I watched the recent PBS documentary on The Mamas and the Papas and they must be right up there with The Beatles as one of the greatest bands of all time. What an incredible collection of music for a group that was only together for a couple of years.
It’s amazing how young they looked in the 1960s. I had a professor who argued people didn’t get old with age, it just appears that way because older cohorts were always wrinkled and grey.
The genius of the group was probably John Phillips (more evidence that height and IQ are correlated) for writing those catchy tunes, but watching the documentary, it was Mamma Cass that stole my heart.
As a fat Jewish girl during a more anti-Semitic era, she must have wanted so desperately to fit in with these cool white kids, and I’m so glad they accepted her, as did America. Cass’s incredible voice added real value to the culture, as did her creative body language. I love when she introduces the song Creeque Alley and says “cue the tape.” Watch how she opens her moth wide and starts bouncing up and down in excitement for the song about to be played (Very Oprah!).
One of the things we learned in the documentary is originally, Cass did not even want to appear on stage with the slim blond blue eyed Michelle Phillips and would only sing from the back of the room. What an ego boost that must have been to Michelle, “you’re so beautiful, I feel ugly being seen with you.” Did Cass really have such low self-esteem, or was she a master manipulator who knew exactly how to make others like her?
After The Mammas and the Pappas broke up, Mamma Cass started a solo career, and probably wondered, who’s going to turn out to see a fat Jewish girl sing solo? The show sold out, two nights in a row! She was so excited she phoned Michelle Phillips to share the happy news. That night she died of a heart attack in her sleep at the age of only 32.
“I do know one thing” said Michelle Phillips. “Cass Elliot died a very happy woman”.
At that point I was so overwhelmed with emotion I had to turn the documentary off.
I’m rerunning a post I wrote back in May because I gained access to some very recent data on the black-white IQ gap (the demographic breakdown of the WISC-V norms which were collected between April 2013 and March 2014 across 49 states). The post has been updated and revised to include the WISC-V data.
The black-white IQ gap fascinates me. The fact that these silly little IQ tests, consisting of childlike jig-saw puzzles, blocks, and cartoon pictures, could measure a one standard deviation gap between American blacks and whites, and the fact that this gap has persisted, generation after generation, throughout an entire century of unprecedented social, cultural and political change, is one of the landmark achievements of science. Even as the raw performance of all Americans on IQ tests has increased by roughly two standard deviations over the 20th century because of better environments, (the Flynn effect), the 1 SD gap between American blacks and whites has stayed constant.
Such incredible consistency over a century of massive social change has caused some to argue that the black-white IQ gap is largely, or perhaps entirely genetic.
But increasingly a consensus is emerging that the black-white IQ gap is finally starting to shrink. You’ll often read that that the black-white IQ gap is no longer one standard deviation (15 points), but is now closer to 10 points. Scholars J.P. Rushton and Arthur Jensen denied the gap was shrinking, but since their deaths, the anti-HBDers have been able to monopolize the debate unchallenged.
Is the black-white IQ gap still roughly 15 points or has it indeed shrunk? I decided to focus on the Wechsler intelligence scales because these are the gold standard of IQ testing, are meticulous in their sampling procedure, and they most resemble the WWI IQ tests that first measured the black-white IQ gap a century ago.
The black-white IQ gap in children
As the chart below shows, the black-white IQ gap has shrunk substantially in children. In the 1972 and 1989 standardizations of the Wechsler Intelligence Scale for Children, whites outperformed blacks by 15.9 IQ points (U.S. norms) but by the time the WISC-V was normed in 2013, the gap had shrunk to 11.6 points.
