Compare the study in question[1] with a list of the nations ranked by IQ[2]. The lists are in about the same order, because math accomplishment is driven by IQ, IQ is strongly influenced by genetics, and different countries tend to have different genetics.
much further separated than kids of different ethnicities in the same country
No. African-Americans, for example, have an average IQ of 85, very much like an up-market African country. Go read The Bell Curve.
only extremely bad families have a major environmental effect
The siblings in the adoptive families have nearly the full range of accomplishments of the general public. The possible explanations are (1) a sibling's accomplishments are mostly determined by internal causes, or (2) a sibling's accomplishments are substantially influenced by environment, and they experience a wide range of environments (even within the same family).
Other studies use twins raised together. The twins are more similar to each other than to their non-twin siblings, and the differences are explained mostly by IQ. This suggests that accomplishment comes mostly from internal causes.
the environments they were looking at were actually very similar
In terms of opportunity, they are extremely similar by adulthood. In the U.S., annual income at age 30 depends mostly on IQ, regardless of race, ethnicity, and other factors.
In fact, although IQ explains more academic achievement and income than any other measurable psychometric factor, it still doesn’t explain very much of them. It explains only about 15-25% of variation in individual income and only about 25% of variation in individual academic achievement. Most of the rest is due to factors that are not measured.
By contrast, per capita income explains 46% of variation in average IQ among countries, and even more of the variation in academic performance. (Average academic performance explains over 50% of the variation in average IQ among countries.)
So, does high average income cause people to have high IQs and good academic performance, or the other way around? http://sitemaker.umich.edu/salas.356/more_money__better_grad... documents a study that found children’s academic performance improves when their parents’ income increases.
The TIMSS data you’re citing show the opposite of what you claim them to. The variations in academic achievement between countries are much larger than variations in IQ between ethnicities.
TIMSS doesn’t include a lot of African countries, but of those it does include, 8th-grade achievement in Ghana is 309, in Botswana is 364, in Egypt is 391, and in Tunisia is 420, compared to 508 in the US. The TIMSS scale is normalized so that one standard deviation is 100 points. If we estimated that the 508 score were a weighted average of 85% “white people” and 15% “black people” one standard deviation lower, then the “white people” would be at 523 and the “black people” would be at 423 --- closer to the “white US people” only one s.d. higher than to the Ghanaians 1.14 s.d.s lower. Even Botswana (perhaps the “upmarket” African country you’re thinking of?) would be 0.59 s.d.s below the “black US people”.
(The above breakdown is roughly correct. In the NAEP, the grade 8 mathematics average score for white students in 2009 was 293, the s.d. was 33, and the average for black students was 261, almost exactly one s.d. lower. http://nces.ed.gov/nationsreportcard/naepdata/ doesn’t let me link the results, though.)
In general, the TIMSS rankings are much better explained by median income than by average intelligence. The Taiwan average is almost two s.d.s above the international average, which would be like an average IQ of 130, while the Qatar average is almost two s.d.s below, which would be like an average IQ of just over 70, “borderline mentally retarded”. If you’ve ever traveled, it’s obvious that Qatar isn’t actually half populated with mentally retarded people and Taiwan isn’t actually half populated with geniuses.
According to the book you linked, the measured average IQs are 105 for Taiwan and 78 for Qatar.
If you look at this map, you can see how little relationship there is between ethnic composition and national IQ:
http://en.wikipedia.org/wiki/File:National_IQ_Lynn_Vanhanen_...
There are enormous differences between Brazil and Congo, between Vietnam and China, between Singapore (which you can’t even see) and Indonesia.
The Bell Curve has been amply debunked already, and in any case is decades out of date.
By contrast, per capita income explains 46% of variation in average IQ among countries, ...
Correlation is not causation. If race, nutrition, and parasite load (e.g., malaria) are causative factors of IQ (they are!), we would expect per capita income to be highly correlated with IQ. And indeed nutrition and parasites explain why equatorial Africa cranks out hideous poverty and average IQs in the 50s.
The Taiwan average is almost two s.d.s above the international average, which would be like an average IQ of 130, ...
Many Asians place a high priority on math practice, which readily increases ability. Islands of high math accomplishment have been created in the U.S. ghettos when a gifted teacher gives the students a culture transplant.
If you’ve ever traveled, it’s obvious that Qatar isn’t actually half populated with mentally retarded people ...
Mental retardation is actually defined by deficient global neurological abilities: physical coordination, verbal fluency, procedural memory, emotional perception, emotional control, and so forth. In other words, what are called skills of daily living. These talents are partially independent of general intelligence. It is perfectly possible for someone to be friendly, well spoken, and a great cook without having much abstract reasoning ability.
... while the Qatar average is almost two s.d.s below, which would be like an average IQ of just over 70, “borderline mentally retarded”.
Those thresholds were designed for tracking white children into special education classes for developmental disabilities. In a homogenous group, IQ is highly correlated with mental retardation, because global brain damage necessarily reduces IQ. However the thresholds are much less meaningful across races. A white kid with an IQ of 70 will probably need special help to learn to tie their shoes, and will lack the curiosity to care. A black kid with an IQ of 70 is nearly normal except for abstract reasoning ability, and has trainable worker written all over them.
The Bell Curve has been amply debunked already, ...
[citation needed]
Preferably a citation that lets hard data do the talking, looks at all the data, and understands the difference between correlation and causation.
much further separated than kids of different ethnicities in the same country
No. African-Americans, for example, have an average IQ of 85, very much like an up-market African country. Go read The Bell Curve.
only extremely bad families have a major environmental effect
The siblings in the adoptive families have nearly the full range of accomplishments of the general public. The possible explanations are (1) a sibling's accomplishments are mostly determined by internal causes, or (2) a sibling's accomplishments are substantially influenced by environment, and they experience a wide range of environments (even within the same family).
Other studies use twins raised together. The twins are more similar to each other than to their non-twin siblings, and the differences are explained mostly by IQ. This suggests that accomplishment comes mostly from internal causes.
the environments they were looking at were actually very similar
In terms of opportunity, they are extremely similar by adulthood. In the U.S., annual income at age 30 depends mostly on IQ, regardless of race, ethnicity, and other factors.
[1] http://pirls.bc.edu/timss2007/PDF/T07_M_IR_Chapter1.pdf
[2] http://en.wikipedia.org/wiki/IQ_and_the_Wealth_of_Nations