The Unz Review - Mobile
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
Email This Page to Someone

 Remember My Information

 Russian Reaction Blog / Paper ReviewTeasers


I don’t know how, but Lynn, Cheng, and Russian psychometricist Grigoriev have managed to find Russian regional results for PISA 2015.


Moscow has plummeted in the rankings and is now fourth, whereas Saint-Petersburg is now first.

I have calculated the correlations with the PISA 2009 results, for regions that participated in both surveys, to be a pretty weak r=0.52. As you can see, the samples for each region are pretty small, typically around 100, though relatively more schoolchildren were tested in the capitals: 245 in Saint-Petersburg, and 373 in Moscow.

The Yakut-majority Sakha Republic has improved drastically, by half an S.D., so it is no longer last, but modestly below average (this ties in with Vladimir Shibaev’s recent work in 2017 which shows that Yakut IQ might be similar to Russian, and not drastically lower, as an earlier study from 2015 had indicated). That “honor” now belongs to Dagestan, which remains stuck at a PISA-equivalent IQ in the high 80s.


Lynn et al. also did their standard correlation exercises.

Other tests of academic achievement (average Unified State Exam results of those admitted to universities from 2014) and historical literacy (1897 census):

Note in particular that the province of Dagestan has the lowest PISA score (424.1) and the second lowest EQ (84); and also that the city of St. Petersburg has the highest PISA score (524.4), the highest EQ (111) and the highest literacy rate in 1897 (61.6%). The city of Moscow has the fourth highest PISA score (516.4), the second highest EQ (110) and the second highest literacy rate in 1897 (53.1%).

GDP per capita:

Second, the PISA scores were correlated at r = .31 with GDP per capita. The correlation falls just short of statistical significance at p<.05 (r = .32 would be statistically significant).

wealth-iq-russia This is because some Russian regions have resource windfalls amidst low populations, e.g. Khanty-Mansyisk AO, which accounts for half of Russia’s oil output and enjoys a Swiss-like standard of living.

If you only consider “normal” Russian regions, the correlation becomes a much more typical r=0.73 (the graph to the right is based on results from PISA 2009 and PPP-adjusted Gross Regional Products from 2008.

Russian ethnicity:

Third, the PISA scores were significantly correlated at r = .45 (p<.01) with the percentage of the population with Russian ethnicity. This result is confirmed by the multiple regression analysis showing that the percentage of Russian ethnicity was a significant predictor of the PISA scores (β = .36, t = 2.68, p<.01).

Cold winters:

Fourth, the PISA scores were significantly correlated at r = .35 (p<.05) with latitude showing that IQs are higher in the more northerly provinces.

• Category: Race/Ethnicity • Tags: IQ, Paper Review, Psychometrics, Russia 


Grigoriev, Andrey & Lynn 2009
Studies of Socioeconomic and Ethnic Differences in Intelligence in the Former Soviet Union in the Early Twentieth Century


This paper reviews the studies of socioeconomic and ethnic and racial differences in intelligence carried out in Russia/USSR during the late 1920s and early 1930s. In these studies the IQs of social classes and of ethnic minorities were tested. These included Tatars (a Caucasoid people), Chuvash and Altai (mixed Caucasoid-Mongoloid peoples), Evenk (a mixed Caucasoid-Arctic people), and Uzbeks (a Central-South Asian people). The results of these studies showed socioeconomic differences of 12 IQ points between the children of white collar and blue collar workers, and that with the exception of the Tartars the ethnic minorities obtained lower IQs than European Russians.

This is essentially a short history of psychometrics in the USSR/Russia.

(1) The first measurement of Russian IQ was performed in 1909 by A.M. Schubert, who used the French Binet test with n=229 children: “She concluded that the Binet test appeared to be too difficult for Russian children and the scale should be moved on 1 to 2 ages to be appropriate for them.” Since Mental age ÷ Physical age × 100 = IQ, this implies their average IQ was perhaps one S.D. lower than that of the French, though later researchers pointed out those children were drawn from lower socio-economic strata.

