more recent value of 0.71 I cited from Dorans? Consider this a sub-component of Exhibit U.

    Regarding

    Note that Dorans in Table A.4 for Cor(V,Z) and Cor(M,Z) gets 0.92 and 0.93 while the numbers should be equal if indeed his SAT I V+M equals to (SAT I Verbal)+(SAT I Math). This must be a numerical rounding off error and and also an indication that Doran is a mindless number cruncher. And you, res, did not catch it?

    It is fascinating that you consider a 1% discrepancy more important than your >2x error. As far as why, it is probably due to round-off, but I think calling it an error is a bit much. Would you have preferred that Dorans change the results from his data to match theory? And yes, I noticed it, but considered it not worth commenting on.

    And

    Note that for a correlation r=0 (which is the case of the lowest possible variance explained in a 2×2 correlation matrix) the variance explained is 50%. You can’t do worse than 50%.

    Right. And that is exactly what should have raised a red flag when you arrived at the 40% variance explained number. I think that shows clearly just how much your characterization of Dorans as a “mindless number cruncher” is projection on your part. That kind of cross check is exactly the sort of thing that people who go beyond mindless number crunching (or mindless equation manipulating) do.

  231. utu says:
    @AaronB

    The added explanatory power merely reflects the fact that each factor explains itself.

    There is something to it. If you have two variables X and Y and if they are uncorrelated (r=0) then the variable Z=X+Y as the first principal components of the correlation matrix [1,r][r,1] explains 50% of variance of the data set composed of X and Y (*). The correlation of Z with X and with Y is then 0.707. Some mindless ‘factor analyst’ may triumphantly exclaim: look, I found g that explains 50% of data but in the case of 2×2 matrix it is trivial, pointless and obviously misleading claim because it can’t be less than 50%. In case of 4×4 matrix it can’t be less than 25%. Clarification: The lower order eigenvectors can explain less but the dominant eigenvector can’t explain less than 1/n where n is number of variables.

    When r increases then the explained variance is larger than 50% and when r reaches 1 then the explained variance is 100% but again this is rather meaningless because then X must be equal to Y, so obviously X explains Y 100% as Z=X+Y=2X=2Y.

    So it seems that at the two extremes when the explained variance is either 50% or 100% (2×2 matrix) then nothing interesting is being said because the cases are trivial. Could one argue that the explained variance of 75% would be in some sense more interesting than the explained variance of 90% because 90% is closer to the meaningless 100%?

    (*) Clarification. What do we mean when saying “variance of the data set composed of X and Y”? When variables are normalized then the variance of the data set is the number of variables because the variances is the sum of all variances.

  232. AaronB says:

    The correlation of Z with X and with Y is then 0.707. Some mindless ‘factor analyst’ may triumphantly exclaim: look, I found g that explains 50% of data

    I confess to not quite following here.

    Yes, since Z is composed of X and Y, it necessarily follows that Z correlates with both. It is the summation of both, it must correlate.

    If X and Y correlate at r=0, and Z correlates to X and Y as it must, isn’t that correlation just a reflection of the fact that Z includes X and Y? This tells us something about the relationship between Z and its component variables, but why does this tell us anything about the relationship between the component variables?

    And if r=0, then Z explaining at least 50% of the variance between X and Y – is this because Z is composed of two variables, and we divide by minimum 2 in the absence of any greater number provided by a higher than zero correlation?

    And that’s why a 4×4 matrix, we divide by minimum 4, to get 25%?

    But why would Z tell us anything about the variance between X and Y, I do not fully understand.

    When r increases then the explained variance is larger than 50% and when r reaches 1 then the explained variance is 100% but again this is rather meaningless because then X must be equal to Y, so obviously X explains Y 100% as Z=X+Y=2X=2Y.

    Yes, if each factor explains 100% of the other, then obviously they are just the same thing. There would be no shared factor and unshared factor, no g+special factor. There would be total identity.

    I guess this is what I was sort of getting at by tautological.

    Could one argue that the explained variance of 75% would be in some sense more interesting than the explained variance of 90% because 90% is closer to the meaningless 100%?

    Certainly, because at least here there is an unshared factor, which allows us to tease out the shared factor, g.

