A little bit of difference goes a long way

In response to my post about the UK Black-White difference (or lack thereof) – and its implications for race realism, I received a virtual laundry list of environmentalist explanations:

The negative effect of White UK culture:

The big thing to look into is the emergence of a gender gap over the last generation in Britain favoring teen girls in run-of-the-mill tests like the GSCE. It’s significantly bigger than in the U.S., and seems to hold across most ethnicities. It’s a smoking gun that something is going wrong with white British culture.

The positive effect of Black Mulatto culture:

Another thing to keep in mind is that while I don’t know about the UK, I know that in the U.S., a lot of West Indians are from the mulatto elite: Attorney General Eric Holder, General Colin Powell, singer Rhianna, journalist Malcolm Gladwell, etc.

Motivation:

I think a likely (partial?) explanation is that the Black students tended to be more motivated to get good grades. They probably took less challenging courses since we can see that the White advantage is much greater when English and Math are included (these are probably more rigorous subjects).

That the tests don’t measure intelligence:

The GCSE data in the Deary et al. study are from 2002. In 2002, about 47% of all students obtained fewer than 5 GCSEs at grade C or above, whereas in 2009 the proportion was only 25% (see graph 1). It’s obvious that there’s been lots of grade inflation

Low expectations for Whites:

So, my overall model of Britain’s social system v. America’s is that the former is dominated by class and the latter by race. The British system tends to be better for blacks because they have less pressure on them to Act Black all the time, while it’s worse for poorer whites because they are constantly pressured by their peers not to Act Toff by like, you know, reading.

And systematic scientific anti-racism and biased research:

The British establishment is so extreme in its anti-white position, and so desperate to degrade its own native population in favor of blacks and Asians, that we should not take any official figures from Britain seriously

Let’s put aside the fact that my original estimate of little difference now seems questionable (i.e., there possibly seems to be a UK Black-White difference of 0.5 or so standard deviations.) The first couple of responses were somewhat unbelievable, given what is typically argued by hereditarians: High within population heritability limits between population environmental influences. This is a staple hereditarian argument, which goes:

The heritability of IQ for both Blacks and Whites is high at older ages. For example, in an analysis of the nationally representative ADD health data, Guo and Stearns (2002) found respective Black and White adolescent heritabilities of .57 and .71 and respective between family environmentalities of .16 and .03; similarly high within population heritabilities have been found in other studies (Rushton and Jensen, 2005). In absence of measurement bias and unique factors which uniformly raise or lower one or the other population’s scores (i.e. X-factors), high within population heritabilities constrain between population environmental influences. . If, for example, two populations have between family within population environmentalities of 0.16 and strict measurement invariance holds, for environmental influences to create a 1 SD difference, 2.5 SDs of environmental influence are needed (i.e., 1/ sqrt(0.16)). Several studies have shown that, in the case of Blacks and Whites, there are no detectable X-factors (e.g. Rowe et al., 1994; Rowe et al., 1995; Carretta, 1995; Carretta and Ree, 1995) and several other studies have shown that differences between Blacks and Whites are not due to measurement bias (Dolan, 2000; Dolan and Hamaker, 2001; Lubke, et al 2003)…..

(See, Jensen, 1973; Jensen, 1998; Murray and Herrnstein, 1994; Levin, 1997; Sesardic, 2005…)

This applies to the presence of differences just as much as it does to the absence. Given a hereditarian model, and given the magnitude of differences typically proposed, the said environmental explanations for an absence are simply implausible.

There is, of course, an out. Flynn and Dickens (2001) have provided it. Flynn and Dickens’ social multipliers model – which is frequently cited by quasi-environmentalists — works just as well in defense of “race realism” as it does in defense of environmentalism, if one is willing to adopt what we might call “geneticism,” where this simply refers to the view that race differences are caused, directly and/or indirectly, by genes.

By the Flynn and Dickens model, small genotypic differences between individuals and/or populations (say, 0.2 SD) can multiply, by way of gene-environment covariance, to produce large phenotypic differences (Say, 1 SD). To the extent that this gene-environment covariance is active, in the sense of individuals selecting their own phenotypic shaping environments, the resulting differences can be said to be causally genetic, in an indirect sense. This model can be applied to numerous behavioral traits. Moreover, between population multiplication is not heavily contingent on within population multiplication. Now that Flynn and Dickens have established the a priori plausibility of this model, it can only be ruled out, as a mechanism of between population phenotypic differentiation, by testing populations and showing the absence of the prerequisite small genetic differences.

