07 Jun The Science and Ethics of Group Differences in Intelligence: Part 1
The Science and Ethics of Group Differences in Intelligence: Part One
Winegard, Winegard, Boutwell, and Shackelford (henceforth referred to as WWBS) have written a rejoinder to the Vox article that I co-authored with Eric Turkheimer and Richard Nisbett. They begin by describing a few lines of research that they allege we “cherry picked.” Next, they review “copious” so-called “data” to advance the hereditarian case that the Black-White gap in average IQ test scores is caused by genetic differences between racial groups, and they conclude by chastising those who would express moral disapproval of hereditarianism’s proponents.
This is the first in a series of blog posts that use their article as a foil to develop some more general ideas regarding the science and ethics of studying genetically-based group differences in complex human phenotypes, including intelligence.
This first post will narrowly hew to the first half of their article. Specifically, I will respond to their contentions regarding: (a) whether the heritability of intelligence differs by social class, (b) the nature of adoption-related and secular gains in intelligence, vis-á-vis the Black-White IQ score difference, and (c) historical trends in the magnitude of the average Black-White IQ score difference. Overall, I disagree with WWBS’ conclusions on these points. As I argued in Vox, I think these general lines of evidence are, on the whole, consistent with an environmental explanation of Black-White intelligence differences.
But, more importantly, I think that this research sheds a dim light on the hereditarian hypothesis. In my next post, I will consider questions that are rarely articulated by either hereditarians or their opponents: what would convincing evidence for WWBS’ hypothesis look like, given contemporary genomic science, and what is empirically and theoretically challenging about resolving this debate more satisfactorily? With a clearer picture of what convincing evidence would look like, I then consider WWBS’ rationale for embracing hereditarianism and find it insufficiently rigorous.
Finally, in my last post, I will revisit the question of moral disapproval: What are scientists’ ethical responsibilities when talking about Black-White differences in intelligence?
Part One: Empirical Quibbles
Social class differences in the heritability of intelligence
In the Vox article, we wrote, “The heritability of intelligence, although never zero, is markedly lower among American children raised in poverty;” this sentence contained a hyperlink to a 2003 study by Turkheimer and colleagues. In response, WWBS note that, in a more recent meta-analysis by Tucker-Drob and Bates, the effect size estimated by Turkheimer et al. (2003) was the largest of the studies that tested the interaction.
At this point, I need to disclose to you that Elliot Tucker-Drob, who authored the meta-analysis, is my husband, and frequent scientific collaborator. You can probably guess how odd it is to have my’s partner’s work cited as a rejoinder to my own. (Perhaps in other marriages, partners do not read each other’s writing?) But, what is even odder is how the key conclusions of the meta-analysis continue to be misrepresented.
WWBS write: “First, many other studies that have examined the heritability of cognitive ability have found no or little evidence that heritability is lower in poor children.” No evidence, except for a meta-analysis of data from 10,831 twin and sibling pairs in the U.S.! Tucker-Drob and Bates (2016) concluded that, “In U.S. studies, we found clear support for moderately sized Gene X SES effects.” Figure 1 of that paper shows the predicted relationship between socioeconomic status and the heritability of intelligence given the meta-analytic effect size for U.S. samples.
WWBS go on, “In a meta-analysis of available studies, Tucker-Drob and Bates found that there was no overall effect of socioeconomic status on the heritability of intelligence (in all studies, including ones from outside of the US), but that there was an effect in studies from the US.” Tucker-Drob and Bates (2016) reported but did not interpret the ‘overall effect’ of socioeconomic status on the heritability of intelligence in all studies, because this parameter is meaningless. They tested whether all the effect sizes from all the different studies could be treated as arising from the same population; they couldn’t. They also tested whether the difference between the effect in U.S. studies and the effect in non-U.S. studies was itself significant; it was. Interpreting the average of the U.S. and non-U.S. parameters doesn’t make sense.
