Race is a human/social construct.

[lomiller]

Just to review, haplogroups and racial groups don't match up very well.

And even if they did race would be redundant, as anything you could use race for could just as easily be done with the haplogroups that comprise that "race"

 

[Skeptic Ginger]

More sources reiterating the lack of racial haplotype groupings.

Most SNP variation is within all groups

For most SNPs, any population has individuals of all possible genotypes for a SNP, but populations differ in the frequencies of individuals with each of the different genotypes. About 85 percent of human SNP variation is within all populations, and about 15 percent is between populations, as shown in the figure below. Thus two random individuals within a village are almost as different in their SNP alleles as any two random individuals from anywhere in the world. Although a small proportion of SNPs have alleles that are common in some groups but rare in others, most SNP alleles that are common in one group will be common in other groups. Under the Common-Disease/Common-Variant theory, common variants that contribute to a disease in one group will also contribute to the disease in other groups, although the amount of the contribution may vary.

Common haplotypes are in all populations

The pattern of variation within and among populations for haplotype structure is just starting to be studied on a large scale. Recent studies show that the common haplotypes are found in all populations studied, and that the population-specific haplotypes are generally rare. African populations generally have more haplotypes than other populations, which generally have subsets of the African ones, due to the origins of other populations from ones that spread out of Africa.

There's a lot more useful discussion in the paper.

 

[lomiller]

The IQ haplogroup paper.

No one's attacked it yet; would be interested in a critique from a biologist's pov.

https://lesacreduprintemps19.files.wordpress.com/2012/05/haprinderm.pdf

Shows correlation not causation
Conciliation is to a questionable metric
Paper was released under a "pay to publish" model indicating it couldn't clear peer review in a real journal

 

[bpesta22]

That doesn't at all address my question. Where do you draw lines? What is your basis for those lines? And keep in mind my point about gradient (range) of differences. Correlate to what exactly?

Variance explained.

If you want further demarcations, I would use effect-size guidelines to label the effects as small, medium or large. If you want discrete scores that precisely demarcate races, I can't help, as we're dealing with fuzzy categories (but, e.g., where on a scale measuring political attitudes does one get labelled "republican" such that all scores below that number = "not republican"?).

I posted a chart above listing a number of things that co-vary with "race". The chart had words, but one could put in numbers for each variable, and then test what categorizing people by race can do in terms of explaining variance.

To see whether race is a meaningful category, we could draw a nice / diverse / representative sample 1000s of people, and then measure all of them on a host of psychological and physical traits (including both self-reported race and predictions about race based on haplogroups).

FA would reveal that scores on many traits all co-vary on a single factor, and the factor seems to represent race. One can then work backwards and predict any person's race (defined socially or genetically) using the other variables.

FA would not reveal (my bet) that hair color, handedness and blood groups load together on any factor. In fact, I'd suspect no factor would emerge that is surprising (i.e., represents a new categorization scheme that humans don't already use). I say this because humans are pretty already obsessed with categorizing things.

Someone is eager to reply that purely social things might explain why Race exists as a factor explaining a crap-ton of variance in the observed scores. That gets us into a massive literature where social variables are controlled to explain a race difference, yet fail to do so. To wit, the IQ/haplogroup paper I cited controlled for income, education and nutrition, yet haplogroups still predicted IQs.

This is also why I think the experiment I proposed above -- if done-- would get lots of people to stfu, one way or the other.

Were I more proficient in genetics, I'd apply for a Pioneer Fund grant.

 

[bpesta22]

Shows correlation not causation
Conciliation is to a questionable metric
Paper was released under a "pay to publish" model indicating it couldn't clear peer review in a real journal

Impossible to show cause here, sadly, as like many variables in social science, random assignment is impossible. Cause implies correlation, though.

The journal is not pay to publish. It's peer-reviewed, Elsevier. It's the premiere journal in intelligence research (why may be like being the tallest midget) and has a relatively high impact factor (which took a hit from 3.2 to 2.7 this year, for some reason).

Questionable metrics usually don't correlate with other things.

 

[bpesta22]

Just to review, haplogroups and racial groups don't match up very well.

How do you confirmation-bias out the studies I cited (secondarily, as cited by Wade) that show near perfect alignment?

 

[bpesta22]

And even if they did race would be redundant, as anything you could use race for could just as easily be done with the haplogroups that comprise that "race"

That's exactly my point. "Race" is a proxy for the genetic effects resulting from humans evolving in different environments for 10s of 1000s of years.

The proxy ("please self report your race") though seems to correlate near unity with the genetic measure.

 

[Skeptic Ginger]

How do you confirmation-bias out the studies I cited (secondarily, as cited by Wade) that show near perfect alignment?

I've told you but you don't seem to understand a very basic research process problem. You fit the data to your hypothesis by ignoring the data that doesn't support it. It's not the first time this kind of error has occurred. I keep trying to find a link describing the error but when I search for 'data-mining' there are too many other hits.