Geneticism, as so conceptualized, is not exclusive of hereditarianism. Hereditarianism is just a subtype of geneticism; it’s the view that race differences are largely caused by additive genetic differences (e.g., a 0.8 geneotypic SD difference to a 1 phenotypic SD difference.)

Adopting a geneticist view, expands greatly the scope of “race realism.” Doing so also changes the contours of the race-genes debate. Arguments which are potent against hereditarianism are impotent against geneticism. Malleability of between group differences; absences of differences in certain environments; unusually small differences between mixed race individuals – are all consistent with variants of geneticism that don’t posit large hereditarian differences. Such models make predictions, of course; they predict the probable occurrence of differences – as opposed to the necessary occurrence (in absence of implausibly large countervailing environmental factors) (i.e., across environmental regimes, difference of various magnitude, going in the same direction., will tend to emerge.)

The responses above, then, are not inherently unreasonable. The explanations offered are plausible given some forms of geneticism, particularly weakly hereditarian forms. I would guess, though, that these forms were not being entertained when the reposes were written.

Advertisements
This entry was posted in Uncategorized. Bookmark the permalink.

27 Responses to A little bit of difference goes a long way

  1. catperson says:

    As I explained in another thread, in my humble opinion, the explanation for the small B-W IQ gap in the UK (if it’s as small as you say) is almost certainly selective immigration. Keep in mind that having a university degree, particular in STEM field, in sub-Saharan Africa, is about as rare as an American having a degree from MIT. When you compare the children of African and West Indian super-elites to average white British reared in the same environment it’s not surprising that the racial gap will diminish or even reverse. We see the same pattern with Indian immigrants to the United States.

  2. catperson says:

    “By the Flynn and Dickens model, small genotypic differences between individuals and/or populations (say, 0.2 SD) can multiply, by way of gene-environment covariance, to produce large phenotypic differences (Say, 1 SD).”

    This doesn’t really make sense. If the differences between individuals multiply because of gene-environment correlation, then the standard deviation will also multiply, so the STANDARDIZED difference between people will remain 0.2.

    • Chuck says:

      There are multiple errors here. To illustrate just one — take the above example. Imagine two individuals, A and B, at time 1 who had IQs 15 and 18 points, respectively. below the population mean, were the SD was 15. The phenotypic difference between them here is 0.2 SD. Imagine at time 2, they had, respectively, IQs 3 and 18 points below the population mean. Assuming no other changes in the population, the population variance would have decreased– by a magnitude dependent on the population size. The resulting standardized difference would be over 1 SD.

      • Catperson says:

        That’s all very random. Your point was about a gene environment correlation. Let’s say we have a population with 3 people with IQ’s 115, 100, and 85. The mean IQ is 100 and the SD is 15 and the smartest person is 2 SD smarter than the dumbest person.

        Now let’s assume because of the gene-environment correlation, the smart person gets a PhD and raises his IQ (on a culture biased test) by 15 points and the dumbest person drops out of elementary school to join a gang which drops his IQ by 15 points. Now the IQ’s in are population are 130, 100 and 70. The gap between the smartest and dumbest has doubled, but you know what else has doubled? The standard deviation. So the difference between the smartest and dumbest remains 2 SD.

    • Chuck says:

      “then the standard deviation will also multiply.”

      So what you mean to say is “standard deviations COULD multiply.” I obviously am not referring to a situation in which they do. Try making sense of the statement — I’m sure you can manage.

      • Catperson says:

        Well you speak of a gene-environment correlation. The higher the correlation between genotype and environmental factors that shape phenotype, the higher the standard deviation of the phenotype.

  3. catperson says:

    To the extent that this gene-environment covariance is active, in the sense of individuals selecting their own phenotypic shaping environments, the resulting differences can be said to be causally genetic, in an indirect sense. This model can be applied to numerous behavioral traits.

