Survey of Expert Opinion on Intelligence: Causes of International Differences in Cognitive Ability

So what. Seriously - do you even know what you are trying to argue? You need to define your races as whole-genome (neutral for anything sociological) clusters. Nobody has done that. The difficulty would be great, as illustrated by your link - imagine somebody using that link to define a bunch of human "races": you'd have a dozen different races, and a situation in which the Japanese and Malaysians are in one race; the French and Ethiopians in another race; the tall Maasai and the pigmy Mbuti in the same genetic race (or the Maasai in their own); the Melanesians in a completely different race from the Papuans, the Andaman Islanders, and the Australian aborigines (and none of them US black); and so forth.

In which case the US standard "black" race - this sociological construct you are trying to assign genetically lower cognitive abilities to - genetically falls apart. Fails to exist as a genetic cluster. And the US standard "white" one, now defined as a genetic cluster, includes Obama's father - a non-Bantu Kenyan (Ethiopian cluster, same race as the French and the Finns).

Can't argue with science, right?

The whole genotype is not used - a few preselected features are used. The selection process is not random, but purposeful.

And the populations are not labeled by race. The makers of that graph are not stupid.

And the visible genetic clusters do not match the US standard sociological racial categories. Nor do they match anything a biologist would presume based on morphology - the Maasai and the Mbuti pigmies are not going to end up in the same biological race under very many taxonomical systems that split anyone else. https://en.wikipedia.org/wiki/Mbuti_people#/media/File:Bambuti.jpg
http://i2.cdn.turner.com/cnnnext/da...inning-team-2012-horizontal-large-gallery.jpg

But familial inheritance does - you have lost track of your own damn argument.

Well no, it probably can't, but that isn't the point. The point is, it hasn't been. Your racial IQ studies did not control for childhood lead exposure.

Minor races and major "race groups" for everybody, but all the blacks are in the same one. Wtf?

Give it up. You're never going to make sense of this racial genetic cognitive ability bs, because it doesn't make any sense. It's a complete waste of time. You're making the rest of us stupider just trying to think about it.

 

Log in or Sign up to hide all adverts.

Yes, I was responding to your claim that the article in Nature establishing the heritability of height was only conducted at the family level, which was false. That's all. Was your claim relevant anyway: not really. You were the one that introduced "no genetic differences in height". I just thought it was funny so went with it. Feel free to drop it.

It's done in the plot I linked to. How can you look at a plot constructed by genomic similarity and say there is no plot?

Yes, the Japanese and Malaysians are in the same major race.

Please Register or Log in to view the hidden image!

Final paragraph:

To unambiguously infer population histories represents a considerable challenge…Although this study does not disprove a two-wave model of migration, the evidence from our autosomal data and the accompanying simulation studies…point toward a history that unites the Negrito and non-Negrito populations of Southeast and East Asia via a single primary wave of entry of humans into the continent.​

Mapping Human Genetic Diversity in Asia.

Maybe you are learning something!

Ethiopians clearly fall between the Caucasoid and Negroid clusters, and are mixed. Are we looking at the same plot? Any number of studies confirm this. A good paper on East African genomes was in Nature last year, check it out. Melanesians are in the same major race area as Papuans, again, do we have the same eyeball function? Where are Andaman Islanders and Aborigines on the plot? None of these are Negroids, they are Australoids or archaic Mongloids, also called Negrito.

I'm obviously not applying the US legal definition. Please try to argue against your opponent, rather than your imagination.

Nonsense. They are using around a million random SNPs at this point. It's a totally reliable sample.

They are labelled by their ethnic classification, and they cluster into races. Funny how the genetic distribution matches Blumenbach's taxonomy.

Yes, we know the US classification does not match the biological one.

In fact the genetic clusters match the Blumenbach partition based on ancestry inferred from skull morphology. Height and limb proportions are not ancestry informative since they are locally adapated and correlate to latitude. I guess you didn't know that.

Why are you taking a point from the top of my post, where it was responding to your heritability of height nonsense, and copying it to the bottom? It seems like you are deliberately trying to confuse things, perhaps because you are finding it difficult to respond to points ingenuously.

Race does not predict language better than it predicts IQ. Your assertion is false and your response about familial inheritance makes no sense. I don't have an argument at this point, I am simply responding to a stream of nonsense.

Yes they did, various studies have eliminated it as a factor.

No, Blacks have minor races too. You are responding to your imaginary opponent again.

You seem to be trying to confuse things. Maybe you really are just confused. I am not confused at all.