In 1930, now in the USSR, another study found the following:

They tested 414 children aged between 8½ and 11½ with the American Stanford– Binet (administered in Russian translation). The sample consisted of 200 children of peasants,141 children of blue collar workers, and 73 children of white-collar workers. All children were from Moscow or the Moscow region. The results were that the children of peasants obtained a mean IQ of 87 (the standard deviation=10), the children of blue-collar workers a mean IQ of 91 (SD=8.6) and the children of white-collar workers a mean IQ 98 (SD=8.4). The mean IQ (unweighted) for three groups was 92… Thus, the total weighted mean for Russian children in this study was 90.3 (these IQs are in relation to American Stanford–Binet norms).

Capture This brings to mind a 1920s study quoted by Anne Anastasi in her book Differential Psychology (pp.524), in which Russian immigrant children to the US got 90.

This 10 point difference was presumably there because Russia was a more economically backwards country, with a more repressed average IQ due to gaps in schooling, malnutrition, parasitic load, etc.

(2) As in the West, consistent differences were found in the IQs of people from different socio-economic strata.

Another study of relation of IQ to social class was carried out by M. Syrkin (М.Сыркин)(Сыркин,1929) who compared the intelligence of fourth grade children (N=338, age approximately 10 years) belonging to six socio-economic groups. The lowest group was described as “ blue collar workers and at least one of parents illiterate”and the highest group was described as “white-collar workers and at least one parent educated in an institute of higher education”. Intelligence was assessed with five verbal tests measuring comprehension and verbal reasoning. There was a difference of 1.42d(equivalent to 21.3 IQ points) between the lowest and highest socioeconomic groups.

The USSR really did expel, kill off, or otherwise limit the reproductive fitness of its best and brightest.

In 1928, E.I. Zverev (Е.И. Зверев)(Зверев, 1931) tested the IQ of 114 children just admitted to school and aged about 7½– 8 years, in and around the city of Kursk, about 500 km south of Moscow. The children were tested with the Binet– Bert test (a Russian adaptation of the Binet). The mean IQ of these children was 80.8. This is much lower than the IQ of children obtained by Gurjanov, Smirnov, Sokolov, & Shevarev (Гурьянов, Смирнов, Соколов,&Шеварев, 1930) for Moscow and the Moscow region. Probably this difference was due to methodological and sample differences, but there is a possibility that the regional factor was also involved.

The latter hypothesis is likely the correct one.

In the 2009 PISA test, there was a 12 IQ point difference between Kursk and Moscow, which is an incredibly concentrated cognitive cluster.

(3) Now we go on to the most “controversial” part – ethnic differences in IQ.

Central Russia

There were also some studies of the IQs of non-Slavonic but predominantly Caucasoid peoples.I. Bektchentay (И .Бикчентай) and Z. Carimowa (З.Каримова )(Бикчентай &Каримова, 1930) tested the IQs of 380 Tartar children aged 8– 18 in fi ve Tartar schools in Moscow with the Boltunow–Binettest(aRussian adaptation of the Binet). The Tartars are indigenous to the Caucasus in the far south of Russia and the former Soviet Union, but a number of them live in central Russian towns and cities. The mean IQ of the Tartar children in this study was approximately the same as that of Russian children. The correlation between the Boltunow– Binet test and school achievements (assessed by teachers’ estimates) in their study was 0.84.

Yes, this is a pretty major distinction.

The Volga Tatars – the Muslim and Christianized Tatars of central Russia – have an average IQ of around 100 (about equal to modern Russia/Europe). Population genetics studies have found them to be basically acculturated Slavs.

The first of these was reported by F.P. Petrov (Е.П. Петров) (Петров, 1928) who tested the IQs of 1398 Chuvash children aged 3–13 in 1926–1927 with the French Binet–Simon test… The figures inTable 2 show a median IQ of 87 for boys and 84 for girls, and means (unweighted) of 89 for boys and 86 for girls. These are in relation to 100 for French norms, but no normative data are reported for Russian children. The IQs of the Chuvash children show a decline with age, with the lowest IQs among the 12 and 13 year olds.

Chuvashia is currently about average for the Russian regions.


