2011 Grade 4 TIMSS/PIRLS Scores by Race/Ethnicty and Nativity

The good news is that 1st and 2nd Generation nominally White immigrants — Chechens, Moroccans, Afganis, etc. — are helping to narrow the gaps. FYI, the Math-Reading-Science composite native B/W gap came out to ~ 1.1 SD or ~ 0.4 SD larger than the composite native H/W gap. Second generation Black immigrants didn’t seem to perform much better than their third generation peers (in this most recent round of international testing). This suggests that the Black (and the White) pool of emigrants to the US is becoming more inclusive.



This entry was posted in Uncategorized. Bookmark the permalink.

27 Responses to 2011 Grade 4 TIMSS/PIRLS Scores by Race/Ethnicty and Nativity

  1. Buuuuuuuuuuuuugugugugugugugu says:

    You know the IQ scores in US staying so stable over time is actually a good thing for the environmentalists not for the HBD guys. For it to be environmental and cultural the gaps should not close gradually it has to close suddenly or in big bursts at a time and the gaps in achievement should close gradually with some erratic ups and downs. The gaps should also close slowly as the income increases.

    • Chuck says:

      I can’t decipher your meaning. First, an environmental hypothesis does not predict stability over time unless there is a stabilizing force. As I explained elsewhere, a simple environmental model predicts rapid narrowing:
      http://humanvarieties.org/2013/01/15/secular-changes-in-the-black-white-cognitive-ability-gap/ Second, this post specifically concerned immigrant differences — not the cause of these. Since Black and Hispanic immigrants continually under-perform White immigrants and White natives, even if the natives gaps were to narrow, the overall gaps would persist — because there would be a continual inflow of low IQ “colored people”. The implication is that, for the foreseeable future, the gaps will not close.

      • Buuuuuuuuuuuuugugugu says:

        There is a stabilizing force and its been there for ages. You guys just can’t see it. Whats the one thing that has stayed the same even though what it refers to has actually been changing?

        Obama being the first black president should tell you everything.

        • Chuck says:

          Don’t waste my time. Measure Invariance (MI) holds for the Black-White psychometric intelligence differential. This implies that the differential is not being conditioned by group specific factors. As for the meaning of MI: “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.” As such, the causes of the between race differential must be a subset of the causes of the within race differentials. Shared environmental differences are not inter-generationally stable. Non-shared environmental influences do not condition stability within populations so it is not obvious to me how they can condition stability between populations between generations. But I am open to this possibility — and I see no other barring genetics. What is it that you have in mind exactly (that is not a group specific factor like “racism“)?

          • Buuuuuuuuuuuuugugugu says:

            Don’t be rude guy.

            Blacks have less black in them now. The gap should have closed somewhat and done so gradually, but it hasn’t. Even people with one white parent will chose black to classify themselves in some cases.

            Hasn’t it increased at some stages too?

          • Chuck says:

            Increased miscegenation is leading to a narrowing of the differential. But “Black”, here, is being defined as having two Black parents. Individuals with one Black and one White parent perform intermediate to “monoracial” individuals. Read my review of the race IQ issue here.

          • Buuuuuuuuuugugugu says:

            You misunderstood. Those who are defined as black have less black in them now than before. The statistics you are getting of people who are black are less black now than before. There are even mixed people who would tick black if asked what race he/she is in those stats with self identification.

            What I am trying to say is the black white gap should have closed somewhat. Not talking about biracial… I am talking about what black is now and what was then, which less black.

          • Buuuuuuuuuugugugu says:

            Thats fine, more white people in the family lines means the environments of the entire families of said blacks will be on average better.

            Still my point that blacks have plenty white in them in US and should htey should have more now than before. Thus that national IQ gap between blacks and whites should have been closing bit by bit, but it hasn’t.

            I hope I’m clear enough.