And you don't understand what people are trying to explain to you, yes, you can choose any number of things and find the cluster of correlating genetic sequences. All you've done is identified a cluster of genetic markers that identify an arbitrarily designated group.

Make up anything you want and call it a race. How about people with type O- blood, blonde hair, light skin and blue eyes? Is that a race? You could find the cluster of genetic markers that successfully predicts such a human group.

Pick another group. A+ blood, brown hair, brown eyes and taller than 5'5". Same thing, you can find the cluster of genetic markers these folks would have in common. Why are they any less of a racial group than the criteria you've chosen?

 

[bpesta22]

I've told you but you don't seem to understand a very basic research process problem. You fit the data to your hypothesis by ignoring the data that doesn't support it. It's not the first time this kind of error has occurred. I keep trying to find a link describing the error but when I search for 'data-mining' there are too many other hits.

And you don't understand what people are trying to explain to you, yes, you can choose any number of things and find the cluster of correlating genetic sequences. All you've done is identified a cluster of genetic markers that identify an arbitrarily designated group.

Make up anything you want and call it a race. How about people with type O- blood, blonde hair, light skin and blue eyes? Is that a race? You could find the cluster of genetic markers that successfully predicts such a human group.

Pick another group. A+ blood, brown hair, brown eyes and taller than 5'5". Same thing, you can find the cluster of genetic markers these folks would have in common. Why are they any less of a racial group than the criteria you've chosen?

We need to agree to disagree-- you may have the last comment, but the variables you picked here do not correlate with themselves and so cannot combine to scale people in any useful or reliable way. That to me defines arbitrary.

 

[RandFan]

Variance explained.

If you want further demarcations, I would use effect-size guidelines to label the effects as small, medium or large. If you want discrete scores that precisely demarcate races, I can't help, as we're dealing with fuzzy categories (but, e.g., where on a scale measuring political attitudes does one get labelled "republican" such that all scores below that number = "not republican"?).

I'm sorry but this just fails to address what I'm saying. I don't know if you simply don't understand or if you are ignoring me.

SG get's it...

And you don't understand what people are trying to explain to you, yes, you can choose any number of things and find the cluster of correlating genetic sequences. All you've done is identified a cluster of genetic markers that identify an arbitrarily designated group.

Bpesta, why is it that your deliniations are not arbitrary?

 

[lomiller]

That's exactly my point. "Race" is a proxy for the genetic effects

Even if you want to redefine it that way all you have done is make it redundant because it doesn't accomplish anything you can't already do

 

[bpesta22]

Even if you want to redefine it that way all you have done is make it redundant because it doesn't accomplish anything you can't already do

How else would you prove that race has a genetic basis? The redundancy *is* the proof.

"Please self report your race." I then show these self reports correlate strongly with physical and psychological traits. I then then show that including a measure of mutations on Y makes these correlations shrink to zero. To me, the more complete the redundancy, the stronger the proof*.

What else do you require?

*It's quite possible that the self-reports-- carrying social baggage-- would be more powerful predictors of social outcomes than the haplogroup measure. In other words, it's quite possible social measures of race are not redundant with genetic ones. In fact, I think this is what you guys must predict.

 

[Kthulhut Fhtagn]

http://psychology.uwo.ca/faculty/rushtonpdfs/pppl1.pdf

(anyone ad homming Rushton or Jensen needs to also explain why an APA journal accepted their article for publication. Conversely, I agree that just because it's published in an APA journal doesn't make it true).

Publishing an article isn't the same thing as the APA endorsing their view. For that matter, I wouldn't consider it "ad homming" to point out that Rushton has published material in the past of severely dubious quality. Suffice to say, I don't know why no one can point out the man made his career on being less than honest and find that irreconcilable with the fact that the APA published an article from him.

ETA: In fact, I consider it unusual that you feel the need to defend Rushton's character in the same post where you present a study from him. That's pretty unusual, I don't normally see individuals providing citations and defending the character of the individual they're citing. My guess, and I could be reaching here, is that you have experience with posting Rushtons papers elsewhere and far too many individuals who have commented on the issue of race are aware the man is a bleeding racist.

EETA: Just as a preemptive response; how do I know Rushton is a racist? Perhaps because of the fact that he spoke at race separatist Michael Harts Preserving Western Civilization whose stated purpose is:

We believe that America’s Judeo-Christian heritage and European identity must be defended. Today, our glorious Western civilization is under assault from many directions. Three such threats will be discussed at this conference. First, the massive influx to the United States and Europe of Third-World immigrants who do not share our fundamental political and cultural values. Second, the threat from Islam, a militant ideology that is hostile to our society and, in principle, committed to destroying it. Third, because of the persistent disappointing performance of blacks (which many whites mistakenly blame on themselves) many whites have guilt feelings that undermine Western morale and deter us from dealing sensibly with the other threats.