Also tests carried out on indigenous tundric peoples, such as the Evenks (Bulanov 1930):

The results are presented as typical for Evenk children, but because of the small samples, their IQs may not be regarded as reliable. The results are as follows. For the Binet test the mean IQ was 70.16 (for 5 children, and in relation to French norms). The results obtained with the Rossolimo test showed lower average IQs of the Evenk (Tungus) compared with a Moscow sample on some abilities, namely, memory for pictures and words, ability to comprehend combined pictures, ability to comprehend visual incongruities, and, according to Bulanow’s interpreta- tion, ability to retain a high level of attention. As regards memory for pictures, the results contradicted the sometimes described capacity of Evenk (Tungus) to remember exactly long routes on wild territory (Encyclopedic Dictionary by Brockhaus & Efron (Энциклопедический словарь Ф .А . Брок – гауза и И.А .Ефрона ), 1902, vol. 67, p. 66)….

Bulanow also reported some observations on Evenk (Tungus) children and adults concerning their great difficulty in understanding the concepts of measurement and number. He reported that when Evenk children were questioned about devices for measurement, they did not have the concept of an absolute unit of measurement. They thought that the unit changed with the material measured. Bulanow reported further that when he asked Evenk adults how many children they had “ It was difficult, almost impossible, to get from parents precise information as to how many of their children were alive, how many of their children had died, what was the age of their children, and so on.” (p. 198).

… and on the Altai (Zaporochets 1930):

The results for the Binet test were as follows: mean IQ for total group was 66.9 (sd. 8.5), mean IQ for children aged 8– 12 was 69.15, and the mean IQ for children aged 13–16 years was 64.8. As noted by Zaporojets, this test was tedious for the Altai children. Some tasks were especially difficult for them. These were tasks involving calculation, logical operations, and the fluency task to name as many as words as possible during 3 min. As for the Rossolimo test, the most diffi cult tests for Altai children were those requiring the ability to retain a high level of attention and to comprehend visual incongruities. Their mean IQ for the Pintner–Peterson test was 75.

Zaporojets noted that the Altai children did not have a clear understanding of units of measurement. He observed that when they were questioned about the length of a meter, the Altai would often ask: “Which meter?”They thought that the meter in one shop could be longer than in another. An adult Altai said about distance: “It is 100 big versts (approximately 100 kilometers)” (he apparently thought that the number of small versts must be more).

Zaporojets’ paper contains some interesting observations on adult Altai. Although adult Altai performed calculations poorly at the time of study, they showed a remarkable ability for visual estimation of large quantities. A herdsman, who could count only to 20–30, noticed very well the absence of one horse, cow or sheep in a herd of many hundreds. He looked at a huge herd and noted that a particular cow was absent. Another example of the great visualization ability of the Altai was that they could remember and showed the way through wild territory, where they had been only once many years previously.

Common theme: No numeracy (they’d have a very bad Whipple’s index), very premodern and non-abstract ways of thinking, but quite well suited for their environment.

In PISA 2009, Yakutia had the lowest score of any tested Russian region, including Dagestan (though Chechnya and Ingushetia were not included). Ethnic Yakuts, who probably have similar IQs to the Altai and Evenks, constitute 50% of its population, though probably more like 2/3 amongst the children taking PISA due to their higher fertility rates. This might imply that the average Yakut IQ is in the low-to-mid 80s.

Central Asia

First test was carried out in 1926 by A. Schtelerman: He did not give IQs but reported that the scores of the Uzbek children were lower than those of children in Moscow.

A series of studies by V.K. Soloviev on Russian and Uzbek army cadets and professionals found that “the test scores and the educational level of the Uzbeks were lower than those of the Europeans.”

