# Race differences in average IQ are largely genetic **URL:** \ **Category:** Great Debates\ **Created:** [May 27, 2012, 12:28pm UTC](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236 "2012-05-27T12:28:24Z")\ **Posts on this page:** 9\ **Page:** 35
**Author:** ![Great\_Antibob](https://avatars.discourse-cdn.com/v4/letter/g/e47c2d/32.png) [@Great\_Antibob](https://boards.straightdope.com/u/Great_Antibob)\ **Post date:** [June 22, 2012, 3:28pm UTC](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236/681 "2012-06-22T15:28:08Z")
> [@ralph124c](#): > > Kinda related, but I remember seeing a TV show, where they were talking about racial genetic characteristics. They were interviewing a black transplant surgeon, who mentioned that of all racial groups, it was hardest to find good tissue matches for African-Americans..this was due to that fact that Africans have much less genetic variability. It seems you’re mixing a few different things. It’s quite likely that, on average, African-Americans have less genetic variability than African-Africans. We know that most of the Africans brought to the US to serve as slaves came predominantly from one region in Africa. There weren’t many folks from East or South Africa who were a major part of the US slave trade. That’s not inconsistent with what we’ve been saying - populations, rather than “races” have been shown to exist genetically. [According to the US Department of Health and Human services](http://www.organdonor.gov/whydonate/minorities.html), it is difficult to find organ matches for African-Americans. Not because of low genetic variability but because of a lack of sufficient donors. While African-Americans donate organs in proportion to their population, they are overrepresented in terms of kidney transplants. You certainly want to minimize genetic variability between organ donors, which is why it is preferable to find a match among African-Americans. I’m guessing you heard that last part and got it conflated with the idea that low genetic variability was somehow bad for finding organ donors, when it’s the reverse. You would expect that if there’s low genetic variability, two random people would actually be closer matches than otherwise. It’s why we need tight health controls on mass-farmed pigs. They are so genetically similar that a bug that hits a single sick animal can easily infect every animal in a farm. It’s less of a problem on farms where there’s more genetic variability. Likewise a similar problem for bananas and other crops/animals where we’ve bred for certain traits. ---
**Author:** ![Paranoid\_Randroid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/paranoid_randroid/32/18_2.png) [@Paranoid\_Randroid](https://boards.straightdope.com/u/Paranoid_Randroid)\ **Post date:** [June 22, 2012, 4:12pm UTC](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236/682 "2012-06-22T16:12:44Z")
> [@iiandyiiii](#): > > Should employers and landlords have the right to reject an applicant strictly because of his or her “race”? I’m curious about this as well. I’ve heard the claim that in such matters the potential employer or landlord will focus on the properties of the individual in question, rather than relying on averages and stereotypes. Thus racial discrimination in those cases carries less legitimacy even if legal. But I can’t really buy that. The moment you must choose between multiple prima facie comparable candidates, if you feel one is likelier to be less intelligent and less moral then then you’ll use that purported information to make your choice. So is racial discrimination licit _everywhere_? ---
**Author:** ![New\_Deal\_Democrat](https://avatars.discourse-cdn.com/v4/letter/n/9d8465/32.png) [@New\_Deal\_Democrat](https://boards.straightdope.com/u/New_Deal_Democrat)\ **Post date:** [June 22, 2012, 6:46pm UTC](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236/683 "2012-06-22T18:46:56Z")
> [@iiandyiiii](#): > > Should employers and landlords have the right to reject an applicant strictly because of his or her “race”? I would say not, but in the case of an employer, these usually have the choice between a number of qualified applicants. Bosses hire people they think they will enjoy working with. Quite a few whites do not like blacks. I do not see how they can be required to hire them without affirmative policies. I am opposed to affirmative action because in practice it lowers standards for blacks and Hispanics. ---