Of course because of mass immigration, it’s more meaningful to express IQ using white norms, not American norms, since the U.S. mean continues to fall below the white mean. Relative to Americans as a whole, Whites have gone from being average Americans to joining Asian and Jewish Americans, as another high IQ sub-group.
white iq (u.s. norms)
black iq (u.s. norms)
white iq (white norms)
black iq (white norms)
black-white iq gap (u.s. norms)
black-white iq gap (white norms)
wisc-r (1972)
102.3 (sd = 14.08)
86.4 (sd = 12.63)
100 (sd = 15)
83 (sd = 13.46)
15.9
17
wisc-iii (1989)
103.5 (sd = 13.86)
88.6 (sd = 12.83)
100 (sd = 15)
84 (sd = 13.89)
14.9
16
wisc-iv (2002)
103.2 (sd = 14.52)
91.7 (sd = 15.73)
100 (sd = 15)
88 (sd = 16.25)
11.5
12
wisc-v (2013)
103.5 (sd = 14.6)
91.9 (sd = 13.3)
100 (sd = 15)
88 (sd = 13.66)
11.6
12
The black-white IQ gap in adults
The bad news is these large gains in black kids have not been matched by gains in black adults. When the Wechsler Adult Intelligence Scale (Revised) (WAIS-R) was normed in 1978, whites outscored blacks by 14.6 points (U.S. norms). By the time the WAIS-IV was normed in 2006, the gap was 14.5 points.
white iq (u.s. norms)
black iq (u.s. norms)
white iq (white norms)
black iq (white norms)
black-white iq gap (u.s. norms)
black-white iqgap (white norms)
wais-r (1978)
101.4 (sd = 14.65)
86.8 (sd = 13.14)
100 (sd = 15)
85 (sd = 13.45)
14.6
15
wais-iii (1995)
102.6 (sd = 14.81)
89.1 (sd = 13.31)
100 (sd = 15)
86 (sd = 13.48)
13.5
14
wais-iv (2006)
103.2
88.7
100 (sd = 15)
14.5
Discussion
The roughly 15 point black-white IQ gap, which has been observed since WWI, has clearly shrunk at least slightly in children, yet remains roughly unchanged in adults. There are several explanations for this. The most obvious is that adult IQs are a delayed indicator; adult IQs tell us how well children scored several decades ago while children IQs tell us how well adults will score several decades in the future.
A more gloomy interpretation is that the black IQ gains (relative to whites) are ephemeral and will fade as the children become adults. One possibility is that affirmative action, political correctness and anti-discrimination laws of the last several decades allowed a critical mass of blacks to enter the middle and upper classes, and their children, raised in these better environments from birth, made gains on IQ tests. However IQ gains caused by the social environment tend to fade completely by later adulthood (at least within groups), so it remains to be seen if the adult black-white IQ gap will narrow.
Another explanation for the narrowing black-white IQ gap in children is structural changes to the test itself. Black tend to do worst on measures of spatial ability, and hard-core spatial tests were 20% of the WISC-R (Block Design and Object Assembly) but only 10% of the WISC-IV (only Block Design). Meanwhile blacks do relatively well on low g measures of clerical speed. These were only 10% of the WISC-R (Coding) but 20% of the WISC-IV (Coding and Symbol Search). Not sure about the WISC-V.
A final possibility is that the narrowing of the black-white IQ gap in children has a genetic explanation. With the vanishing taboo against interracial marriage, the average level of white and Asian admixture in “black” American children has likely reached record highs, as has the level of non-white admixture among American “whites”. Also, with the rise of immigration, many black Americans are not “African Americans” (descendants of slaves), but “American Africans” (descendants of Africa’s elite who voluntarily migrated to the West). President Obama, the mixed race son of an elite African immigrant, embodies both of these trends.
Conclusion
Despite unimaginable social change, there’s no evidence that the roughly 15 point black-white IQ gap among American adults has changed in the last 100 years. There’s some evidence that the future generation of American adults will have a reduced black-white IQ gap, but even this small reduction might be spurious because structural changes in the tests and the genetic watering down of racial differences.
I haven’t forgotten about reader requests to estimate the IQs of Eminem, Ben Shapiro, the Founding Fathers or even Clark Ashton Smith, but with Halloween only weeks away, I wanted to write a quick post about GG Allin while he’s still fresh in my memory (hat-tip to commenter jeanbedelbokassa for suggesting this person).
