Unfortunately we cannot avoid the question of Race and IQ when discussing racism because modern day racists, those who call themselves “race realists” today, are actively abusing science to push for policies such as:
- Immigration quotas based on race
- No bussing
- Ending affirmative action
- Cut or scale back programs aimed at disadvantaged children.
Basically the thrust of their argument is that all the inequality seen in America is “natural” and not caused by discrimination, unequal opportunities or racism. In their view it is just a reflection of racial inferiority.
This means it is very hard to discuss racism without addressing these issues. Here I have tried to show an accumulation of various studies and facts which questions the narrative of the “race realists.”
Remark on Race
When I write about race, I mean the social construct of race. What people identify as or the racial categories people put others in based on the established conventions of the society they live in. To understand better why most biologists don’t consider race biological, despite the fact that you can clearly see the difference between a Chinese and a Swede, read this explanation. The short answer is that your appearance is a tiny fraction of your genetic makup. A Swede can in principle be more similar to an Asian genetically, than another Swede. Most of your gene variation is not exposed on your exterior.
Minnesota Transracial Adoption Study
This is an important study popular among “race realists.” It is supposed to show black intellectual inferiority.
Researcher Drew Thomas provides a counterpoint here by reinterpreting the study, he gets the following scores:
Child and Parent | IQ Score
White-White | 101.8
Black-White | 98.5
Black-Black | 89.4
Which should that the difference between a white child adopted by a white family and a black child adopted by a white family is minimal. The difference could easily be explained by discrimination that exist in society at large.
How did Drew Thomas reach this conclusion? By accounting for the following flaws in the original study:
- There was several white adopted kids with low IQ score which the researchers lost track of. Followup studies did not include them. The black adopted kids they lost track of did not have unusually low IQ score. Hence this skewed the results in favor of the white kids.
- One had not taken into account the different times the various adopted children where studied, and hence ignoring the Flynn effect. If you take an IQ test today you must get more correct results to achieve an IQ score of 100 than if you took the test 50 years ago. Hence older test scores cannot be compared to newer test scored. Scores must be relative to the population at that time.
Last point needs some elaboration. Say an Asian in the year 2000 takes an IQ test which was made in 1950. He gets a score of 60/80, and in the year 2000 this is the score the average person would get on this test. That means his IQ would be judged as 100 in the year 2000. This is because IQ tests measure you deviation from an average in a given population.
Now imagine that back in the 1950s the average score was instead 40/80 for this same test, and that a score of 60/80 would have given him an IQ measure of 120 back in the 1950s.
So assume the average white person in 1950 scored 40/80 on the IQ test, while in 2000 the average white person scored 60/80 on the IQ test. Then it is obvious that you cannot take the year 2000 test results of an Asian man and compare it to 1950 test results for whites to conclude Asians are more intelligent.
This is basically the kind of mistake that the Minnesota study did, and which Drew Thomas is trying to correct in his paper. When comparing two different racial groups, you got to compare their score against the reference group for the right time period. A year 2000 score cannot use a 1950s score average as a reference group.
Multiracial Student Score
“Race realists” frequently have a fear of racial mixing which they claim pollutes the gene pool of whites and make the population inferior. The results above should suggest that will not happen, but do we have studies of mixed children? In fact we do. Here are the results from a Brookings study:
There is no statistically significant difference between whites and multiracial students on tests in science, math, and writing. For reading, multiracial students slightly outperformed both white and Asian students, becoming the highest-performing racial or ethnic group with reported data by a small but statistically significant margin
Heritability of IQ Score
How much of your IQ is heritable. Many studies show around 80% heritability. This means 80% of the variations we see are explained by genes. However Eric Turkheimer has a study that shows it is heritability varies with socio economic status (SES).
- High SES kids. 0.72 heritability. Shared Environment 0.15.
- Low SES kids. 0.10 heritability. Shared Environment 0.58.
Not everyone get this result. But he is not alone. E.g. this study of twins in Sudan gives heritability of 17% and influence of shared environment of 60%.
Is Heritability a Sensible Concept?
However heritability itself is a questionable concept. Heritability of obesity is measure to 6–85% depending on population.
Why is this important to mention? Because people easily think 80% heritability of IQ somehow suggest environment cannot be changed to significantly alter your IQ. The Flynn effect suggest otherwise and so does the heritability of obesity and height.
We know for a fact e.g. that you can drastically alter your obesity by changing your diet. You are not locked to being obese because of genetics. The way heritability is measured makes it look like that.