**Author:** ![gamerunknown](https://avatars.discourse-cdn.com/v4/letter/g/6bbea6/32.png) [@gamerunknown](https://boards.straightdope.com/u/gamerunknown)\ **Post date:** [June 22, 2012, 11:04pm UTC](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236/684 "2012-06-22T23:04:30Z")
> [@Chuck11](#): > > geographically separated populations There are no significantly separated [human populations](http://www.pandasthumb.org/archives/2012/02/continuous-geog.html). > [@Chuck11](#): > > just as one can discuss congenital class difference in IQ. Class is not determined before birth (with the notable exception of the aristocracy) - class distinctions are permeable. > [@Chuck11](#): > > there have been studies which have linked genetic differences to race differences. Which does not demonstrate a genetic component to race. It demonstrates differences between races while controlling for socioeconomic status, but does not address the possibility of [systematic](http://users.nber.org/~kling/surveys/NLSY79.html) [bias](http://en.wikipedia.org/wiki/Stereotype_threat). A more thorough study would correct for the stereotype threat by using a single blind administration of a test with standardised instructions. Ignoring the validity of the test measure and focusing only on its reliability, the degree to which the difference in outcomes for attainment in that measure were socioeconomic could be determined. However, this would not change the fact that no genetic component would be isolated, as there are biological factors which are not genetic. Which wasn’t accounted for in the study either. [This section](http://en.wikipedia.org/wiki/Structural_equation_modeling#Interpretation_and_communication) in the wiki you linked stresses the necessity of experimentation in order to determine between causal hypotheses in order to control for confounding variables. ---
**Author:** ![gamerunknown](https://avatars.discourse-cdn.com/v4/letter/g/6bbea6/32.png) [@gamerunknown](https://boards.straightdope.com/u/gamerunknown)\ **Post date:** [June 22, 2012, 11:29pm UTC](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236/685 "2012-06-22T23:29:48Z")
Forgot: > [@Chuck11](#): > > …50/50 nature/nurture). If you can find an inconsistency name it. > [@Chuck11](#): > > we would conclude: more nature than nurture. Your specific example contradicts your proposed finding. ---
**Author:** ![Chuck11](https://avatars.discourse-cdn.com/v4/letter/c/96bed5/32.png) [@Chuck11](https://boards.straightdope.com/u/Chuck11)\ **Post date:** [June 23, 2012, 2:01am UTC](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236/686 "2012-06-23T02:01:17Z")
[quote=“gamerunknown, post:684, topic:623236”] There are no significantly separated [human populations](http://www.pandasthumb.org/archives/2012/02/continuous-geog.html). I can’t believe that you cited a post which I extensively debunked in the post’s very comment section. Refer to the comments by acoolcoolhippo and my interchange with the author. Moreover, I can’t believe that you cited _that_ post in good faith in support of what you are now saying. Either you are being deceitful or you’re unable to think for yourself. Instead of linking to a post, and making a cryptic critique, just explain in what way you think that I’m incorrect. And while you’re at it, explain what it means to say that there are no “significantly separated human populations” when the separation that we are discussion is geographic separation. (Note, I discussed the issue of “significance” in my comment section. As I noted here, when discussing it you need a metric. By at least some commonly used zoological metrics, my races are “significantly different.” > [@](#): > > Class is not determined before birth (with the notable exception of the aristocracy) - class distinctions are permeable. In a meritocratic society, class is largely determined by genes. So that would be “before birth” and my point stands. > [@](#): > > Which does not demonstrate a genetic component to race. It demonstrates differences between races while controlling for socioeconomic status, but does not address the possibility of bias. Good. Let’s discuss possible environmental causes. But let’s do this systematically. Read my [discussion first](http://occidentalascent.wordpress.com/2012/06/10/the-facts-that-need-to-be-explained/). I noted: **Of the plausible explanations for the Black-White score difference on measures of intelligence, at least in the United States, psychometric bias is not one of them. The most detailed work on this issue was Jensen’s 700+ page “Bias in mental testing,” which prompted both the National Academy of Science and the National Research Council to set up special committees to determine the issue; after exhaustive review of the evidence, both substantially agreed with Jensen (1980) (e.g., Wlgdor and Garner, eds, 1982). Since, a plethora of studies have investigated this issue using diverse methods and confirmed the conclusion. The most compelling evidence for the absence of psychometric bias comes from studies of measure invariance in standardization samples (e.g., Dolan, 2000; Dolan & Hamaker, 2001; Lubke, Dolan, Kelderman, & Mellenbergh, 2003; Edwards and Oakland, 2006). The finding of measure invariance implies that the factorial difference between groups are of the same nature as differences within groups (Wu et al., 2007). Wu et al. explain: “Mellenburgh (1989), Meredith (1993), and Meredith and Millsap (1992) provided a statistical definition of [Measure Invariance]. Namely, an observed score is said to be measurement invariant if a person’s probability of an observed score does not depend on his/her group membership, conditional on the true score. That is, respondents from different groups, but with the same true score, will have the same observed score.”