  2. Chuck says:

    “Thats fine, more white people in the family lines …”

    Well, of course, that would be the environmental reply — but we just discussed how shared environmental influences are weakly inter-generationally transmitted. The argument here is multi-level — try to follow: If you wish to account for the mixed race data by positing shared environmental effects, then you must maintain that the between race difference is due to shared environment. If you maintain this, then you need to explain the large between race differential despite the relatively weak correlation between shared environment and IQ within races, given Measurement Invariance.

    • Buuuuuuuuuugugugu says:

      “but we just discussed how shared environmental influences are weakly inter-generationally transmitted”.

      Uh no it not weak at all. Literally everything effecting your environment can be passed on through generations. Its family the closest thing to a person, they provide the place you grow up in, the stories you are told, the toys you play with, the books you read, even have an influence on the friends you meet,. Family and family history influences somebodies environment the most… On average.

      Still the mixed race part is not what I was talking about.

      The main part of what I am trying to say is that the black population that keeps being tested with IQ is LESS black than before BUT the IQ gap has not changed. It has even increased at some points. This is not what should happen if it was genetic instead It should close gradually, but that has NOT happened.

      • Chuck says:

        First, by adulthood, less than 20% of the within race variance in general mental ability is explained by shared environmental influences. Read up. After refer back here. This is an inescapable point.

        Second, with regards to your point: “The main part of what I am trying to say is that the black population that keeps being tested with IQ is LESS black than before BUT the IQ gap has not changed. It has even increased at some points. This is not what should happen if it was genetic instead It should close gradually, but that has NOT happened.”

        The “multi-generational” Black population (i.e., two Black parents) represents the subset of the Black population that doesn’t marry out. This portion of the greater Black population (i.e., all people the identify as being Black) is about 20% European. The percent of European = White Admixture in this portion would only decrease if individuals who tended to marry out tended to be more admixed in the first place (e.g., 35% European) and if individuals who had one White parent tended not to marry in. I see no evidence that either condition is true.

        Regardless, you are making too much of small differences. The “pure” West African “pure” European (i.e., West Eurasian) difference in genotypic mental ability can be no more than 1.2 standard deviation — i.e., about the difference between random people within any given population. I graphed the phenotypic differences that would be produced by this and other genotypic differences as a function of within population heritability coefficients (i.e., the amount of genetic influence on tests within population). The heritability of the tests that we are talking about range from 0.4 to 0.8, age depending. The tests mentioned above — for Grade 4 — probably have a heritability of at most about 0.5. So the magnitude of the difference should — were no environmental factors involved at most be 0.75 to 1.05 SD, and for the tests that we are talking about –above — maybe 0.85 SD between “pure” European and “pure” Black West Africans — or at least pure descendents of the founding stock of African Americans (who weren’t necessarily representative of the Black West African population, given the nature of the migration history). Now African Americans, as we said are 20% or so European, so we should expect maybe a difference of 0.85 SD* (1-0.2) = ~ 0.7 SD. This can be compared to 1.1 SD discussed above. So it’s already granted that at least one third of the difference is due to environmental factors (0.7/1.1) = 65%. So there is some variability here. So the trivial differences that you are talking about really would be impossible to detect. (A 10% shift in ancestry would only be associated with a ~ 0.07 SD difference — completely unnoticeable.)

      • Chuck says:

        Generally, your argument is unimpressive. Search through my blog and you will find truly devastating environmental counters, such as: http://occidentalascent.wordpress.com/2012/02/06/is-global-race-realism-still-tenable/ Even the most elite racial hereditarians have been unable to come up with a reply to some of these. Have I? Maybe, maybe not.

        • Buuuuukugugugu says:

          I’ve read everything from a while back already.

          “Have I? Maybe, maybe not.” Have you devastatingly countered the hereditarians?

          I think you have countered them somewhat but not with a killer blow…yet.