Identified with serial killers
In studying GG Allin over the past week or so, I was struck by his claim that if he hadn’t been an entertainer, he would have been a serial killer and was reportedly friends with John Wayne Gacy. Of course I should stress that I’m in no way equating an artist like GG to an actual killer, but there are aspects of GG’s childhood and adult persona that remind me of some of the most grotesque murderers in recent history.
When heavy metal director and GG Allin fan Rob Zombie remade John Carpenter’s classic Halloween, he angered fans of the original by turning villain Michael Myers from a quiet suburban boy, into a white trash sexually ambiguous bullied heavy metal fan, but Zombie based the character on real life cases.
In an interview with Vanity Fair, Zombie explained what he knew about serial killers:
I love ’em all. Not, you know, as people or anything, but they all make for great stories. I think Henry Lee Lucas is probably one of my favorites.
He added:
He and his buddy Ottis Toole were just a couple of deranged rednecks. But given his upbringing, y’know, it’s just not that surprising. Some of these guys, you think, “What would make a person do something like this?” And then you read about their upbringing and you’re like, “Oh, okay, well I guess that might do it.”
He also said:
I’ve read so many books about these guys, I start confusing their backstories. But with Henry and Ottis, I remember it was pretty horrible. Stripper moms, alcoholic dads, I think they were both forced to dress up like girls at some point. Henry killed his mom and raped her corpse, and Ottis had a thing for arson and cannibalism. They were into some really perverted stuff, like having sex with dead animals and that kinda thing.
Compare these descriptions with GG Allen, who was raised in the woods without running water or electricity, with a father so mentally ill, he reportedly named his son Jesus, because he had a vision of Jesus asking him to (the name got changed to GG). In school GG was bullied by other kids, and by high school would start dressing like a girl (shocking for the time period).
Like Lucas and Toole, GG would engage in the most sexually perverse of behavior such as self-reportedly raping people on stage (never confirmed), defecating on stage and throwing it at the audience, and shoving bananas into the most disgusting parts of his body and then eating them. In preparing for this post, I watched a documentary about him and it was so disturbing, I couldn’t even look at the screen and felt physically nauseous for hours afterwards.
In other words, GG, who says he would have been a serial killer, is very much of that ilk, both behaviorally and biographically, and the most depraved of serial killers, seem to average IQ 80. Note, I’m not talking about all serial killers, just the really perverted ones, since the same prenatal brain damage that causes paraphilias, also tends to lower IQ.
And yet GG did not become a killer, but rather an extremely successful man, who at his peak, claimed to have a million followers. Such fame and adulation, surely made him more powerful than many city mayors, though not quite as powerful as a U.S. congressman, governor, or national media personality.
Demographic prediction: Statistically expected IQ of a supremely powerful would-be serial killer: IQ 104
Let’s say GG was three standard deviations more powerful than the average white American on a normalized curve, which would make him about 4 SD more power than the average perverted would-be serial killer, since (would-be) serial killers are of low-middle socio-economic status.
If there were a perfect correlation between IQ and power, we’d expect GG’s IQ to be 60 points above the would-be depraved serial killer mean of 80,but since the correlation between IQ and power only seems to be about 0.4, his IQ would likely be only 40% so extreme. For example, U.S. presidents are only about 40% as far above the mean in IQ as they are in power.
So instead of GG’s IQ being 60 points above 80, he’d likely be about 24 points above 80, or IQ 104. This is consistent with a psychologist who interviewed him in prison and found his intelligence to be at least average (hat-tip to the omniscient ruhkukah).
Historiometric scholastic IQ
Because the demographic prediction made above has such a large standard error, it must be substantiated by historiometric evidence to be at all credible. Although historiometric IQs have been ridiculed, when competently done, they are perhaps one of the most important concepts in social science.
When I first read that GG had repeated the third grade, I didn’t give it much weight because judging by his adult persona, I assume he was a rebellious kid who constantly misbehaved. When interviewed, his mother confirmed that young GG simply didn’t care about school, but what she also said was that it was her decision to have him repeat the third grade because he was getting so far behind the other kids.
