May 31, 2023
Is the white-black IQ gap shrinking in the United States?
Although many assume that the existence of sizable disparities in average IQ among the races simply must be “pseudoscience,” that different ethnicities average different levels of mental ability is one of the most overwhelmingly well-established findings of the social sciences.
Exactly why ancestral groups have varying IQs remains controversial, but the fact that they do is not disputed among the well-informed. A general rule of thumb among researchers has long been that whites and blacks differ in average IQ by about one standard deviation.
For example, the 2001 meta-analysis by Roth et al. of 105 cognitive test studies with a sample size totaling over 6 million found an average gap of 16.5 IQ points, or 1.1 standard deviations.
IQ analyses are often performed with the benchmark norm for whites being set to 100 and a standard deviation of 15. Hence, if blacks averaged 85, that would mean that 84 percent of whites are smarter than the typical black. But it would also mean that one out of six blacks, or approaching 7 million these days, would be smarter than the median white.
It’s natural to be interested in whether the white-black gap has been narrowing over the generations. If the cause of the IQ gap is racism, then the obvious huge decline in racism over recent generations should have boosted black IQ scores relative to whites. Has it?
Note that it’s by no means impossible for better conditions to boost IQ, just as many populations are taller than in the past. In fact, around the world, raw scores on IQ tests tended to go up over the course of the 20th century (the now-famous Flynn Effect), requiring test publishers to toughen their scoring standards.
In general, though, the Flynn Effect drove up scores comparable amounts in most places around the world. However, some examples of sharply improved living standards boosting national average test scores relative to global averages have been observed; for example, relative IQs are up a lot, as is height, in South Korea over the past two generations. (Unfortunately, the birth rate of these taller, smarter South Koreans has dropped precipitously.)
Similarly, African-American athletes likely suffered in the increasingly distant segregated past from subpar nutrition and a lack of advanced training. My impression is that black players were not quite as outstanding combinations of speed and strength relative to whites within living memory.
Or consider the two best ballplayers of the 1950s, back when baseball was still the leading sport, Willie Mays and Mickey Mantle. Statistics savant Bill James’ judgment is that the white center-fielder, while less durable, was even faster and stronger than the great black slugger and ball hawk.
This is not to say that mid-century blacks were small or weak, just that their current athletic dominance only became really clear later, perhaps in the late 1960s. That probably had something to do with improvements in the black standard of living and with the integration of teams from the 1940s onward.
On the other hand, the O.J. Simpson Era when blacks fulfilled their athletic potential was a long time ago. If black cognition was being held back for much the same reasons as black athleticism once was, we should probably be seeing evidence of it catching up lately.
Are black IQs going up relative to white IQs?
Perhaps slightly, due to elite immigration from Africa and increased numbers of part-black children with one white parent. For example, Barack Obama, who is a reasonably smart guy, is both the child of an African elite and of a white.
But the traditional gap between whites and descendants of American slaves with no recent white ancestry (for instance, Michelle Obama) appears to be about as large as ever in a lavish National Institutes of Health study of over 10,000 American children called Adolescent Brain Cognitive Development (or ABCD) that recruited and extensively tested a nationally representative sample of 9- and 10-year-olds in 2016–2018.
This is a gold-standard project intended to be the biggest and best study of the teenage brain. The NIH recently ponied up an additional $290 million for the second seven years of the ABCD. Data sources utilized include the NIH Toolbox of cognition tests (an elaborate multi-hour state-of-the-art version of the traditional IQ test), full genomes, in-depth surveying of socioeconomic status, and five (count ’em) MRIs, one every two years from age 10 through 18.
Because the taxpayers are shelling out for the ABCD, it was intended to be operated on the “open science” model with the data broadly available. But that freedom is being squeezed because some scientists have lately used these huge new longitudinal databases to look into key questions of the age and found politically unwelcome answers. For example, tenured Cleveland State professor Bryan J. Pesta recently was fired after publishing a paper using the Philadelphia Neurodevelopment Cohort to examine the impact of racial admixture on IQ.
(To learn more about these new databases, read my 2021 column about a study of how white are self-identified blacks and how that correlates with IQ.)
At the Human Varieties blog last week, “Chuck” posted a summary of cognition scores from the ABCD by ethnicity.
He set the average score of children identified by their parents as non-Hispanic whites (and nothing else) to 100.0, with a standard deviation of 15.
On this metric, the 437 Asians and part-Asians averaged 105.1, which would be the 63rd percentile for whites. The 59 pure-blooded (non-hapa) northeast Asians (Chinese, Koreans, and Japanese) scored a spectacular 111.2 (77th percentile among whites), although that sample size is only modestly confidence-inducing.