They continue, “That effect, however, was small, and much smaller than in the original Turkheimer study that inspired this line of research.” As Turkheimer et al. (2003) makes clear, theirs was not the original study that inspired this line of research. This line of research dates back to Sandra Scarr (1971, whose original hypothesis was quoted in the first paragraph of Tucker-Drob & Bates, 2016), and there were subsequent tests of the hypothesis by David Rowe and colleagues. Given this, why should the Turkheimer et al. (2003) effect size be the anchoring point? The interaction effect is small, but not zero. Again, please see Figure 1 from Tucker-Drob and Bates (2016): among the poorest children in the U.S., the heritability of intelligence is estimated to be ~26% versus ~ 61% in children from the wealthiest families.
WWBS then question whether the interaction would hold past childhood: “…even if the heritability of intelligence in some groups was low in childhood (say, 10% or so), it is not clear that it would remain low into adulthood.”
This exact question was tested in the meta-analysis! From Tucker-Drob and Bates (2016), emphasis added: “We examined whether test performance was measured in childhood or adulthood, childhood age of testing, whether the tests measured either achievement and knowledge or intelligence, whether a single or composite indicator of SES was used, and whether the tests were of single ability or a composite cognitive measure. None of these additional moderators achieved statistical significance, and the cross-national difference in the Gene × SES effect remained when each of these possible moderators was entered into the meta-analytic model. Thus, the cross-national difference identified does not appear to be an epiphenomenon of cross-national differences in the age ranges examined or the particular intelligence or achievement outcomes measured.”
WWBS then continue: “Third, researchers recently found evidence that the heritability of intelligence is the same in Blacks and Whites. In fact, this was found in the same sample of youths (but also including non-twin siblings and half-siblings) that Turkheimer used for his original analysis.” There are no twin datasets with enough Black participants to provide a strong test of racial differences heritability of IQ is lower in Blacks. (To give you an idea of how large the necessary sample size might be, Tucker-Drob and Bates, based on their meta-analysis, estimated that a sample of 3300 twin pairs would be necessary to detect the gene × socioeconomic status interaction with sufficient statistical power.) Power for designs that compare full-siblings vs. half-siblings (such as in the study they cited) is even lower than for a twin design.
They then suggest that the Black-White IQ gap persists even controlling for social class: “However, even if there is a small race difference in heritability estimates, researchers could limit their analyses to middle- and upper-class Blacks and Whites, where the IQ gap is about the same, if not larger, than it is among lower-class Blacks and Whites.”
This comment implies that the environmental experiences of Blacks and Whites in America can be equated by simply equating groups on income. This is sociologically uninformed. Pages and pages could be written about this, but I will limit myself to two examples. First, middle- and upper-income Black households, on average, live in neighborhoods that are much more disadvantaged than the neighborhoods that Whites with similar incomes live in. For example, Black households earning $60,000 per year live in neighborhoods that are comparable to the neighborhoods where White households making only $11,800 live. Second, even when you limit comparisons to the top quintile of income, the median wealth of Black households is 43% of the median wealth of White households.
So, a child growing up in a middle- or upper-class Black home, as defined by household income, still begins life with significant material and contextual disadvantages compared to a child growing up in a White household with comparable income. As LeBron James recently said, in response to having a racial slur painted on his home, “…no matter how much money you have, no matter how famous you are, no matter how many people admire you, being black in America is — it’s tough.”
Finally, they conclude this topic by writing, “In other words, the fact that intelligence may be less heritable in impoverished youths is irrelevant to the debate about the cause(s) of race differences in intelligence.” For a fact that is irrelevant to their argument, WWBS spend a lot of time objecting to it! But actually, here is a point of agreement. The hypothesis about gene X socioeconomic status interactions is about poverty, not about race per se. It informs the debate about race differences because Black households are poorer than White households, and because it illustrates the point that heritability is not a static property of a trait. Still, even if the heritability of intelligence in the poorest Black children is 40% rather than 20%, the entirety of the difference in IQ scores between Black and White children could still be due to environmental rather than genetic differences between groups.
Adoption-related and historical gains in intelligence
In our Vox article, we described evidence that (1) children adopted into wealthier homes had higher IQs, on average, than children who remained with their disadvantaged birth parents, and (2) the average IQ is increasing over historical time. The point being – even moderately to highly heritable phenotypes like intelligence can be substantially improved via environmental changes.