This implies that had GG taken a scholastic achievement test at age nine, he probably would have scored like an average white eight-year-old (since he was a year behind), thus implying a ratio IQ equivalent of 89 (8/9 = 0.89). This is considerably lower than the crude demographic prediction of 104 (white norms).
Historiometric Draw-a Man-IQ
The website ggallin.com claims to have original art work drawn by GG. This was an excellent opportunity for me to score GG on Florence Goodenough’s legendary Draw-A-Man IQ test, which was revised by Dale Harris in 1963. The test aimed not to measure artistic talent, but basic knowledge of the human body, which serves as a rough and ready measure of intelligence.
Original drawing by GG Allin
Applying the painstaking scoring procedure of the Goodenough-Harris Drawing Test to the above picture, I found that of the 73 items in the 1963 manual, GG failed 13 (items #6,7,8,12,13,16,21,25,37,49,60,61,69), for a raw score of 60/73.
The smoothed mean and standard deviation for American fifteen-year-olds (considered roughly adult level on tests like this) was 45.2 and 9.83 respectively, making GG’s score 1.51 SD above the mean (IQ 123, U.S. norms; about 122, white norms). However because the norms were from fifteen-year-olds circa 1963, they belonged to a birth cohort eight years older than GG’s.
According to scholar Richard Lynn, scores on the Draw-A-Man test become inflated at a rate of 3 points per decade (see The Flynn effect), which means GG would have scored 2.4 points lower if compared to his actual birth cohort, so we must reduce his IQ on this test to about 120 (white norms).
Historiometric Composite IQ: 105
Historiometric evidence suggests GG would have scored an IQ equivalent of 89 on a childhood scholastic achievement test, and when his drawing was scored on The Draw-A-Man test, he clocked in at IQ 120.
The Draw-A-Man test correlates between 0.48 to 0.72 with the Stanford Binet (Harris, 1963, pg 96). Assuming it also correlates about 0.6 with measures of scholastic achievement, which are essentially IQ tests in populations with similar schooling, then someone with a scholastic IQ of 88, and a Draw-A-Man IQ of 120, would have a composite IQ of 105 (white norms). This is consistent with both the demographic prediction and with the opinion of the psychologist who interviewed him, however given the speculative nature of historiometrics, combined with the mediocre validity of the Draw-A-Man IQ test, this should only be considered a crude estimate.
So I turned on Comedy Central’s Rob Lowe roast, expecting to see Lowe get roasted, but instead all of the jokes were aimed at Ann Coulter, who for some strange reason was attending.
I was watching with a woman who hated Ann Coulter, but by the end of the evening she felt sorry for her.
Coulter seemed to be making the same mistake her hero Trump made when he was roasted by President Obama: She didn’t laugh!
By laughing, you make yourself in on the joke, but by just sitting there smiling, staring nervously, you are merely the butt of the joke. At one point one of the comics mocking Ann, turned to her and said something like “don’t stare at me with that freaky bitch face,” causing the audience to explode into more cruel laughter at Ann.
When Peton Peyton Manning was mercilessly mocked for his freakishly huge head, Manning’s brain size gave him the intelligence to adapt, by slapping his knees laughing, but Ann just sat there all night with that creepy smile.
Knowing Ann’s obsession with Ivy League elitism, she was probably thinking the whole thing was beneath her. Who cares if these low SAT score state college losers mock me, she probably thought.
When it was finally Ann’s turn to take the stage, you expected her to turn the tables on her tormenters, but sadly her one liners seemed to bomb, and she was booed by the audience and heckled by the comics.
Because of Coulter’s anti-immigration views, many of the loudest laughs at Coulter’s expense came from Hispanics in the audience. This is yet another example of ethnic genetic interests.
Part of the problem was probably Coulter’s arrogance. The comedy writers had given her material, but she probably arrogantly thought “I don’t need low SAT scores state school losers to write jokes for me. I will write my own jokes.”