(Note that these are parental identifications of their children’s race and ethnicity and not determined by genome analysis. But an earlier paper looked at parental identification compared to Ancestry.com-style genomic determination of race and didn’t find all that much difference in impact on IQ. The federal government’s system of racial self-identification tends to be good enough for government work.)
The 2,096 Hispanics had a mean IQ of 91.7, or the 29th percentile of the white distribution.
The 1,926 blacks scored 85.2, or the 16th percentile for whites: a 0.99 standard deviation gap.
So this 2016–18 black score is only a tiny bit better than the 20th-century rule of thumb of a one standard deviation (15 points) racial gap. Still, it might depict modest improvement from the 1.1 s.d. gap (16.5 points) cited by the more rigorous researchers in the past.
The black score is being slowly boosted by the growth in the numbers of three groups who aren’t core African-Americans (i.e., the descendants of American slaves). The 397 children who are identified by their parents as both black and white average 91.9 (29th percentile), the 57 black Africans 90.6 (27th), and the 46 black Caribbeans (e.g., Jamaicans and other non-Hispanics) 88.7 (23rd).
Still, these Hispanic-level scores suggest that immigration and intermarriage aren’t going to solve America’s racial gap in IQ problem anytime soon.
Worse, the 1,426 kids identified by their nonimmigrant parents as black, nothing else, and non-Hispanic (e.g., Michael Jordan and most other mainstream African-Americans) averaged only 83.0 (13th percentile). Seventeen points is a 1.13 standard deviation gap.
That’s not good: 83.0 is the lowest of the 28 ancestral groups listed by Chuck. (The second lowest is Black and Other Puerto Rican at 88.0.)
That 17-point gap might even be a little larger than that typically seen in the second half of the 20th century.
Why? One reason might be that in the second half of the 20th century, downscale blacks tended to have a lot more kids than upscale blacks. However, poor blacks stopped having so many more children than affluent blacks in the 1990s, so that dysgenic and/or dyscultural mechanism shouldn’t be as strong at depressing black scores going forward.
On the other hand, I think this white-black gap could be inflated somewhat by the white score being a little higher than is realistic due to a few factors pushing up the socioeconomic status and IQ scores of whites in the ABCD sample. In general, the white parents’ education levels struck me as higher than I expected.
One reason: Because kids were only recruited from school districts within fifty miles of one of the 21 research centers with MRI scanners, only 12.3 percent of the ABCD subjects are rural vs. 17.5 percent in national data.
Another potential issue is that 15 percent of the sample are twins because the human sciences love twins for all they can learn from comparing them. My impression is that at least among whites, twins tend to be more likely born to well-educated older mothers who delayed childbirth for career reasons and then had to undergo fertility treatments.
A third (but quite tiny) factor inflating white test scores is that Chuck’s analysis breaks out the 41 participants of North Africa and Mideast descent separate from whites: The NAMEs average 96.4.
So, these subtle factors might be expanding the white-black gap more than is real, although likely not by a sizable amount.
What should be done about racial gaps in IQ?
Progressives have this insane tendency to assume that if it really is true that blacks aren’t as smart as whites on average, then the only logical thing to do would be to murder all of our fellow black citizens in Treblinka-style death camps. Why? Because, they apparently reason, only Nazis, as they’ve so often said, think blacks have lower mean IQs, so if it turns out that the IQ Nazis are right, well, that means Hitler should be our role model.
Or something. You can never quite get liberals to articulate why they are convinced it would be the end of the world if there are racial differences in intelligence, other than that’s the ditch they’ve decided to die in and it would be embarrassing for them to turn out to be wrong.
My view is that the IQ gap is more like the sprinting gap, which helps explain much about the realities of sports (why are strong-armed African-American baseball players more likely to play right field than third base?) but is not, to say the least, a good reason to kill all the whites.
In reality, the main public-spirited step we can take regarding racial gaps in intelligence is to Raise Awareness: Racial intelligence differences are real, large, crucial for understanding American society, and not going away anytime soon. It’s something we have to live with. Just as the foot-speed gap doesn’t justify witch-hunting football coaches for being racist against white cornerbacks, the manifold effects of the intelligence gap don’t justify the current paranoid conspiracy theorizing about white racism.
Much of the madness of the 2020s, such as the post–George Floyd murder and car crash sprees and California’s reparations plans, is likely attributable to blacks figuring out that they weren’t going to catch up but that they could guilt-trip whites into paying reparations because whites have been disarmed of mentioning the IQ and crime gaps.
A new culture in which the basic truths of the social sciences are widely discussed would be a better one than what we have been saddled with in the present decade.