 

Log in or Sign up to hide all adverts.

No, it was not false. And it was not quite as you misrepresent it. My point was that your study linked had nothing to do with the heritability of population height at the cultural, racial, or national level, as Komlos analyzed. You were attempting to argue from familial heritability within a population to comparative population heritability between populations not defined by common familial inheritance, direct familial ancestry.

That is invalid. Its accuracy would be presumptively - barring discovery of mechanism - coincidental. Komlos et al has cast serious doubt on its accuracy, and established conditions for any proposed mechanism.

You asserted that strength of familial inheritance was evidence of genetic influence, and pointed to height as an example. I pointed out that religion and language were more strongly correlated with familial heritage than IQ (or height, actually), and height was a problematic example for you since it did not scale up to the population level genetically, as your racial bs requires.

There is a plot. It is not a sociologically neutral whole genome clustering plot that exhibits human races. It is not used by anyone to define human races, and if used in that way it generates racial classifications radically different from any of the sociological ones extant (and taxonomically near useless as well). It is not what you need, and it does not generate the clusters you require for your "racial" IQ bs.

There is no US legal definition. There is no genetic definition. There is a US sociological definition, and you are using it. (You are not, for example, using the Brazilian sociological definition in which Michael Jordan, say, would be "white" in many contexts).

They are together in a tight cluster, in your graph. So that is your "major race" cluster distance? That will hand you many dozens of "minor races".

Sociologically, they are different races entirely. And they differ significantly, as sociological populations, in measured IQ.

So you now recognize that "black" is not a racial category of any use to you - even for inferring genetic IQ variance between "races", let alone cognitive ability. You do see that, right? The "black" race is not even a cluster in your graph there, while the French and the Ethiopians, Japanese and Malaysians, are closely assembled.

As you recognize:

So do the Maasai. The biggest gap - the natural disjunction, objectively - is midway between different populations of Ethiopians - upper group goes with the French, lower group with the Maasai and south into Africa. Or you could call the Ethiopians and Maasai two clusters of their own, two "major races" separate from the (your term, bile rising) "Negroid" with the tall Bantu and all four disparate and otherwise genetically wide-ranging pigmy Bushmen populations (a nonsense cluster, from a taxonomic pov).

They are all "black", US standard racial classification. (As are a variety of other little dots in that graph, nowhere near the Bantu). And if you try to draw one of those arbitrary little circles around the lot of them, you will find yourself excluding the French and so forth only by sociological presupposition (Which is exactly how the little boundaries we see drawn in your examples were chosen - they do not reliably correspond to actual distance or connectivity measures visible on the graph itself)

No distance or connectivity measure on your graph there will hand you the US sociological races. US black is not a genetically definable race.

Ha! - I was right about the motive: I guessed they were doing a dispersal and migration analysis, just by looking at the clusters they generated with their criteria.

None of your IQ studies controlled for childhood lead exposure - or any other such environmental factor. Deal with it.

This is now officially comedy. Do you recall your response to my earlier posts noting that if whites were a "race" there had to be several "black" races?

Moving along then: So noting the IQ difference between Japanese and Malaysians, French and Ethiopians, and so forth, you are abandoning the attempt to assign genetic IQ levels to the "major race" of "blacks" - recognizing that since it includes the French or excludes the Ethiopians and Melanesians, it is not a valid genetic population for the purpose.

 

Log in or Sign up to hide all adverts.

Taking one point at a time, I would invite board members to look at this plot, and say whether they agree with the poster above that:

1) the French and Ethiopians are "closely assembled"
2) that SS Africans (Negroids or Blacks) excluding Ethiopians (mixed) don't form a cluster (ie. a non arbitrary-spectral-division) vis a vis the major cluster at the top (Caucasoids)

To make it more clear:

Please Register or Log in to view the hidden image!

 

lol, WTF?

 

Clarity!