Although she did have one great line: ““Welcome to the Ann Coulter Roast with Rob Lowe”
But one of the advantages of being an Ivy League lawyer, is no matter how badly you screw up, people assume it must have been part of your brilliant master plan. Some are suggesting that she deliberately bombed as a way of getting more publicity.
As blogger Race realist noted on this blog, many Afrocentrists want to believe that the Australoid peoples of the Andaman Islands, Papua New Guinea, and of course Australia, among other places, are part of one large pan-African race.
By contrast so many HBDers want to deny this.
The root of this debate is not so much genuine scientific disagreement, but ethnic genetic interests. The black Afrocentrists believe it’s good for black people to be seen as part of this larger, global black community, thus dispelling the claim that blacks only left Africa in chains. For leaving Africa was one of the milestones that separates humans from our closest living relatives (chimpanzees), so the claim that blacks never left Africa is seen as racist indeed.
By contrast, the non-black HBDers think it elevates their racial status to promote the idea that their ancestors did something that blacks couldn’t accomplish: leaving Africa.
It thus becomes very important to deny Australoids are black or Negroird, because admitting that they are, means blacks were smart enough to figure out how to leave Africa some 70,000 to 130,000 years ago, and they simply don’t want to give blacks that much credit. On a subconscious level they fear, if even blacks figured out how to get out of Africa, then what makes my ancestors so special?
Both sides need to take a deep breath, and realize, any fool can leave Africa. Homo Erectus left Africa.
But perhaps leaving Africa 70,000 to 130,000 years ago was much more challenging than we can ever imagine. After all, the Sahara desert had become at times a dry wasteland, so perhaps the only way out was to build a raft that could carry them across the red sea from Ethiopia all the way to Yemen. This may sound trivial today, but given the technology and intelligence of the time, it must have felt like traveling to the moon. This is speculation of course; leaving Africa may have been easy if the Sahara had briefly become vegetated or if sea levels hit record lows.
But if the mere leaving of Africa did indeed screen for IQ, why do the Australoids score even lower on IQ tests than sub-Saharan Africans do, and have somewhat smaller brains?
In order to answer this question, we begin with the fact that only the congoid Africans score higher than the Australoids; the bushmen and pygmies score significantly lower, and the pygmies have by far, the smallest brains of all.
Perhaps what happened was, 70,000 years ago, the human species was about as smart as bushmen and pygmies, but those who left Africa were as smart as Australoids (and many became Australoids). Meanwhile, the population that stayed behind in Africa remained at Buhmen/Pygmy IQ level until the advent of agriculture and civilization in the middle east several thousand years ago. I believe once dark caucasoids invented agriculture, their population exploded giving rise to new mutations that raised their IQs by 13 points, allowing the creation of the first civilization: Mesopotamia.
With agriculture/civilization, the dark caucasoids traveled the World, spreading their high IQ mutation to every corner of the globe except those that are most isolated. So the mean genetic IQ of the entire World increased by 13 points, with the exception of places that were too hard to get to, such as Australia and Papua New Guineas, the Congo rain forest, the Southern tip of Africa, and the Americas. Since most of sub-Saharan Africa got the genes, they suddenly leaped from having lower IQs than the Austaloids, to being smarter (only the bushmen and pygmies, who like the Australoids, were too isolated to get the genes, remained behind the Australoids).
The downfall for the dark caucasoids was spreading these genes to whites and East Asians. Because of cold winters, East Asians and whites were smarter than pre-mutation dark caucasoids, and now with these mutations, they were smarter once again.
In honor of Labor Day, I wanted to write a quick post on Marxism. I’m not anti-Marxist; in fact I’ve endorsed ONLY Marxists for President of the United States on this blog (Bernie Sanders and Jill Stein), and given the U.S. Supreme Court’s ridiculous Citizen’s United ruling, the U.S. needs Marxists now more than ever. But even though Marxists probably tend to be quite smart given the correlation between IQ and liberalism and the difficulty of reading Marx, there are two ways in which Marxists seem clueless.