    The Flynn-Dickens model can be applied to traits that can be influenced by phenotypic shaping environments such as muscles and academic achievement. For example someone who notices genetic potential for muscles or educational success can spend a lot of time in the gym or the library respectively and maximize these genetic gifts. However other traits like height and IQ are much less responsive to the post-natal environment and are largely predetermined, so while a tall person is free to do stretch exercises to maximize his height and a high IQ person is free to do brain teasers to maximize her IQ, such environment self-selection will have virtually no impact on height or IQ respectively.

    • Chuck says:

      “However other traits like height and IQ are much less responsive to the post-natal environment and are largely predetermined”

      Obviously, quasi-environmentalists don’t agree with you. Refer here:
      http://occidentalascent.wordpress.com/2012/01/10/the-bell-curve-wars-18-years-on/

      Generally, it’s a complex issue which I have discussed extensively elsewhere. Don’t assume you know more on this subject than me.

      • Catperson says:

        Obviously the social environmentalists don’t agree with me otherwise they wouldn’t be social environmentalists. But if you read “The g Factor” by Jensen he cites very persuasive evidence that cognitive abilities are wholly biological variables not amenable to any significant degree by the social, cultural or intellectual environments you choose or that are chosen for you. Adoption into a higher class environment has zero impact on IQ by later life. Intensive stimulation in early childhood has virtually no impact on IQ. People who practice repeating numbers from memory all day every day only improve their narrow ability to remember numbers but it does nothing to improve their ability to remember letters let alone improve their memory in general or IQ.

  4. What we need, and don’t have, is data on regression to the mean and on the relative size(s) of the right-tail(s) of the distribution(s). We have that data for the USA, and that data confirms the hereditarian hypothesis. In order to test the hypothesis that the smaller UK IQ gap is the result of immigrant selectivity, or other forms of unrepresentativeness, we need that data for the UK.

    As for the impact of environmental factors on the UK results, we were proposing that the GCSE results were insufficiently g-loaded and that the thresholds involved were too low. That is not the same sort of argument as put forward by the anti-hereditarians. They argue that g itself is environmentally determined, and even at the right-tail of the distribution.

    Immigrant selectivity is also relevant to the issue of motivation. If immigrants are selected for IQ and motivation, then it would be reasonable to expect that their children would have greater motivation than the general population.

    • Catperson says:

      What we need, and don’t have, is data on regression to the mean and on the relative size(s) of the right-tail(s) of the distribution(s). We have that data for the USA, and that data confirms the hereditarian hypothesis

      Can you elaborate? What data do we have in the U.S. that we don’t have for the U.K.

    • Chuck says:

      “As for the impact of environmental factors on the UK results, we were proposing that the GCSE results were insufficiently g-loaded and that the thresholds involved were too low.”

      Ok, but we’re still left with only a 0.5 SD g –FSIQ — gap based on age 11 2009/2010 CAT scores. So, assuming this difference is totally due to genes, the Black-African, White-European genetic gap would be this plus the phenotypic selectivity of the Black immigrants times 0.6 for regression. This is were the issue of immigrant selection comes in. I argue that African immigrants represent no more than the 70th to the 83rd or so cognitive percentile. So the absolute ceiling of the genetic gap is 0.8 SD to 1.1 SD. That’s the absolute ceiling. I would note that the UK Black-White gap seems to have been decreasing — from 0.9 SD as reported in Lynn (2008) to the present.– so we might infer that it will continue to decrease somewhat. And it might be the case that these estimates of selectivity are unduly high. Based on the educational qualifications of Black immigrants to the UK (circa 2001), we might reasonably conclude this

    • Chuck says:

      “As for the impact of environmental factors on the UK results, we were proposing..”

      As for that, I pointed to both standardized tests and grades. And in some of the data, test differences were presented separately. And the gaps there, which are comparable to NAEP gaps, were small. See here, specifically KS2: http://occidentalascent.files.wordpress.com/2012/02/screen-shot-2012-02-07-at-10-17-43-am.png In that same data, which came from the Longitudinal Study of Young People in England, grades were found to be highly correlated with g (Deary et al. 2006). So I didn’t take seriously claims that GCSE results were insufficiently g-loaded.