Your nonsense circles - arbitrarily drawn on apparently sociological grounds - feature exactly my point:

you have a couple of Ethiopian dots in with the French cluster dots (and no visible reason for excluding the rest of them),

a large number of US standard "blacks" a very long way away from your circled Bantu, but a completely arbitrary inclusion of the Maasai ( on visual grounds, which who knows the meaning of since the scale is chosen on unrelated criteria, they form a cluster all their own) in with the several Pigmy tribes (whom we know to be genetically disparate themselves),

a "Caucasoid" cluster that includes some of southern India's darker skinned folks but not all of them (a wise choice for a racial agenda) and is drawn on no visible distance or connectivity criteria whatsoever,

all in all a perfect demonstration of what happens when somebody tries to extract human "races" out of genetic surveys like that. What you see is their sociological presumptions. The next step for this guy will be trying to explain the lack of correlation of IQ with his eventual full panoply of arbitrary circles - because he's not going to be able to draw boundaries around his IQ scores without making nonsense of his races, and vice versa. Not even using deceptive "average" IQ scores.

btw: I have never asserted that the genetic cluster with the Pigmies in it does not exist on that graph, or that there is no genetic cluster with the French in it, or anything like that. I just said that such clusters would not align with the US sociological races, and they would not support the concept of an average racial (or national) IQ one could compare with other races and nations.

Further, watching you circle the Maasai in with the Mbuti, and lump the different Bantu in with all of the disparate Pigmy tribes, and Lord knows what else if you continued circling, it's apparent this graph is not much good for taxonomy purposes even. It's apparently a residence time/migration/interbreeding map.

In other words:

You chose to exclude the Ethiopians and the Melanesians, rather than include the French. So you have no genetically defined "black" race.

 

I notice Bells has closed the other thread on IQ, and included slander about me, claiming I denied what I said in my OP. In fact I denied the relationship of race to IQ being my only topic, and was in the middle of discussing whether IQ subtests tended to correlate, with someone who has points other than chanting "racist". I refused to engage on the "sociological" race point in that thread:

which iceaura was trying to bring up, since it is being discussed in this thread. Bells then pretended I was refusing to address a different claim, that what it means to be intelligent is similar across cultures, and closed the thread. It's important to take one point at a time, preferably in separate threads. The validity of IQ is one of the central points. Bells accuses me of doing "everything but discuss this thread's topic as per the OP", notably ignoring the people chanting "racist", while I was in the middle of discussing the point of subtest correlations, and had not been challenged on cultural differences in intelligence in that thread. This point is raised by Bells for the first time in that thread, I am accused of failing to address it, and the thread is immediately closed. Of course I would have been happy to discuss your POV cherry picked Sternberg references.

In short your moderator Bells has a minority POV, and has abused his/her mod powers to push that POV.

I request the thread is reopened, where I will address your three references to Sternberg.

 

Sure, you just wanted to make a point about races and then let it go unchallenged. You wanted to present, as fact, that IQ was a good measure of intelligence because, you claimed, different races demonstrated different levels of intelligence and IQ tracked this. I suppose a racist would have defended these racial differences, but you were just interested in taking them as a fact.

I'm sorry that people won't let you just say the same thing that racists do and then silently go about their business.

 

I've heard of quote mining but sentence mining? Here's the full sentence again, for the hard of reading.

What does it say about your scientific position when you have to resort to tactics like this?

 

What does it have to say about your position that you simply repeat a racist claim, then try to stop anyone from discussing the merits of that claim? What does it say about your position when I point this out and you try to weasel out of this by trying the ridiculous tactic of trying to claim I'm quote-mining you? Your position does not read any better when you add in the rest of your sentence. There is no apology needed for cutting off the rest of your attempt to deny the facts of the other thread.

 

The circles are drawn roughly around vertice and endpoint clusters to point out what I am talking about. Obviously I did not run a PCA centroid algorithm on them.

Explain what you mean by "the French cluster"? I think this may be the source of the confusion. Do you mean the cluster labelled "French"? In which case your assertion is false. Or do you mean the Caucasoid cluster, which has French in it, in which case yes some of the mixed Ethiopians are closer to it than the Negroid cluster. Your "point" is something I never claimed otherwise?

Nope, you are making stuff up.

Nope, included due to centroid proximity. Minor clusters nest within major clusters.

Yes, Bantu are descended from Pygmies.

As I said I just highlighted the area since your "French cluster" is even more meaningless and confusing, I am just trying to clarify what you mean. Do you mean the huge cluster containing the French, or the French cluster?

Not at all. We are just establishing that it's possible to cluster by genetics, and explaining the IQ distribution is next.

No one is talking about US sociological races except you.

Taxonomy means "lumping in" or classifying together. So you make no sense.

No, it's a genomic PCA. They all are. All of the things you mention can be sometimes be weakly inferred from any genomic PCA, very preferably with other lines of evidence.