          [With regards to the UK, I have been unable to acquire more scores — but I am still investigating the issue. I have argued, elsewhere, based on migrant differences within the US, that inter national migrants are slightly selected. If so, and granting an about 0.5 SD UK difference and taking into account the point about phenotype, genotype, and heritability, I feel that the UK results can be reconciled with my claimed US B/W genotypic g differential of 1 SD.]


          “First, by adulthood, less than 20% of the within race variance in general mental ability is explained by shared environmental influences. Read up. After refer back here. This is an inescapable point.”

          Its not so inescapable.
          Twins? Identical twins are not so identical as previously believed and they weren’t really reared in different environments in any of the studies. Also you are comparing MZ twins with DZ twins. DZ twins are like normal siblings and will be treated as such. They as individuals will think of themselves as such. MZ twins on the other hand won’t, they are treated different from birth by parents and everyone else. You really think when they were reared apart they never saw each other? How many times do you think they hung out together, played together, talked to each other? As opposed to DZ twins?

          [If you persist in denying the fact of low shared environmentality, I will have to end this discussion. The estimate I used in my argument was an upper bounds estimate (i.e., c^2 = 20%). Based parent-adoptee correlations, the estimate is 0 by adulthood. Based on Unrelated Siblings Reared Together and Virtual Twin correlations, the estimate is about 0.2 in childhood-adolescence and it decreases with age. Adoptive parent – adoptee and Unrelated Sibling correlations are direct measures of the influence of shared environment. Estimates based on MZ twins reared apart, MZ-DZ twin differences, full-half sibling differences and other exotic relations corroborate this c^2 estimate of </= 0.2. Moreover, GCTA estimates, based on the genetic similarity between random individuals, confirms this. GCTA shows a narrow heritability of about 0.45-0.5 (at older ages). The remainder of the differences is split between (a) shared environment (b) unshared environment (c) non-additive genetic effect (d) additive genetic effect not tagged and (d) measurement error. In short, there simply is no room for an increased influence of shared environment relative to what is found based on traditional kinship studies. As for the MZ look alike hypothesis, this has already been tested. So this situation won’t due – if you can’t even agree on the upper bounds c^2 estimate of 0.2, then, barring some spectacular argument, I can only conclude that you are being anti-empirical.]


          No man calm down, Jesus Christ. Thats not what I said you got it the other way around.

          What I am saying is that genes changed and the SD was stable decade to decade.

          I never said it was 100% environmental. I never said it was 100% anything.

          So what you are saying is that its too much of a small difference with white admixture. I however don’t mean only white admixture, I mean all other admixture too. More non black is in the black than before and should be more than everyone likes to think.

          Again I am not 100% on the environmental side. All I am saying is that its less genetic with IQ because of the SD being so stable. I’m saying that the the admixture in the black is bigger than you think.

          [We aren’t getting anywhere here. Give me some specific numbers. The time range is 1920 to 2010 since this is the period for which we have IQ data. So the birth years would be about 1905 to 1995, age depending. Give your estimated African/European/Amerindian/ admixture per generation. The Black generation span was about 22 years averaged across last century. So we have 4-5 generations, age depending. I want to see 4-5 estimates of mean admixture for “Blacks” with 2 Black Parents. We can then check your estimates with estimates derived from genetic analyses. And we can then estimate the upper bounds effect that the change in admixture would have given a 1 SD differences in genotypic g.]

          • Chuck says:

            Replied above.

            Generally, my diagnosis is that you are going amiss because (a) you are not looking at the full body of evidence simultaneously and (b) you are not quantifying your claims. A case in point was with your rejection of the c^2 estimate.

            From now on, try to quantify your claims.

            (Do you notice that my replies are very number filled? And that I spell out my arguments in almost syllogistic form? This is why my arguments convince me, but your arguments don’t.)

          • Buuuuuuukugugug says:

            Ok to make you happy I will agree on the c^2 estimate somewhat. Though only at high or average SES, within the same populations and cultures. I still think even that is higher than 20% though, but I will accept it for now.