  5. Steve Sailer says:

    The obvious analogy is to the Indian diaspora. The Indian masses back home do badly on IQ and school achievement tests. The highly selected Indian community in the U.S. does extremely well. The less selected Indian communities in places where the British brought in Indian laborers, such as Mauritius and Trinidad and Fiji, do fair to middling.

    This suggests to me that if Indians back in India can get their act together, that overall cognitive and achievement scores will rise.

    • Catperson says:

      Correct. Indians born in America probably have a mean IQ of 110 even though India has a mean IQ of 81 and the explanation for this is extremely selective immigration of their parents combined with extremely superior first world nutrition enhancing brain development. If second generation Indian immigrants can score 2 SD above India’s average, then second generation sub-saharans can score 2 SD above sub-sahara’s average IQ of 67. You’re exactly right as usual Steve; it’s perfectly analogous.

      • Chuck says:

        “Correct. Indians born in America probably have a mean IQ of 110 even though India has a mean IQ of 81 and the explanation for this is extremely selective immigration of their parents combined with extremely superior first world nutrition enhancing brain development”

        See Catperson understands that given the high within population heritability of g, extreme differences are needed. Otherwise “such environment self-selection will have virtually no impact on height or IQ respectively” since “traits like height and IQ are much less responsive to the post-natal environment and are largely predetermined.” .

    • Chuck says:

      The problem one runs into is the whole elaborate within-between heritability argument. Refer to Jensen’s “The G-factor” or Sesardic’s “Making sense of heritability.” For what you say to be true, either there must be a MASSIVE environmental difference, there must be an implausible between population X factor, or the narrow heritability of g in India must be low and IQ must be weakly correlated with genes. If the latter is true for India, it’s likely true for Africa and a central race realist claim is violated.

      • Catperson says:

        There is a MASSIVE environmental difference between the first world and third world. The average black in Africa is 1 SD shorter than blacks born in America so the nutrition in Africa must be vastly inferior since the genes between Dark African Americans and west Africans are virtually identical. However within America, there’s virtually no height difference between blacks and whites implying no nutrition difference, implying the 1 SD IQ gap must be virtually 100% genetic since biological environment is equal.

        There’s also MASSIVE environmental differences between Americans today and Americans during world war I which is why IQ, height, and brain size have probably all increased over 1 SD since that time.

  6. Steve Sailer says:

    Why the strawman argument? I’ve assumed a roughly 50-50 divide between heredity and environment since the 1990s. For example, here’s my 1998 review of Jensen’s “The g Factor.”

    http://www.isteve.com/jensen.htm

    I helped Flynn polish the Flynn-Dickens theory a decade ago.

    • Catperson says:

      A 50-50 divide between heredity and environment? Perhaps on a global scale where you have vast differences in nutrition between the West and the third world, but certainly not within America, where 80% of the variation in later IQ is explained by genes, 20% is explained by the biological environment, and roughly 0% is explained by the social environment.

      In my humble opinion, the Flynn-Dickens model is not tenable with respect to IQ.

  7. Steve Sailer says:

    As I pointed out in my review of Lynn and Vanhanen’s IQ and the Wealth of Nations in 2002:

    It appears likely that some combination of malnutrition, disease, inbreeding, lack of education, lack of mental stimulation, lack of familiarity with abstract reasoning and so forth can keep people from reaching their genetic potential for IQ. Lynn himself did early studies demonstrating that malnutrition drives down IQ. The co-authors conclude their book by recommending that

    “The rich countries’ economic aid programs for the poor countries should be continued and some of these should be directed at attempting to increase the intelligence levels of the populations of the poorer countries by improvements in nutrition and the like.”

    A clear example of how a bad environment can hurt IQ can be seen in the IQ scores for sub-Saharan African countries. They average only around 70. In contrast, African-Americans average about 85. It appears unlikely that African-Americans’ white admixture can account for most of this 15-point gap because they are only around 17%-18% white on average, according to the latest genetic research. (Thus African-Americans white genes probably couldn’t account for more than 3 points of the gap between African-Americans and African-Africans.) This suggests that the harshness of life in Africa might be cutting ten points or more off African IQ scores.

    http://www.vdare.com/articles/a-few-thoughts-on-iq-and-the-wealth-of-nations

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s