Marxists assume that enormous economic inequality is in and of itself, proof that the market is rigged. This ignores the fact that there is enormous inequality in human productivity. For example, a member of Prometheus brilliantly noted that because the human mind operates in parallel, complex learning and problem solving speed doubles every 5 or 10 IQ points. What that means is that is that in complex jobs, we should expect an IQ 170 to be up to 15,625 times more productive than an IQ 100. Further, if the IQ 170 is ten times more motivated than the IQ 100, he becomes perhaps 156,250 times more productive than the IQ 100.
The other factor that Marxists don’t seem to get is the role of technology in creating enormous inequality. In the distant past, a writer would take years to write only one book, which would severely diminish his productivity, but with the advent of the printing press, writers can produce MILLIONS of copies of their books. So you have this huge divide between those whose work can be multiplied a million fold by technology, and those whose work can only be done once per unit of effort. This divide seems most unfair when we compare dumb athletes making millions entertaining sports fans to brilliant doctors who make only six figures saving lives.
But what people don’t get is that a brilliant doctor, will save maybe five lives a year, while thanks to television, the dumb athlete is entertaining TENS OF MILLIONS of people a year: a trivial service multiplied by tens of millions is indeed worth more than a valuable service for only five people. So in a very objective sense, the dumb athlete deserves more money than the brilliant doctor.
When you combine the fact that complex problem solving speed doubles every 5 -10 IQ points, and then gets multiplied by differences in motivation and the use of technologies like the printing press, we should expect unbelievably large differences in wealth and income between the rich and the poor, even if everyone were playing fair (which they’re not). Yes the system is rigged, but mere inequality doesn’t prove anything; a truly fair system might result in even more inequality!
But at the same time, the athlete did not invent the television and the writer did not invent the printing press, nor does he enforce the arbitrary intellectual property laws that allow him to monopolize all the profits from reprints of his work. All success is the product of both the individual and the society in which he lives, which is why I don’t object to a 50% tax rate for all who can afford it. In theory I would even support a 50% tax rate on investment income, but that’s stupid because the government actually collects more tax dollars when they keep the capital gains tax low because more rich people then invest.
I also support a 50% inheritance tax. Some object to this because they’ve already been taxed 50% on their income, so everything they have left at death should be tax free. However I don’t see the inheritance tax as a tax on the dead, I see it as a tax on the person who inherits the money. If the tax rate on earning a million dollars in 2016 is for example 50%, why should the person who didn’t even earn his million in 2016, but was given it because his father died, be spared that 50% tax rate?
What should be done with all those tax dollars? Above all, I support Charles Murray’s idea of a negative tax for the relatively poor that would replace the welfare state and income transfer programs like social security, Medicare, Medicaid, Obamacare etc, (though a certain percent would have to be earmarked for health costs as Murray was reluctantly pursuaded).
I do NOT support a minimum wage. If a consenting adult is willing to work for less than a penny an hour, the government has no right to prevent it. A minimum wage unfairly places the burden of helping the poor on job creators, rather than distributing it equally among all tax payers, and with a negative tax for the relatively poor, it becomes redundant.
It also destroys jobs. Cashiers are being replaced by automated checkout machines and McDonalds has introduced automated ordering machines, though they say the new gourmet burger you can order on them create new jobs for chefs.
The IQ scale is often defined so that the mean, in First World whites is set at 100 and the standard deviation is set at 15. Using this scale, what is the mean and SD of the entire human species?
In order to answer this question, we need to estimate the highest and lowest IQ in the entire World (excluding organic retardates who skew the curve).
The highest IQ in the entire World is likely an East Asian given the high mean IQ and large population. Assuming there are 1.6 billion East Asians with a mean IQ of 105 and an SD of presumably 15, the normal curve predicts the highest IQ in East Asia, and by extension the World, would be 196.