    • Replies: @utu
  233. utu says:
    @AaronB

    The cases of 50% and 100% are not interesting when it comes to factor analysis of 2×2 correlation matrix because (1) 100% means you have just one variable not two so there is nothing to factorize and (2) 50% means that the two variables are uncorrelated and thus they themselves are orthogonal factors so no factor analysis is necessary either. Furthermore 50% is a lower bound, you can’t lower than that, so getting excited that your factor can yield, say 55% is not a big deal because it means that your are close to the lower bound and the two variables are pretty much independent and they do not have much in common.

    I brought this up only to add some perspective on how to see the numbers to avoid the unwarranted excitement and hype and also because I have just finished responding to res where I had to figure out correct formulas for the 2×2 case with pen and paper, so it was still fresh in my mind.

    Sayonara. I have to leave you now for the seppuku ceremony because res caught me making a factor of two error. Though, wait a minute, on the second thought the factor of two error when the erring is on the low side is only 50%, so skipping the desert this evening, I think, should be an adequate punishment for 50% error.

    • Replies: @AaronB
  234. AaronB says:
    @utu

    The cases of 50% and 100% are not interesting when it comes to factor analysis of 2×2 correlation matrix because (1) 100% means you have just one variable not two so there is nothing to factorize and (2) 50% means that the two variables are uncorrelated and thus they themselves are orthogonal factors so no factor analysis is necessary either. Furthermore 50% is a lower bound, you can’t lower than that, so getting excited that your factor can yield, say 55% is not a big deal because it means that your are close to the lower bound and the two variables are pretty much independent and they do not have much in common.

    Thank you. This makes sense.

    so skipping the desert this evening, I think, should be an adequate punishment for 50% error.

    As long as you’re still able to live with yourself.

  235. Factorize says:

    An academic scandal is breaking news in the US. Cheating, why bother? Finding one’s psychometric niche in life leads to happiness and success: not finding one’s psychometric niche leads to misery and failure. On life’s IQ test, cheating is the wrong answer. Isn’t this right up there on the psychometric taboo list with breaking into prison? Breaching security of Nerd U, simply would entitle normies to all night coding sessions, interactions with those on the spectrum and vitamin D deficiency. Hey, that sounds great to me! To others, though it would be torture, in this instance self-torture. Cheating doesn’t pay!

  236. Factorize says:

    This scandal is rich in psychometric meaning. Some of these parents were willing to pay almost 7 million dollars in order to merely create an impression of cognitive competence? Inflating one’s SATs by 1000 points would create a psychometric incongruity with observable ability much too large to overlook. Fair warning has now be given for the lengths parents might be expected to go to give their children a genetic advantage in life. What will be the response 20 years from now when children with 7 million dollars of genetic enhancement of IQ no longer need to cheat to be admitted to college? It should be obvious to all that demand for IQ uplift by parents who want their kids to have an edge will be substantial. A blog for this scandal would be welcome.

  237. “Now the redoubtable Russel Warne” (JT) has published a compact article about Stephen J. Gould in the online magazine Quillette

    https://quillette.com/2019/03/19/the-mismeasurements-of-stephen-jay-gould/

  238. @nickels

    If anyone ever showed evolution for the scam it is, Gould is the man.

    Gould did nothing of the sort, no matter how fervently creationists wish he had.  Gould grossly misrepresented evolution, in part by claiming that it did not work above the neck in Homo sapiens sapiens; his denial of racial differences in skull volume and the entire concept of IQ was Marxist, not scientific and certainly not Darwinian.  Darwin wrote The Descent of Man which contradicts practically every claim Gould made about the different strains of genus Homo and arrives at conclusions Gould considered anathema… despite them being factually true and irrefutable.

    S. J. Gould was a red diaper baby whose publicly-held opinions on many issues were fixed by Marxist ideology.  As anti-Christian as he was, you should be careful of using him for support.  Like the severed head of a snapping turtle, he can still bite you.

Current Commenter
says:

Leave a Reply - Comments on articles more than two weeks old will be judged much more strictly on quality and tone


 Remember My InformationWhy?
 Email Replies to my Comment
Submitted comments become the property of The Unz Review and may be republished elsewhere at the sole discretion of the latter
Subscribe to This Comment Thread via RSS Subscribe to All James Thompson Comments via RSS