E.g. consider countries A and B containing people who are genetically the same. In A fast food is sold at every corner. In country B, only healthy food is sold. People in A will be much fatter than in B. However if you measure heritability in A and B separately you will find in each country that the main explanation for the variation in obesity within each country will be the result of genes.
This makes it look as environment does not matter. But the effect of environment is underestimated because you measure heritability in a more or less homogenous environment.
High Test Scores of Black British Immigrant Children
When looking at British scores, if you lump all blacks into one group they score slightly below the British whites. However if you look at specific nationalities then we have the following GCSE scores above whites of:
- Nigerians 21.8%
- Ghanalan 5.5%
Some other non-white groups do very well as well:
- Chinese 38%
- Sri Lankan Tamil 32.5%
- Iranian 31.9%
- Vietnamese 31.5%
- Indian 29.9%
Interestingly some white groups do worse:
- Italian -2.28
- Eastern European -23.9
- Portuguese -45.9%
If you look at subgroups of Africans, like Igbo do as good as Indians and Chinese.
Africans speaking Luganda and Krio did better than the Chinese students in 2011
Highest performing pupil in GCSE was a a Nigerian Igbo girl. She was also the fastest runner. So achieving best score despite spending more time on sports.
Is Crime Rate Among African-Americans Cause by Genetics or Environment?
Racists are eager to claim that the injustice one sees in America about the number of African-Americans shot by police and incarcerated is simply the natural order of things. It is simply what they view as a natural outcome of what they deem people with inferior genes predisposed to crime.
Let us examine the validity of this claim. African-American have 16–24% white genes. According to the “race realists” (modern day racists) that means African-Americans are essentially up to 25% better than Africans. More white genes is supposed to make you smarter and less criminal according to them.
Let us take two simple metrics related to crime, homicide rate and incarceration rate for the whole US population:
However for the African-American population it is:
If these is all due to African genes, these numbers should be much worse in Africa. Africans do after all have 100% African genes. But this is not the case at all.
- Worst incarceration rate in Africa is Rwanda with 464 out of 100 000. That is 5x lower than the US rate!
- Central African Republic has a rate of 16. Almost 150x lower than the US.
- Congo despite its terrible history has homicide rates of 13.55 and incarceration rate of 29.
These are bad places, but what about more developed African countries? Surely these make more sense to compare with America? After all African-Americans don’t live at African levels of poverty.
- Ghana has an incarceration rate of 51, which is lower than Norway. The homicide rate is 1.68, which is dramatically lower than the 20.9 rate among black in America.
To not be accused of cherry picking we can look at the average homicide rate for all of Africa which 12.5. That is quite a lot higher than the US homicide rate of 20.9 for African-Americans. In fact the only country in Africa beating US homicide rates is South Africa at 36.4. But SA is an extreme outlier which in my opinion should not be included. Its highly problematic history with both blacks and whites makes the country in many ways more similar to the US.
Perhaps there is simply something wrong with how African-Americans are treated in the US? Perhaps institutionalized racism is a real thing? How could we check that?
One idea would be to see outcomes for African-Americans relative to countries much poorer than the US for some key metrics.
- Infant mortality rates for blacks in the US is a staggering 14.
- Sri Lanka infant mortality is 8.23.
- Cuban infant mortality rate is 5.5.
Hence before jumping to conclusions and assume that genetics explain the disadvantage of African-Americans, one has to be able to explain why African-Americans living in one of the riches countries in the world has worse outcomes than poor people and Africans in other parts of the world. It points to a cultural and societal explanation.
Proof of Racism in the American Justice System
Proving racism in the justice system is hard for several reasons. If say African-Americans are arrested at a higher percentage than whites this can have several possible explanations:
- African-Americans simply commit more crimes.
- There is not difference between African-Americans are subject of racism.
- Both factors are present. African-Americans commit more crimes but are also subject to racism causing higher arrest numbers.
I think the latter is the more realistic explanation. But how do you prove it? The University of Maryland has an excellent study proving this.
It gets around a difficult problem. When cops arrest an African-American it is very hard to determine in an individual case whether he is grossly exaggerating the crime to get a higher sentence because the defendant is black rather than white.
However under the right circumstances such a bias can cause a statistical anomaly proving the prejudice. This is what we see here. The convictions around 280–290 gram for blacks is far above other sentences. This is highly unrealistic. There is no reason why criminals should choose to have a quantity of drugs exactly within this range. However because there are advantages to the police starting from 2010 in term of getting higher convinctions, they have an incentive to lie and and exaggurate to put defendants in that bracket.