The lowest (non-organic) IQ in the World is probably in sub-Sahara, given the low mean and large population. Assuming there are 1 billion sub-Saharans, with a mean IQ of 67 (per Richard Lynn) and an SD of presumably 15, then the lowest non-organic IQ in sub-Sahara, and by extension the World, would be -23.
So the highest IQ in the World is 196 and the lowest (non-organic) IQ in the World is -23, and in theory, the mean should be in between these extremes: IQ 87
Thus the difference between the highest and lowest (non-organic) IQ in the World is 219 points. The World population is 7.4 population. In a population this size, the difference between the highest and lowest data point is theoretically 12.66 SD.
Thus, dividing the IQ gap between the World’s highest and lowest IQ (219 points) by 12.66, gives the estimated standard deviation for the entire World, and that estimate is 17.3
So while an IQ of 87 is below average for First World whites, (mean IQ 100, SD 15), it’s totally average Worldwide (mean IQ 87, SD 17.3).
Many non-scientists have a great interest in heritability, but lack the science education and/or cognitive ability to understand modern techniques like Genome-wide Complex Trait Analysis (GCTA), so this post is a quick attempt to explain it. Full Disclosure: I have virtually no formal science training beyond high school but this is just an oversimplified explanation.
GCTA gives a measure of the squared correlation between additive genotype and phenotype. The reason it’s so confusing is that you can’t directly correlate a phenotype with a genotype if you haven’t found the genes that code for that phenotype, and thus you can’t determine if someone is genetically high on a given trait.
So for example, you can’t determine if someone’s genetic IQ matches their actual IQ, if you don’t know if they have the genes for IQ. Since a correlation, by definition, is how close the rank order of two variables (i.e. genetic IQ and actual IQ) agree, it can’t be directly calculated if one of said variables (i.e. genetic IQ) can’t be ranked. It would be like trying to calculate the correlation between height and weight, but all the weights were reported in a language you didn’t speak.
To sidestep this problem, GCTA was invented by a scientist of East Asian heritage. In GCTA, instead of ranking everyone in your sample from highest to lowest on each trait, you simply randomly assign people to pairs, and for each pair, calculate the genetic distance and the phenotype distance. So for example, if the people who differ by 100 single nucleotide polymorphisms (SNPs), on average, differ by one standard deviation in IQ, and if people who differ by one standard deviation in IQ differ, on average, by 39 SNPs, then perhaps it can be inferred that (in this sample) the correlation between genetic IQ and actual IQ is whatever number when squared and multiplied by 100, equals 39.
That number is 0.62
This is because in a bivariate normal distribution, the slope of the standardized regression line equals the correlation between two variables, so if a genetic difference of 100 SNPs regresses to a one standard deviation difference in IQ, then one standard deviation must be only 62% as extreme as 100 SNPs and if a one standard deviation difference in IQ regresses to a 39 SNP difference, then 39 must be only 62% as extreme as one standard deviation.
Once we have the correlation of say 0.62 between additive genotype and phenotype , we square it to get the amount of variation explained which in this example would be 0.38 (the real number is probably much higher, and even higher still for broad-sense heritability).
Of course what very few people realize is that heritability is technically NOT the percentage of the phenotypic variation explained by genes, it’s the percentage explained by genes when environment is held constant or allowed to vary randomly.
Recently commenter Trumpocalypse (aka Mugabe), also commented on GCTA:
if it could be shown that the joint distribution, f, of some measure of genetic distance between individuals, d, and phenotypic difference, p, has a linear conditional mean (that is, f(d|p) is a straight line), then it would be possible to avoid all the problems of twin studies and the shared womb of twins…
the conditional mean could simply be extrapolated to d = 0. and from the mean difference, p, at d = 0, the heritability could be calculated easily.
furthermore the perfect-ness of the measure of genetic distance would be irrelevant. the joint probability density function would “include” this error. so with a large enough sample the extrapolation would have very little error.
this is essentially what GCTA is.
but whether the joint distribution is bivariate normal or some other distribution with linear conditional mean…i don’t know.