            Also LOL I saw the first look alike study before you replaced it. Still though the new one doesn’t convince me of much. Mainly because its done by self report by teenagers who I doubt really know or care about the difference between MZ and DZ in the first place. Sample size is small too.

            It doesn’t really matter that much for what am trying to show you anyway which is about the black admixture and the black IQ score.

            The one drop rule counted for blacks in the majority of the time period in which the data was taken. It counted most before and less only very recently. Before 1960 it was most in effect and after 1960 too. Only recently, perhaps after about 2005 that biracials began separating themselves in classification.

            then on top of that there is further admixture that has increased over time in African Americans too.

            Thus the IQ should not have been so stable. Even though the genetic make up and even the classifications themselves have not been stable.

            Just saying its more cultural than you think in America.

          • Chuck says:

            You are overestimating the magnitude of the shift in admixture — and so the possible effect that this could have. Do the math and you will see this.

          • Buuuuuuukugugu says:

            Its not just counting the admixture. Its the definition of black and how that has changed… and that the admixture is not only from Europeans and the rate of mixing in present day.

            Either way you could be right since it could be small enough not to show. I will bet it isn’t though.

            So what do you think the actual average gap in points is, between the traditional black Americans and white Americans once the environment is accounted for?

            Mine as a guesstimate is about 10 maybe 6.


            Anyway this is my last post especially with race. Its actually nonsensical the whole thing. More dumb people in one place as opposed to another has got nothing to do with race. Technically there are no races between humans. All genes are transferable and if blacks wanted they can just bang the clever instead of dumb. Black people can literally give birth to a white and vice versa. Heck you can even get black and white twins too. Seriously check if you don’t believe me.

            Leave a reply with your point gap estimate I would like to see it.


  3. Chuck says:

    “You misunderstood. Those who are defined as black have less black in them now than before. ”

    On this topic, I misunderstand nothing. Within the last 20 years there has been substantially increased miscegenation, As such, the differential is reduced, somewhat, when you look at all “Blacks” under 20 (defined as a person with at least one Black parent — where “Black parent” is defined as a person with at least on Black parent). (At the current rate of miscegenation, African Americans are becoming about 10% Whiter every generation).

    But many of the studies – for example NAEP, TIMSS, PIRLS — now distinguish between “Blacks” with two biological Black parents and “Blacks” with one biological Black parent. These studies — which separate “multi-generational” Blacks from “mixed race” Blacks show less narrowing — because, of course, they exclude “mixed race” Blacks — who perform superior to “multi-generational Blacks” . Now, as for “multi-generational” Blacks, the research clearly shows that the average admixture is about 20% with a standard deviation of about 15% (so, of course, “multi-generational” Blacks are really “mixed race” Blacks in the sense that they have some White ancestors — they just don’t have one White parent). Now, the research clearly indicates that most of the admixture was introduced pre-1860. See, for example: Jin, et al. (2012). Genome-wide detection of natural selection in African Americans pre-and post-admixture. Genome research, 22(3), 519-527.

    • Buuuuuuuuuugugugu says:

      The black gap, the ones with two black parents. Their gap has not closed. That is what I mean.

      Admixture being added most in 1860 doesn’t mean they have stopped getting whiter, and more of everything else other than black. There was an increase in the gap for some time too wasn’t there?

      See thats what I mean. Such things should not happen, you should instead see a gradual decrease over time in IQ if it was genetic. A gradual one. Not a sudden one with kids in one generation.

      I believe in UK there was closing in the gap recently?

      • Chuck says:

        You said: “The black gap, the ones with two black parents. Their gap has not narrowed. … Admixture being added most in 1860 doesn’t mean they have stopped getting whiter, and more of everything else other than black.”