The third study of the intelligence of the Uzbeks was carried out in 1931 by A.R. Luria (А.Р . Лурия ), at that time at the Institute of Psychology in Moscow. Luria did not use intelligence tests but gave a descriptive analysis of the Uzbeks’ cognitive abilities. He distinguished two modes of thought designated graphic recall (memories of how objects in the individual’s personal experience are related) and ca- tegorical relationships (categorisation by abstract concepts). He found that the thought processes of illiterate Uzbek peasants were confined to graphic recall and that they were not able to form abstract concepts. For example, they were shown a hammer, an axe, a log and a saw, and asked which of these did not belong. The typical Uzbek answer was that they all belonged together because they are all needed to make firewood. People who are able to think in terms of categorical relationships identify the log as the answer because the other three are tools (an abstract concept). Illiterate Uzbeks peasants were unable to form concepts of this kind. They were also unable to solve syllogisms. For instance, given the syllogism “There are no camels in Germany; the city of B is in Germany; are there camels there?” Luria gave as a typical Uzbeks answer “I don’t know, I have never seen German cities. If B is a large city, there should be camels there.” Similarly, Luria asked “In the far north, where there is snow, all bears are white; Novia Zemlya is in the far north; what color are the bears in Novia Zemlya?”. A typical Uzbek answer was “I’ve never been to the far north and never seen bears”(Luria,1979, p. 77–8). Thus, Luria concluded that these peoples were not capable of abstract thought: “ the processes of abstraction and generalization are not invariant at all stages of socioeconomic and cultural development. Rather, such processes are pro- ducts of the cultural environment” (Luria, 1979, p. 74). Luria proposed that the ability to think in terms of categorical relationships is acquired through education. He did not suggest that the Uzbeks have any genetic cognitive deficiency.

I wrote about Luria back in the late 2000s when I still agnostic about genetic racial differences in IQ.

Today those factors no longer really hold, but Central Asians do very poorly on international standardized tests.

Kyrgyzstan came at the very bottom of PISA 2009, with a PISA-equivalent IQ of around 75.

Table below is from David Becker’s database of national IQs:

National Ethnic Age N Test IQ Study
Kazakhstan 8 to 16 617 SPM+ 87.30 Grigoriev & Lynn (2014)
Kyrgyz 85.60 Lynn & Cheng (2014)
Tajikistan 13 to 15 674 SPM+ 88.00 Khosimov & Lynn (2017)
Uzbekistan 10 to 15 51 SPM+ 86.00 Grigoriev & Lynn (2014 )
Uzbekistan 11 to 13 614 SPM+ 85.00 Salahodjaev et al. (2017)

Still, Luria has some of the best arguments against that position, so its a bit surprising that the blank slatists don’t cite him more.

stalin-the-tajik(4) Or maybe not, because it still didn’t save him him from the SJWs’ ideological predecessors, Sovok Justice Warriors:

These early studies carried out in the years 1926– 1931 found that there were substantial socioeconomic and ethnic/ racial differences in intelligence in the Soviet Union. These conclusions were not consistent with Marxist orthodoxy which held that these differences would disappear under communism. Accordingly, these studies, particularly that of Luria, attracted a great deal of criticism in the Soviet Union in the early 1930s. This has been described by Kozulin (1984): “Critics accused Luria of insulting the national minorities of Soviet Asia whom he had ostensibly depicted as an inferior race. The results of the expedition were refused publication and the very theme of cultural development was forbidden” . In 1936 intelligence testing was banned in the Soviet Union. It was not until the 1960s and early 1970s that this prohibition was progressively relaxed (Grigorenko & Kornilova, 1997). Luria’s work was not published in Russian until 1974 and English translations were published in 1976 and 1979 (Luria, 1976, 1979).

As Lynn and Grigoriev point out, this was closely correlated to the suppression of genetics research, though at least Luria and Co. weren’t outright murdered like Vavilov.

The history of work on intelligence in the former Soviet Union parallels that of genetics, where mainstream Mendelian theory represented by Nikolai Vavilov in the 1920s was likewise suppressed in the 1930s and replaced by the environmentalist pseudo-genetics of Trofi m Lysenko. The domination of science by political theory was relaxed in the 1960s and 1970s, and in recent decades both intelligence research and Mendelian genetics have been rehabilitated in Russia.

Scientifically, there is real work being done on psychometrics in Russia, though in comparison to the US it is very meager and basically inconsequential.

Since it is not politicized in the US, it is neither promoted nor prosecuted.

