Race, IQ, and The Bell Curve

Before working on the mix of projects that I currently do, I taught psychological assessment for a couple of years. I taught graduate students the basics of how to administer and score a range of psychological tests including IQ tests. In every class I taught on intelligence and achievement testing, I would review the evidence and controversy surrounding race and IQ. I always tried to include a couple of questions about the topic on one of the tests or midterms.

Being the only slightly reformed packrat that I am, I still have some of my old tests saved in my filing cabinet. I was able to find this question amongst them:

Years from now, you do a weekly radio call-in show. A caller tells you and the rest of the listeners that they read a book called The Bell Curve about IQ and are convinced that people of color, as a whole, have lower IQ scores. How do you respond? (5 points)

a.  “Not true. No studies show meaningful differences between races on tests of IQ.”

b.  “True. People of color just have lower IQs and we need to acknowledge this fact, end of story.”

c.  “True, but all of the differences between races can be accounted for solely in terms of poverty and nutrition.”

d.  “True, but it is only due to cultural biases in the tests themselves.”

e.  “True, due to a range of different social, cultural, and historical factors, but there is no good reason to believe in intrinsic intelligence differences between races.”

Do you know the answer? If not, no worries. After reading the rest of this blog article, you should be well-equipped to answer this particular question. Before we press on, how about you make a note of which of those answers intuitively seems to be the best.

The truth is that there are significant differences between Whites and African Americans on IQ test scores. The overall differences measured across a range of studies are about 15 points (one standard deviation).[1] In everyday terms, this is a meaningful difference. One standard deviation in male height for example is about 3 inches. One standard deviation in IQ is a little more than the difference between a high school graduate and a college graduate or that of a college graduate and a typical person receiving a PhD or MD. In short, a one standard deviation difference is usually large enough to be both obvious and meaningful.

Now a facile approach would be to simply say something along the lines of “Well Whites are just smarter than African Americans” and leave it at that, but this is not a click-bait article and we are going to journey a little deeper into the available research on the topic. We do know that there are numerous other factors that might contribute to such a difference that do not rely on an essentialist view of race.

In some very thorough research on intelligence testing in social context, Weiss et al looked at a range different factors to tease out a more nuanced understanding of intelligence and why there are differences between racial groups in their performance on standard tests of intelligence.[2] Most of what follows will largely be a summary of their work for the lay reader. Professionals are highly encouraged to read the text in its entirety, especially chapter 4. In any event, the data that follow are drawn from Weiss et al unless otherwise noted, but the interpretations are my own.

The IQ difference of about 15 points between White and African American adults drops to about 11 points when controlling for mediating variables such as education, occupation, income, geographic region, and gender. That is these factors all play a role in IQ differences, but do not fully explain the differences observed.

Drilling down deeper, Weiss and colleagues found that differences in IQ between Whites and African Americans are stratified by age cohort. For those 65-90 years old, the difference was 19.3 points. For those 45-64 years old, the difference was 17.2 points. For those 20-44 years old, the difference was about 13 points. Finally, for those 16-19 years old, the difference was 10 points. Other studies have shown this trend continuing down to younger children where the difference is only 6 points and the differences between White children and African American children have dropped by about 40% over the last three to four decades.[1]

So what is going on here? If we were to take seriously the notion of racial essentialism and the idea that ‘some races are just smarter than others,’ we would expect to see differences that are stable across age cohorts and across the decades. If race is an essential and immutable variable it should be consistent. We do not see that. What we see instead are differences based on age and time.

Why? What has significantly changed in the last 30-60 years that might account for a notably different environmental experience related to race?

Well one obvious answer is the dismantling of de jure segregation and the structures of racial oppression associated with it. Jim Crow, that is segregation by law, was actively enforced well into the mid 1960’s and was probably not officially dead until the Swann v. Charlotte-Mecklenburg Board of Education case was affirmed 9-0 by the Supreme Court in 1971.

African American adults between the ages of 65 and 90 would have lived through active and legal discrimination as a part of their formative years and early adulthood. They would have been subjected to the extreme racial terrorism of Jim Crow and been denied numerous educational and occupational opportunities. They may have come of age in neighborhoods that were actively neglected under the perverse logic of “separate but equal” that ensured that things were always very unequal. These experiences left their imprints on the minds and brains of generations of people of color, most notably the eldest among us.

Racism is not ‘over,’ though the days of de jure segregation are. These differences are becoming apparent in the rapid gains that young African American children have been making in closing the IQ gap with white students. Differences still remain and are likely to persist for a significant period of time, not because of essential racial differences but because of the enduring and deleterious effects of de facto racism and systematic biases.

Another important point to make before we wrap up is that race, while a meaningful factor in discussions of IQ, is not the most important factor. In adults, race accounts for about 15% of the variability in IQ score whereas education accounts for roughly double that (29%).

So back to that question again only modified slightly:

You are on social media. A poster tells you and the rest of the readers that they read a book called The Bell Curve about IQ and are convinced that people of color, as a whole, have lower IQ scores. How do you respond? (5 points)

a.  “Not true. No studies show meaningful differences between races on tests of IQ.”

b.  “True. People of color just have lower IQs and we need to acknowledge this fact, end of story.”

c.  “True, but all of the differences between races can be accounted for solely in terms of poverty and nutrition.”

d.  “True, but it is only due to cultural biases in the tests themselves.”

e.  “True, due to a range of different social, cultural, and historical factors, but there is no good reason to believe in intrinsic intelligence differences between races.”

How would you answer now? Has your answer changed?

[1]Lichtenberger, E. O., & Kaufman, A. S. (2012). Essentials of WAIS-IV Assessment (Vol. 96). John Wiley & Sons.

[2]Weiss, L. G., Saklofske, D. H., Coalson, D., & Raiford, S. E. (Eds.). (2010). WAIS-IV clinical use and interpretation: Scientist-practitioner perspectives. Academic Press.