        Between the 1860s and 1980s the European Admixture in the Black population increased trivially. Maybe 5%. The effect of this would be virtually undetectable. Especially because there are other factors involved (e.g., dysgenic breeding (i.e., low IQ blacks having more kids than high IQ backs) and huge changes in the environment. You’re being silly about this admixture issue.

        You said: “There was an increase in the gap for some time too wasn’t there?”

        As for my review, there is inherent imprecision in meta-analyses of these types. A good summary of the difference is 1 +/- 0.2 SD, decade after decade. One can’t get more precise because samples differed in test taken, age, representativeness, and so on. Again, you are making too much of small differences. Go look up the concept of measurement error.

        You said: “I believe in UK there was closing in the gap recently?”

        But that was none other than me who brought this small inconsistency to light. So that doesn’t count as a very good argument.

      • Chuck says:

        Your ridiculous argument is: “Well, Chuck, the gap varied by 0.2 SD decade to decade …and genes didn’t .. so the whole 1 SD gap must be due to environmental factors.” That’s inane. Don’t waste my time with silly arguments.

  4. Chuck says:

    As for the magnitude of the genetic g gap, I told you my estimate: 1 standard deviation (or 15 IQ equivalent points). This would produce a 0.7 SD gap (or a gap of about 10 IQ equivalent points on a typical test because tests do not perfectly index genes). The standard deviation of IQ is usually set to 15 points; in terms of height that’s equivalent to 2.5 inches in industrialized countries. So, I am arguing that the current phenotypic differential is about 2/3rds nature. This is more or less consistent with the biometric estimates which are considered to be genetically informative in other instances.

    “Its not just counting the admixture. Its the definition of black and how that has changed…”

    I think that my 2/3rds estimate is consistent with this. It’s at least not inconsistent with it.

    You said: “Mine as a guesstimate is about 10 maybe 6.”

    So you think 1/2 to 2/3rds and I think 2/3rds.

    You said: “Anyway this is my last post especially with race. Its actually nonsensical the whole thing. More dumb people in one place as opposed to another has got nothing to do with race. Technically there are no races between humans. All genes are transferable and if blacks wanted they can just bang the clever instead of dumb. Black people can literally give birth to a white and vice versa. Heck you can even get black and white twins too. Seriously check if you don’t believe me.”

    This is a really irritating comment. IQ differences largely explain outcome differences between races in the US (differences which are not super large in the first place). For some reason many people make a huge deal about the outcome differences. When you hear discussions on “discrimination” or “racism”, that’s what’s usually being discussed. This topic is even number 4 on the social science list of fundamental and transformational issues. The issue then tuns back to the cause of the IQ differences. That is: The reason nature/nurture is important is because it explains IQ differences; the reason IQ differences are important is because they explain outcome differences; so if anything is silly, it must be the concern with outcome differences between visibly different groups. Is that silly? Well, ok, but if so, then we live in a silly world.

    As for your comment: “Technically there are no races between humans.” Here is a typical discussion of race in the sense of subspecies in the context of primates: “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 (Groves, 2002. The What, Why and How of Primate Taxonomy).
    What part of the definition is inconsistent with the idea of classic Human continental races?

    • Buuuukugugu says:

      EH ok you sure want to go there with me? I didn’t want to reply but I have to now.

      First and MOST IMPORTANT:
      The problem is the word race itself. It denotes species and the history of what that word meant and still means denotes something very different to one another. Thats why you use Human race and Klingon race, Elf race, Orc race or an alien race. It does not work within humans and should not be used because of what it means to humans.

      Call them African Americans, Sub Saharan Africans, then divide those up into Yoruba, Sadist, Gay, Tall, Short and whatever else you want. Those can fit into sub species too by the definition you produced.

      The problem is that you are trying to divide and label what is too heavily interwoven(increasingly so). You are dealing with humans not fucking guppies. Some nerds sitting around dividing peoples families, friends and even biological brothers and sisters into different “races” is not going to go down well, it never will.