If psychometric considerations were to move closer to politics, e.g. by tying them to the hot potato that is Central Asian immigration, things can go any which way. Although Russians have a more commonsense take on these matters – if 25% of Americans seriously think intelligence is a “social construct,” it’s probably more like 5% in Russia. On the other hand, the Leftists, Stalinists, and even many Eurasianists are aggressively opposed to the idea that intelligence is heritable and differs significantly between races, and in the event that the authorities side with them, Russian scientists don’t have the First Amendment or an fair and impartial court system to hide behind.






Here we report the development of a system that incorporates a pumpless oxygenator circuit connected to the fetus of a lamb via an umbilical cord interface that is maintained within a closed ‘amniotic fluid’ circuit that closely reproduces the environment of the womb. We show that fetal lambs that are developmentally equivalent to the extreme premature human infant can be physiologically supported in this extra-uterine device for up to 4 weeks. Lambs on support maintain stable haemodynamics, have normal blood gas and oxygenation parameters and maintain patency of the fetal circulation. With appropriate nutritional support, lambs on the system demonstrate normal somatic growth, lung maturation and brain growth and myelination.

This is really cool.

twitter-artificial-wombs I have been advocating this technology since I started blogging in 2008.

The immediate benefits, which the authors cite, are a reduction in infant mortality caused by extreme prematurity. This is good, though not that big of a deal, since it is very low in First World countries anyway, while poorer countries will probably not be able to afford the technology anyway.

The real promise is in its eugenic potential.

It is common knowledge that the well-educated reproduce less than the poorly educated, and that has resulted in decades of dysgenic decline throughout the developed world. This dysgenic effect has overtaken the Flynn effect. One of the reasons the well-educated, and especially well-educated women, have few or zero children is because it is bad for their career prospects. There are also some women who are just uncomfortable with the idea of pregnancy and childbirth.

There are essentially just a few solutions to this problem:

(1) Do nothing, deny heritability of IQ. Import Afro-Muslims to breed the next generation of doctors and engineers.

(2) Do nothing, hope for a literal deus ex machina solution, such as Musk’s neural lace or superintelligence.

(3) The Alt Right solution: Send the women back to the kitchen.

Ethical considerations aside, there’s also the matter of practicality – you’d have to be really hardcore at enforcing your “White Sharia” to make any substantive difference. Even most conservative Muslim societies, where female labor participation is very low, have seen plummeting fertility rates. And, needless to say, it does nothing about the dysgenic aspect of modern fertility patterns, which are a significantly bigger problem than falling fertility rates anyway.

(4) Develop artificial wombs.

This is a good idea from all sorts of ideological perspectives.

Everyone: Immediate higher fertility rates in the countries that develop them, especially amongst well-educated women. This might cancel out dysgenic decline at a single stroke.

Liberals: Alternate option for women who don’t want to undergo pregnancy/childbirth for whatever reason. No more market for surrogate mothers – an end to a particularly icky form of Third World exploitation.

Libertarians: People with the means to pay – that is, millionaires and especially billionaires – will no longer be bounded in their reproductive capacity by the biology of their female partner or by the culture of their society (generally, no polygamy). Since wealth is moderately correlated with IQ, this will be eugenic. That said, this might strike some as dystopian. Maybe one could start taxing additional artificial womb-grown offspring past the first five or ten? Then you’d get “offshore hatcheries.” Okay, I suppose that’s even more dystopian.

Zensunnis: I suppose cultures that really dislike women can just gradually start making do without them by replacing them with the equivalent of Axlotl tanks. Conversely, (almost) all female “Amazonian” societies will also become possible. Let’s make sci-fi tropes real.

Futurists: Combining artificial wombs with CRISPR gene-editing for IQ on a mass scale pretty much directly leads to a biosingularity.

As I pointed out, a biosingularity may be preferable to one born of machine superintelligence because it bypasses the AI alignment problem and doesn’t risk the end of conscious experience.

