How to interpret a BBC news article on the effect of race on intelligence?

Question

In spring 1998, Robert Plomin claimed to have discovered a gene linked with intelligence. More recently, the Human Genome Project was cautious when approaching areas implying racial differences since research actually shows greater genetic differences within races than between races. However, not all individuals are endowed with the same intelligence and many believe this must have something to do with our genes and the way they interact with the environment. Identical twins are more likely to obtain the same score in an IQ test than twins from two separate eggs that have a different genetic make- up.

It is important to remember that genes work by interacting with the environment, so social factors will also infl uence intelligence. Intelligence tests may be more of an assessment of social factors, such as your educational background. Black children adopted into white middle-class families score signifi cantly higher on average than those in working-class families – implying a cultural slant to tests.

My classmate thinks that this article suggests that race is suggested as determining intelligence, because race is mentioned several times in the article as being linked to intelligence, whether genetically or through social factors. I want to check this before confronting her, but I think not. Who's right?

In my view, the first bolded sentence dismisses race as a determinant, because there are 'greater genetic differences within' each race than between different races.

The second bolded sentence also rejects race, because it only shows a cultural slant to tests and the first sentence of the last para states 'social factors will also infl uence intelligence.'

Source: ‘Are you born brainy?’, 17 November 2004, from BBC News at www.bbcnews.com

Answer 1

The relationships between race and IQ is a sensitive topic for many good reasons. It also raises a lot of deep questions about how to assess the role of genetics and environment.

A good summary of the literature on Intelligence with a discussion of racial differences can be found in the article "Intelligence: Knowns and Unknowns" (FREE PDF).

Within-group versus between-group variance

You write:

In my view, the first bolded sentence dismisses race as a determinant, because there are > 'greater genetic differences within' each race than between different races.

Group differences can be quantified. They are often quantified in psychology in terms of cohen's d (i.e., the size of group differences in terms of the standard deviation). You sometimes hear the claim that within group differences are larger than between group differences. This point is often highlights that just because a group on average is higher on a variable, there is a large proportion of people in the lower group than are higher than people in the higher group. The claim is often made to remind people not to stereotype. However, from a scientific perspective, we are often quite interested in the size of the difference between group means and the causes of such differences.

Importantly, group differences need to be very large before between-group variance is larger than within group variance. A grouping variable can have an effect of an outcome variable without needing to achieve this degree of group difference.

The size of the group differences says very little about whether the effect is causal.

Race as cultural slant

The second bolded sentence also rejects race, because it only shows a cultural slant to tests and the first sentence of the last para states 'social factors will also influence intelligence.'

I don't think the second bolded sentence entirely rejects race. However, the wording "cultural slant" does go some way towards this argument.

However, I'd challenge the "cultural slant" argument. In particular, if some environments lead to higher IQ, I would be more inclined to interpret that as an effect of environment on IQ rather than saying that IQ tests have a cultural bias. You might argue that the kind of intellectual functioning represented by IQ tests is more valued by some cultures, but that's quite different. Once you operationalise IQ in terms of a general factor that emerges from a wide range of cognitive ability measures, then it makes sense to proceed with this. This is the variable that has been validated (i.e., predicts educational attainment; entry into prestigious professions; job performance particularly in cognitively demanding job; future income; etc.)

Also, to show that environment influences IQ is interesting, but it does not refute the role of genetics. Researchers in this space (e.g., looking at twin studies) often aim to estimate the proportion of variance in IQ due to environment and genetics. To show an effect of $A$ does not mean that there is not an effect of $B$.

Answer 2

Simple minds

We believe the burden of proof must shift to those who argue for a 100% culture-only position... they need to address why, if important minority-specific developmental processes such as stereotype threat (Steele, 1997) and racial stigma (Loury, 2002) exert such a powerful influence on school achievement, the correlation matrices representing developmental processes can be so similar across ethnic and racial groups (Section 5). They need to explain why, if gene–environment interactions are as widespread and difficult to disen- tangle as often claimed (e.g., Block, 1995), identical twins reared apart grow to be so similar (Bouchard, 1996; Bouchard & Loehlin, 2001). Some culture-only hypotheses are too ambiguous to be tested.

These strike me as quite simplistic and ad hoc arguments. Bouchard is adding further empirical twin data about personality and "special mental abilities", rather than IQ. Neither are "developmental processes" the same thing as IQ. And yet he thinks the IQ gap is a real - unitary - difference in intelligence. Why?

A lot of researchers (this is a highly cited article) seem to have invested a lot in IQ tests being an inherently fair, and complete, measurement of G. When in reality these very same arguments can be used to support the claim that the IQ gap is not representative of intelligence.

Such is 'psychology'. If personality and so on is not much of a cultural artefact, culture and genetics are really pretty distinct here, then are the effects of a racist culture mediated by the genetics of a person? If different ethnicities only differ in IQ and not pace of development, do IQ scores reflect a norm or definition of G which is valid for all groups?

These are only "too ambiguous to be tested" if we believe IQ tests are the gold standard which cannot be added to, even interpreted. IMHO it's from a false need for simplicity: one score with the best correlations. Sadly, social and individual psychology surely doesn't work like that