I chose to exclude X rather than include Y? Great logic. I excluded the Melanesians and Ethiopians from the Caucasoid cluster simply because a) the Melanesians are nowhere near it b) The Ethiopians fall between two clusters, centroids, or maxima ie. they are mixed. You know how the Matterhorn is not in the Himalayas, because it is nowhere near there, and how intergalactic material is not in a galaxy, because it's between. Same principle. Do you have trouble understanding the concepts between and distant?

No you were not, this PCA is from a different study to the one in Science Mag. It was created by Doug MacDonald (who gave me permission to reproduce it) purely to show global genetic similarity. Please email him and ask if you got it right. You cannot guess the motive for making a genetic PCA, since it is just a plot of genome data. You are pretending to know the field.

 

And they do a good job of that. You just don't like what they point out.

The cluster boundary you drew with the French, and a couple of Ethiopian dots (you tried to keep them all out, but you couldn't) in it. Along with some, but not all, of tightly clustered southern India, and none of the tightly associated mass of dots just below India, in it - for no visible reason. And so forth.

No one who talks about human races here is talking about anything else. We don't have a large Brazilian audience, for example.

The Maasai cluster does not "nest" in anything. The entire cluster, including any possible "centroid" (of what?) is clearly isolated. It is separate by significant visible distance from the Mbuti/Blaka/Bantu cluster (and there's some incoming comedy, folks - check this out: "Yes, Bantu are descended from Pygmies.". All four different Pigmy peoples? Or is he really talking about the San, ancestors of almost all of us?)

The principle is not the problem. The problem is that you need a scale and a criterion - a measurement of distance, or connectivity, or something. And you don't have one - or at least, you don't have a visible one that explains those boundary lines you drew. And no visible scale, certainly not whatever was used to plot the graph in the first place, gives you the clusters you need. Your graph is not clustered the way you want it to be, and drawing little lines on it just illustrates your own sociological preconceptions.

In other words: that featured genetic clustering destroys your sociological "racial" classifications (you have no US standard black, red, brown, yellow, or white, "racial" clusters in your graph. Your attempted white race probably includes Obama's father, depending on the exact dot). Your IQ classification is not going to align either, if Sculptor's nationstate breakdowns or your own unsupported claims or things like the Flynn effect are any kind of a clue. So what are you attempting to show? Looking at that graph, what are your "races" even going to be?

Next, is it.

Nobody ever said you couldn't cluster humans, or any biological organism on the planet, by genetics. That was agreed, from the beginning. In fact, it was observed, by me, that until you did that and defined your races by it, any attempt to compare genetic races by genetic IQ variance even (let alone "cognitive abilities") was a blunder. So take a look: it's a blunder.

 

Not at all. There still exists IQ variation between nations and self-identified race which needs to be explained, regardless of biologically defined race. I was just showing you that humans cluster genetically, and the clusters match self-identified race concepts. I don't expect you to admit that. What do you think is causing IQ differences between nations and self-identified races?

 

They don't. That's your immediate problem, in furthering your agenda: your genetic clusters don't match your sociological races.

Good question, if you're interested in IQ scores for some reason.

Obviously - since we are talking about sociologically defined populations and a sociological trait (IQ score) - the first place to look would be some kind of sociologically mediated influences. Personally, I would look first, and very carefully, at known culturally mediated influences on IQ score - including but not limited to:

neurotoxin exposure (lead, insecticides, common household chemicals, etc), diet and exercise (especially for girls), music lessons and reading, peer group vocabulary and literacy, breastfeeding and prenatal care, income inequality in the local community, noise and light levels during sleep.

And of course combinations of all these, and parental (especially maternal) status in all these combinations.

So that will keep you busy for a while.

 

You are assuming there are no biological differences between sociological races and nations? Why?

Do you think international adoption would control for these?

 

Why are you assuming there are?

 

A consistent international pattern.

 

I'm not. Obviously not. Where would you come up with something like that?

It's a matter of priorities. Even with biologically defined traits that vary by sociological category - like nearsightedness, bad teeth, or asthma - the first place to look for causes of the variance is in the sociology. And you are starting with a sociological trait.

Unlikely and very difficult to analyze (adoption is dominated by noise factors) but worth checking out if you have the time and resources. Of course adoption would be something you would have to control for anyway - so you'd be keeping track of it.

 

You have not clearly defined the difference between "biological" and "sociological" traits. Could you do so with some examples? Can you explain why we should look at the "sociology" and what this means? Is it different from looking at the environment and generating a heritability estimate?

What noise factors is adoption dominated by? How does one control for adoption?

 

Ok, anybody have any idea what's really going on here?