** Stereotype threat (ST) is a form of psychometric bias. If strict measurement invariance (MI) holds, there is no psychometric bias and therefore no ST (at least in the samples tested). For a number of IQ standardization samples, which show an IQ gap of 15 points, strict MI has been found to hold. > [@](#): > > A more thorough study would correct for the stereotype threat by using a single blind administration of a test with standardised instructions. Ignoring the validity of the test measure and focusing only on its reliability, the degree to which the difference in outcomes for attainment in that measure were socioeconomic could be determined. However, this would not change the fact that no genetic component would be isolated, as there are biological factors which are not genetic. Which wasn’t accounted for in the study either. I don’t understand what you are saying here. And I don’t know what you mean by “the study.” I’m not talking about “a study.” I’m talking about hundreds of studies. And as I noted above you can statistically test for psychometric bias. Go read a book on psychometrics. And read by discussion. Environmental biological caused can not explain the Black-White gap in the US. > [@](#): > > This section in the wiki you linked stresses the necessity of experimentation in order to determine between causal hypotheses in order to control for confounding variables. All that I was saying is that there is evidence linking group differences – at least in the US, to genes. That’s all. My argument is very very simple: If you are going to call me [a racist asshole on the account of Black people in the US sucking](http://unfaircampaign.org/) than prove [my semi-supported simple to disprove](http://www.annualreviews.org/doi/abs/10.1146/annurev-genom-082509-141523) alternative hypothesis incorrect. No one wants to because they are afraid of results like [this](http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2651992/). ---
**Author:** ![Chuck11](https://avatars.discourse-cdn.com/v4/letter/c/96bed5/32.png) [@Chuck11](https://boards.straightdope.com/u/Chuck11)\ **Post date:** [June 23, 2012, 2:26am UTC](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236/687 "2012-06-23T02:26:49Z")
> [@gamerunknown](#): > > Forgot: > Your specific example contradicts your proposed finding. Explain. I said: Were we to treat Black people as a arbitrary sub-population, we would conclude that the difference was _probably_ more genetically conditioned than not. [http://occidentalascent.files.wordpress.com/2011/05/probability-iq.png?w=309&h=58](http://occidentalascent.files.wordpress.com/2011/05/probability-iq.png?w=309&h=58) This is a probabilistic statement and it does not contradict anything that I said. ---
**Author:** ![gamerunknown](https://avatars.discourse-cdn.com/v4/letter/g/6bbea6/32.png) [@gamerunknown](https://boards.straightdope.com/u/gamerunknown)\ **Post date:** [June 23, 2012, 8:40am UTC](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236/688 "2012-06-23T08:40:59Z")
> [@Chuck11](#): > > the post’s very comment section Hadn’t read it, sorry. > [@Chuck11](#): > > By at least some commonly used zoological metrics, my races are “significantly different.” Reading the comments on the article I linked, are those metrics skull and teeth size or shape? Do they hold consistently to the “five classical races” discussed? > [@Chuck11](#): > > In a meritocratic society, class is largely determined by genes. The number of unsupported premises you’re propounding are expanding. If you consider genetic class differences in IQ to be as valid as race differences in IQ, then you now need to provide support for the fact that we live in a meritocratic society and that class is determined by genes. > [@Chuck11](#): > > Environmental biological caused can not explain the Black-White gap in the US. Which doesn’t change the fact that we cannot determine a causative relationship from the specific study you linked. It did not isolate genes or gene clusters and measure their effects in isolation. It assumed the premise that black and white populations form coherent genetic groups and worked to build the conclusions from there. > [@Chuck11](#): > > I don’t know what you mean by "the study.” You may be discussing hundreds of studies, I’m discussing the methodology of the study you linked. ---
**Author:** ![Chuck11](https://avatars.discourse-cdn.com/v4/letter/c/96bed5/32.png) [@Chuck11](https://boards.straightdope.com/u/Chuck11)\ **Post date:** [June 23, 2012, 8:11pm UTC](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236/689 "2012-06-23T20:11:10Z")
> [@gamerunknown](#): > > Hadn’t read it, sorry.Reading the comments on the article I linked, are those metrics skull and teeth size or shape? Do they hold consistently to the “five classical races” discussed? Here is an exposition of the race definition that I’m working with: “The keys here are (1) subspecies are populations, geographic segments of a species, not morphs co-occurring with other variants, and (2) they differ from each other on average, not absolutely. The so-called 75% rule, which I have used above, is only a rule-of-thumb, but it becomes rather meaningless to single out populations in which much less than this proportion is distinctive. Unlike