• Category: Science • Tags: Fertility, Paper Review, Transhumanism 


Tang, Lichun et al. 2017
CRISPR/Cas9-mediated gene editing in human zygotes using Cas9 protein


Previous works using human tripronuclear zygotes suggested that the clustered regularly interspaced short palindromic repeat (CRISPR)/Cas9 system could be a tool in correcting disease-causing mutations. However, whether this system was applicable in normal human (dual pronuclear, 2PN) zygotes was unclear. Here we demonstrate that CRISPR/Cas9 is also effective as a gene-editing tool in human 2PN zygotes. By injection of Cas9 protein complexed with the appropriate sgRNAs and homology donors into one-cell human embryos, we demonstrated efficient homologous recombination-mediated correction of point mutations in HBB and G6PD. However, our results also reveal limitations of this correction procedure and highlight the need for further research.

Gwern Branwen’s comments:

Even nicer: another human-embryo CRISPR paper. Some old 2015 work – results: no off-target mutations and efficiencies of 20/50/100% for various edits. (As I predicted, the older papers, Liang et al 2015 / Kang et al 2016 / Komor et al 2016, were not state of the art and would be improved on considerably.)

Back in February 2015, qualia researcher Mike Johnson predicted that dedicated billionaire with scant regard for legalistic regulations could start genetically “spellchecking” their offspring within 5-7 years.

But if anything, he might have overestimated the timeframe.


• Category: Science • Tags: Crispr, Genetic Load, Paper Review, Transhumanism 


Ashburn-Nardo, Leslie 2017
Parenthood as a Moral Imperative? Moral Outrage and the Stigmatization of Voluntarily Childfree Women and Men


Nationally representative data indicate that adults in the United States are increasingly delaying the decision to have children or are forgoing parenthood entirely. Although some empirical research has examined the social consequences of adults’ decision to be childfree, few studies have identified explanatory mechanisms for the stigma this population experiences. Based on the logic of backlash theory and research on retributive justice, the present research examined moral outrage as a mechanism through which voluntarily childfree targets are perceived less favorably than are targets with children for violating the prescribed social role of parenthood. In a between-subjects experiment, 197 undergraduates (147 women, 49 men, 1 participant with missing gender data) from a large U.S. Midwestern urban university were randomly assigned to evaluate a male or female married target who had chosen to have zero or two children. Participants completed measures of the target’s perceived psychological fulfillment and their affective reactions to the target. Consistent with earlier studies, voluntarily childfree targets were perceived as significantly less psychologically fulfilled than targets with two children. Extending past research, voluntarily childfree targets elicited significantly greater moral outrage than did targets with two children. My findings were not qualified by targets’ gender. Moral outrage mediated the effect of target parenthood status on perceived fulfillment. Collectively, these findings offer the first known empirical evidence of perceptions of parenthood as a moral imperative.

The author herself doesn’t seem to be happy with her own findings:

Practically speaking, the present findings have some troubling potential implications for howpeople transition to parenthood. For example, the present findings, obtained with college students in the Midwestern United States, suggest that many young people view children as a necessary ingredient for fulfilling lives. Thus, they may feel tremendous pressure to have children, not only from others as this literature suggests (Mueller and Yoder 1999), but also internally. Ironically, these perceptions have absolutely no basis in reality. Meta-analyses reveal that parents report significantly less marital satisfaction than do non-parents, and as their number of children increases, marital satisfaction decreases (Twenge et al. 2003).

That maybe so, but reality definitely seems to have a basis in those perceptions.

For instance, people without those perceptions didn’t tend to pass on their genes.

• Category: Science • Tags: Demographics, Paper Review, Psychology 


Ritchie, Stuart – 2017 – Review of The Rationality Quotient by Stanovich et al.


From Stuart Ritchie’s review of “The Rationality Quotient” by Keith Stanovich et al.:

But it was the reported correlation of the [Comprehensive Assessment of Rational Thinking] with IQ-type tests that was really unexpected, given the authors’ argument that they measure very different constructs. A cognitive composite—made up of tests of analogies, antonyms, and a word checklist (Table 13.11)—was found to have a correlation with the full-scale CART of 0.695. 0.695!

That’s the extent to which actual IQ tests typically load on the g factor and each other. One might even go so far as to propose that rationality is intelligence.

The notion that intelligent people are more prone to irrationality is a cognitive bias, though a very understandable one. The Newton who obsesses over the occult is just considerably more noticeable than some nutter ranting about the End Times.