      There is no need for it at all. Take the Black White IQ gap for instance its the same thing as the gap between a group of low IQ whites and high IQ whites. Like with Irish travellers and British. It can be genetic but its naturally changeable just by having sex even if that gap is 100% genetic.

      It is illogical to divide humans into different races because of what race means to humans and illogical to divide them into sub species because its too simplistic and unstable. The genetics of a group changes way too fast. Whats black here is not black somewhere else and that also changes rapidly.

      Homo sapiens sapiens are the sub species in human taxonomy. Everyone is Homo sapien sapien. Now what race does try to divide that into another set of sub species based on very rapidly changing variables.

      You are trying to divide peas in a pot with a chopstick. Its a waste of time.

      • Chuck says:

        I am glad that we agree that the between race differential can be partially genetically conditioned irrespective of whether the populations under consideration represent valid nature science classes. This is a trivially obvious point, yet many people are incapable of comprehending it — even after I painstakingly explain and mathematically demonstrate that the difference between any two arbitrarily picked socially constructed groups within a population is likely to be partially genetic since an individual’s deviation from the mean is more likely than not to be more genetically conditioned than not, given the heritability of the trait under questions. That is, it’s only by treating “Blacks” and “Whites” as specially different groups relative to arbitrary sub-populations within races that one can make the case that the differences are not like those between arbitrary sub-populations (e.g., social classes) within populations — which are more likely than not to be partially genetically conditioned. Most people have the logic reversed. They ague: “races” are arbitrary “social constructs”, therefore race differences can’t be genetically conditioned — when, in fact, it is the non-arbitrary non-“socially constructed” aspect of race that makes a racial genetic hypothesis for a difference less straightforward — because this requires one to explain how those difference came about in the first place — One can not just point out the normal genetic variance within a population and say: “Well, Blacks are just less intelligent White people with a suntan” and “The difference between more intelligent and less intelligent Whites tends to be substantially genetically conditioned.”

        So, let us be clear. what we are now discussing is the meaning and the validity of the race concept in context to human sub populations. And: this topic is only tangentially related to the race-genes-IQ issue.

      • Chuck says:

        You said: “The problem is the word race itself. It denotes species”

        I would agree that there is currently only one species in the Homo genus. I, nonetheless, disagree with your claim that “It is illogical to divide contemporaneous humans into different races.” In fact, I would argue that it is consistent to, given how many other species are divided into races. Before discussing this topic further, we must first specify the race concept under consideration. For example, one race concept, in the natural sciences, is simply “breeding populations.” It seems obvious to me that there are such human populations. These might not cohere with socially defined human races but they “exist” — where that means that there are human populations which can validly be called “breeding populations” given the criteria which defines what it is to be a breeding population. To start, would you agree or disagree with the above?

        Basically, I disagree with your prior claim that “Human races don’t exist” — for one, because there is no one set concept of Human race. And there surely are race concepts by which multiple human races exist e.g., ecotypes, breeding populations, etc. You seem to reify the term “race”. This term is a word which describes multiple concepts. We need to specify which concept it is that we wish to discuss.

        So, state the specific race concept that you wish to discuss and it’s qualifying criteria and why you don’t think that contemporaneous human populations qualify as races according to the criteria of this specific concept.

        • Buuuuuuukugugu says:

          You can’t even call blacks black if you can call other supposed races what they are because SSA are more distant in genes within their group than other “races” are to each other and even that won’t stay like that for too long.

          [Again, you didn’t specify a race concept. For example, by the geographic race concept the point which you are making is irrelevant. Why is it so difficult for you to specify the concept which you are working with?]

          Either way man, this is a waste of your time. I went through the whole the thing, trust me. I went through everything trying to quantify this and that and what did I gain, what did anyone that supported me gain?”

          [I like to deal with one issue at a time. This is how I make progress. Either withdraw your claim, concede my point, or cease discussion.]

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