Here is some genetic distance data on nine major human populations:
According to Richard Lynn’s controversial meta-analysis of these nine genetic clusters, Africans average IQ 67, Non-European Caucasoids average 84, European Caucasoids average 99, Northeast Asians average 105, Arctic Northeast Asians average 91, Amerindians average 86, Southeast Asians average 87, Pacific Islanders average 85, and New Guineans and Australians average 62.
Although this is group data, not individual data, it would be interesting to compare group genetic distance to group IQ difference. If for example, I knew both the average IQ difference between two random humans from anywhere in the World, and if I also knew the Fst distance between two random individuals from anywhere in the World, I think I could probably use my crude understanding of GCTA, to estimate the IQ phenotype-genotype individual level correlation of the entire human species from Sforza’s and Lynn’s group level data. Squaring this correlation might be a good proxy for IQ’s Worldwide heritability.
On the other hand, the fact that Sforza intentionally tried to use non-selected genes to calculate genetic distance (since these mutate at a regular rate creating a reliable molecular clock for splitting off dates) might make the exercise pointless.
I’m reading James Flynn’s new book Does Your Family Make You Smarter? and I’m absolutely blown away by how subtle and creative his ideas are. Unlike many environmentalists, who simply nitpick the research of IQ hereditists, Flynn actually makes his points with great statistical creativity.
In order to understand this book, you must first understand that according to the conventional wisdom, our IQs have three main causes:
Genes: The effect of genes on IQ can be measured by correlating the IQs of identical twins raised apart (though the correlation should be reduced a bit for shared prenatal environment)
2. Family environment: The effect of family environment can be measured by correlating the IQs of unrelated people raised in the same home.
3. Chance environment: These are random environmental events that affect our individual IQs, but don’t effect our siblings raised in the same family. For biological determinists like Arthur Jensen and I, these might be getting hit on the head with a golf ball or not getting enough oxygen in the womb. For cultural determinists like Flynn, these might be having a school teacher who inspires you to learn. The effect of chance environment on IQ can be measured by noting that there’s a higher IQ correlation between the same person tested twice, then there is between two identical twins raised in the same home. In other words, you are more similar to you, then you are even to your identical twin raised in the exact same family, so there are obviously unique environmental effects that make even identical twins raised together, individuals.
Now according to Jensen’s (1998) summary of kinship and adoption studies, genes explain about 45% of the IQ variation in childhood, 65% of the variation in adolescence, and about 80% in later maturity. Family environment explains about 35% of the IQ variation in childhood, and near zero by late adolescence. Meanwhile chance environment explains about 25% at all ages.
Why does family environment go from 35% to zero percent? The conventional wisdom is that people with high IQ genes do well in school and end up in stimulating adult environments, and these adult environments replace the effects of family environment. Further, because their genes caused these adult environments, the effect shows up as genetic variance.
Flynn’s brilliant insight
I’m going to do my best to explain Flynn’s ground-breaking new method. Although so far I’ve only read the first few chapters, I hope I’ve at least partly grasped the concept (it’s pretty subtle):
High IQ adults outperform high IQ children on tests, both because of age differences, but also because by adulthood, high IQ genes create high IQ environments, creating a multiplier effect.
By contrast average IQ adults score higher than average IQ children, ONLY because of age differences. In a normal curve, by definition the average person is, on average, average in genes and environment.
Thus by subtracting the test performance gap between average IQ adults and average IQ kids, from the performance gap between high IQ adults and high IQ kids, you get a pure measure of the effect of non-chance environment on test performance.
What this brilliant method shows is that on tests that I have always known are quite culture reduced, like using colored blocks to form designs, the effect of non-chance environment vanishes by the early teens, while on tests that are obviously cultural (vocabulary), non-chance environment persists into adulthood.
To me, the beauty of this method is that two tests could have the exact same heritability, but one test could be heritable for indirect reasons (genes causing environments that improve test performance) while the other test is heritable because genes directly affect performance.
How ironic that Flynn, the man best known for debunking the notion of culture reduced tests via the Flynn effect, may have unintentionally proven they exist after all.