To criticize this line of thinking, WWBS write, “… the ‘poor families’ from which the children were adopted were so poor that the children were suffering from primary undernutrition at the time of adoption.” WWBS seem to believe that being so poor that one suffers from undernutrition is a rare event that couldn’t possibly be relevant to understanding the diverging life outcomes of Americans today. In fact, in about 10% of American households with children (and 17% of Black households with children), the children are food insecure.
“Second, most of the studies examined IQ at young ages, which is important because the genetic effect (heritability) of intelligence increases with age, a phenomenon dubbed the “Wilson effect” by behavior geneticist Thomas Bouchard; therefore, the gains might be smaller if the subjects had been tested later in life.” Doubtful. In the Swedish adoption study co-authored by Turkheimer, the intelligence of the adoptees was tested at age 18, when the heritability and rank-order stability of intelligence are already high.
“Third, even if one accepted the large IQ gains uncritically, those gains are still consistent with heritability estimates of 60%, as was clearly noted by Herrnstein and Murray in The Bell Curve (see page 771, note 86).” Yes, although I’m not sure what the point of this comment is, as no one involved in this conversation is denying the heritability of intelligence. Also, this point was clearly noted, three years before The Bell Curve, by Turkheimer.
“And, fourth, as with the “Flynn effect” the adoption gains are primarily on subtests that are not strongly associated with g, which means that the cause(s) of the Black-White gap probably are different from the cause(s) of adoption IQ gains….The increase in IQ across the 20th century has little to do with the IQ disparity between Blacks and Whites.” This is where objections to talking about adoption converge with objections to talking about the Flynn Effect. Even if the cause(s) of the Black-White gap are different from the cause(s) of adoption IQ gains, and even if those are still yet different from the cause(s) of Flynn effect gains, there is no compelling reason to think that all three cannot be due to environmental causes. Our original point still stands – environmental changes can cause large and meaningful gains in IQ scores.
Cohort changes in the Black-White IQ score gap
WWBS acknowledge that our claim that the Black-White IQ score difference has shrunk “is supported by a few data points”, but also point to other data suggesting that the difference has stagnated in recent years. Richard Nisbett recently responded to this criticism (and I will link to it when it’s published), and I refer the interested reader there for a detailed rebuttal.
But let’s say that WWBS are correct (I don’t think they are), and the average IQ score difference between Black and White Americans has stagnated at 12 points rather than 10. What does this mean? To some, this clearly points to the intractability of the difference, and thus underscores the hereditarian case. To many others, this is further evidence that racism is alive and well in 2017 America. All of that to say, the observation of no change is inconclusive regarding cause, and we are left with nothing more than our politically-charged intuitions about whether genetic influence or racism is more likely to be an immutable force in American society.
What data are necessary?
Astute readers will have noticed from this conversation that there are rhetorical points that can be made by both hereditarians like WWBS and their opponents: the heritability of intelligence is lower in poor people but maybe not Black people; the IQ score gap has narrowed but it’s debatable how much; IQ can increase with the environment but the gains perhaps do not conform to the pattern of loadings on g. More crucially, none of the evidence makes much of a dent in either side’s thinking. If the heritability of intelligence in poor children was 10% rather than 20%, would WWBS give up on the hereditarian hypothesis of group differences? If they convincingly showed that adoption-related gains are limited to 6 IQ points rather than 12, would I care? Decisive data constrain one’s thinking, and the fact that neither WWBS’ thinking nor mine about the origin of group differences is much constrained by the data at hand is a sign that the data at hand are inconclusive about such questions.
Another sign that it is weak data is that the terms of the debate haven’t changed that much since the original Jensen / Rushton / Murray formulation. The Bell Curve was published nearly a quarter-century ago. In the intervening years, we have sequenced the human genome and made large strides in human genetics research generally. Yet here we are, still hashing out the old debates on race and IQ almost as if nothing has changed since 1994. A naïve reader, coming to our conversation about race and genetics, would probably be surprised at how little genetic content there actually is. Instead, there are a lot of inconclusive and abstruse conversations about measurement invariance.
In my next post, this is the question that I will consider in more detail: what would strong evidence regarding genetically-based group differences look like?