Greg Cochran counters that Western intellectuals were more likely to fall for “destructive nonsense” than plumbers during the 20th century. I suspect that was more due to intellectuals not understanding plumbers, neither then nor now, rather than any failure of rationality per se. In everyday life, people tend to associate with people of similar intelligence, and have a social circle of about 150 friends and acquaintances.

And guess what? Communism works great within monasteries and universities.

• Category: Race/Ethnicity • Tags: Paper Review, Psychometrics, Rationality 


Whitley, Elise et al. – 2016 – Variations in cognitive abilities across the life course



New paper by Elise Whitley et al. on age and sex differences in IQ for n=~40,000 British sample.

  • Five tests: Word recall, verbal fluency, and subtraction (loading ~0.5 on g), and number sequence and numerical problem solving (loading ~0.7 on g).
  • Males score about 4 IQ points more on the derived g-factor of cognitive ability.
  • … though this result should be treated with caution on account of: (a) g having different structure across the sexes; (b) it is not an exception to a common problem in IQ and sex studies, namely, the undersampling of men with lower cognitive ability.
  • Better subjective health was associated with higher IQ.
  • The overall pattern across age was a plateau from the late teens to age 65, then a steep fall soon thereafter.

I would say that the ultimate and really the only reason we have mandatory retirement policies are cognitive ones.

EDIT: Emil Kirkegaard had a closer look at the results, including a nicer graph of the age/sex results:

My guess is that the intercept bias/invariance has to do with the composition of the battery. There were only 5 tests, and their breakdown was: 3 math, 1 verbal, 1 memory. Women had better memory but there was no difference in verbal fluency (this is a common finding despite what you have been told). So, the problem likely is that the g factor is colored because 60% of the tests were about math, and that men have an advantage on the math group factor.




Kong, Augustine et al. – 2016 – Selection against variants in the genome associated with educational attainment


This paper makes the case that there has been a decline in the prevalence of genes increasing propensity for more education (POLY EDU) in Iceland from 1910-1975.


Here are some of the key points:

  • The main mechanism was greater age at first child, not total number of children (i.e. the clever are breeding more slowly).
  • As in many such studies, the effect is stronger for women.
  • One allele associated with more children and having them earlier also tags a haplotype associated with “reduced intercranial volume” and neuroticism: “… thus a striking case where a variant associated with a phenotype typically regarded as unfavorable could nonetheless be also associated with increased “ fitness” in the evolutionary sense.
  • The decrease in POLY EDU prevalence was faster earlier this century, but this is an artifact of the higher survival schedules of people with a higher propensity for education (i.e. tying in with the well known finding that higher IQ is associated with higher life expectancy). The decline from 1940 onwards becomes linear, and is a better measure of estimating the change of the average polygenic score over time.
  • It is estimated that is POLY EDU declining by 0.010 SUs per decade, but this rises to 0.028 SUs per decade because the measure captures only a fraction of the full genetic component of education attainment (POLY FULL).
  • The trends in POLY FULL are estimated to be causing a decline of 0.30 IQ points per decade.
  • The authors note that this has entirely canceled out and then some by the Flynn effect, but it could still have “a very substantial effect if the trend persists for centuries.”

Many other studies indicate that the FLynn effect has ended or gone into reverse across the developed world around the 2000s by the latest.

If it’s a permanent plateau, we could be seeing 3 IQ point declines per century. Extend that out for two or three centuries, add some more Third World immigration, and you get the 1 S.D. IQ decline that I posited for the Age of Malthusian Industrialism aka the business as usual scenario.

Anatoly Karlin
About Anatoly Karlin

I am a blogger, thinker, and businessman in the SF Bay Area. I’m originally from Russia, spent many years in Britain, and studied at U.C. Berkeley.

One of my tenets is that ideologies tend to suck. As such, I hesitate about attaching labels to myself. That said, if it’s really necessary, I suppose “liberal-conservative neoreactionary” would be close enough.

Though I consider myself part of the Orthodox Church, my philosophy and spiritual views are more influenced by digital physics, Gnosticism, and Russian cosmism than anything specifically Judeo-Christian.