species, subspecies have no whatness. They share genes with other subspecies of the same species, so their interrelationships are genetically reticulate. In some taxonomic schools of thought they have no place at all, though it seems to me that it is useful to focus on populations that differ as whole but not absolutely. Subspecies should not be reified: they are simply the point along the continuum of population differentiation, from identity to species, at which it becomes worthwhile to give them a scientific name” By a common interpretation of the 75% rule, the classic 5 races qualify as races. That’s all that I’m saying. (There are other interpretations of the 75% rule in which they don’t.) Are the differences that I cite consistent? Some are, but they don’t have to be – per zoological standards. (People don’t seem to get that the standards for classifying zoological subspecies are currently rather lax. ) For example, hair morphological differences are not consistent in the sense that my Australoids do not differ from my Negroids in this trait. But my Negroids differ from my Mongoloids. At any rate this is irrelevant to my point, since I’m rather critical of what I call “global race realism.” I discussed this in my prior comments. You said: > [@](#): > > If you consider genetic class differences in IQ to be as valid as race differences in IQ, then you now need to provide support for the fact that we live in a meritocratic society and that class is determined by genes. Do you agree that within racial populations, IQ has a non-trivial heritability? And do you agree that IQ is correlated with SES? And do you agree that there is social mobility in meritocratic societies such that people with a higher IQ move up in SES and people with a lower IQ move down? I rest my case. You said: > [@](#): > > Which doesn’t change the fact that we cannot determine a causative relationship from the specific study you linked. It did not isolate genes or gene clusters and measure their effects in isolation. You can \*determine \* a conditional causative relation using SEM. To note, this same methodology has been used to determine if between sex differences in Blood pressure are partially heritable. The condition is that you accept the SEM assumptions. If you have good reason to reject those assumptions in this case, let’s hear them. In my extended discussion that I linked to, I discussed this point extensively. As I knew some would object to the assumptions, I restricted my claim both in my posts here and in my extended discussion. I said that there is a link between the magnitude of the difference and genes, specifically within group heritability, and that this needs to be explained. My point is that this gives grounds for further investigation using more genetically informative methods such as admixture mapping. Would you not agree that my claims are pretty nuanced? You said: > [@](#): > > It assumed the premise that black and white populations form coherent genetic groups and worked to build the conclusions from there. I don’t know what you mean by “coherent genetic groups.” It should be obvious that groups that are not delineated genetically can differ genetically. Moreover it should be obvious that genetically heterogeneous groups can differ genetically _on average_. For example, “South Americans” on average are genetically predisposed towards being more darkly pigmented than “North Americans.” As for the SEM methodology, though, it makes no prior assumptions about genetic differences. It tests for them. The assumptions that it makes are explicit and clear. And I’ve discussed them already. > [@](#): > > You: You may be discussing hundreds of studies, I’m discussing the methodology of the study you linked. Here was what you originally said: > [@](#): > > It demonstrates differences between races while controlling for socioeconomic status, but does not address the possibility of systematic bias. A more thorough study would correct for the stereotype threat by using a single blind administration of a test with standardized instructions. Ignoring the validity of the test measure and focusing only on its reliability, the degree to which the difference in outcomes for attainment in that measure were socioeconomic could be determined. However, this would not change the fact that no genetic component would be isolated, as there are biological factors which are not genetic. Which wasn’t accounted for in the study either. The study that you linked to in your comment, NLSY79, was a study that I never discussed. I discussed a study by Rowe an Cleveland which used data from **NLSY97**. As I noted before, the Rowe and Cleveland study build off of a study, [Rowe et al.](https://lesacreduprintemps19.files.wordpress.com/2012/05/no-more-than-skin-deep-ethnic-and-racial-similarity-in-developmental-process.pdf) which explicitly tested for psychometric bias (though in a different way than discussed above). Read the original study first. If you keep misrepresenting my comments, I’m going to end this conversation. [Previous page](https://boards.straightdope.com/t/race-differences-in-average-iq-are-largely-genetic/623236.md?page=34)