 

[DC]

Impossible to show cause here, sadly, as like many variables in social science, random assignment is impossible. Cause implies correlation, though.

The journal is not pay to publish. It's peer-reviewed, Elsevier. It's the premiere journal in intelligence research (why may be like being the tallest midget) and has a relatively high impact factor (which took a hit from 3.2 to 2.7 this year, for some reason).

Questionable metrics usually don't correlate with other things.

the journal is not called Elsevier, that is the publisher of the journal that is called Intelligence.

 

[DC]

Shows correlation not causation
Conciliation is to a questionable metric
Paper was released under a "pay to publish" model indicating it couldn't clear peer review in a real journal

they have a pay to publish model? the only thing i am aware of is the open access model where the author has to pay. but that option is only available once the paper passed peer review, according to the Journals homepage.

 

[DC]

That's exactly my point. "Race" is a proxy for the genetic effects resulting from humans evolving in different environments for 10s of 1000s of years.

The proxy ("please self report your race") though seems to correlate near unity with the genetic measure.

evidence for this claim? i have read/heard from several scientists that this actually not the case at all.

 

[bpesta22]

I'm sorry but this just fails to address what I'm saying. I don't know if you simply don't understand or if you are ignoring me.

SG get's it...

Bpesta, why is it that your deliniations are not arbitrary?

In post 172, I gave an arm-chair definition of race:

"a continuum (probably several) rather than an all or none classification...My definition: skin color and it's covariates arising from shared ancestry and evolution via reproductive isolation over time...Steve Sailor's definition is not bad either: A racial group is an extended family that is inbred to some degree."

***

To me, race is a constellation of **inter-correlated** traits that must map near perfectly (in time, in the genome, historically) to evolution resulting from different selection pressures our ancestors faced because they were geographically isolated. I must predict historical accuracy. Lumping people based on SG’s arbitrary examples do not require this.

This is 100% dust-bowl empiricism; it is completely data driven; there is no theoretical lens here that potentially blinds me in a confirmation-bias sort of way. The nexus of traits either both (a) exist and (b) map to the genome, or they do not.

Re: Arbitrary.

My delineations are not arbitrary because they go-together / vary across a category humans have created named "race". Turns out, the essential features that comprise the social category have very reliable markers in the genome. Nobody has ever created the category that includes blood groups, handedness and hair color (for good reason).

I get the counterargument, I think. We can point to the genome and say "here's where blond hair occurred...over there, we get blue eyes..., and here again, this blood type". Why not use these dimensions to classify humans / why obsess over skin color and its covariates / why give it preferential treatment over other variables whose emergence we can ID in the genome?

1) These other variables do not map on to human history like race does. Afaik, there were no selection pressures that gave left handers an advantage in Africa, but not in Europe. Same goes for hair color or blood groups. [Ironically, I suspect hair color does contradict what I just said, but only because it co-varies with race (and therefore our evolutionary history).]

2) These other variables do not correlate with each other. Handedness and hair color are independent; and I suspect you cannot trace presence/absence of both traits to the same place in the genome (afaik).

Whatever the effect of blond hair, it is irrelevant to the effect of being left handed. You can’t therefore lump these variables into the same classification scheme, and a one-variable classification scheme seems pretty worthless.

That’s all I got.

 

[bpesta22]

Publishing an article isn't the same thing as the APA endorsing their view. For that matter, I wouldn't consider it "ad homming" to point out that Rushton has published material in the past of severely dubious quality. Suffice to say, I don't know why no one can point out the man made his career on being less than honest and find that irreconcilable with the fact that the APA published an article from him.

ETA: In fact, I consider it unusual that you feel the need to defend Rushton's character in the same post where you present a study from him. That's pretty unusual, I don't normally see individuals providing citations and defending the character of the individual they're citing. My guess, and I could be reaching here, is that you have experience with posting Rushtons papers elsewhere and far too many individuals who have commented on the issue of race are aware the man is a bleeding racist.

EETA: Just as a preemptive response; how do I know Rushton is a racist? Perhaps because of the fact that he spoke at race separatist Michael Harts Preserving Western Civilization whose stated purpose is:

Met Rushton once; talked for over an hour; liked him; think he's been vilified (even if everything he ever wrote turns out to be wrong); will defend him because his vita is almost always longer and nicer than the CVs of those who dismiss him as racist.

Realize the penalty one takes for even doing research in this area. That explains why citations often involve the same small set of players. That's why I think journal quality is key here.

 

[bpesta22]

the journal is not called Elsevier, that is the publisher of the journal that is called Intelligence.

Yes, sorry if I implied otherwise. The publisher is well-respected (of course the implication is not "so everything they publish is true").

 

[bpesta22]

evidence for this claim? i have read/heard from several scientists that this actually not the case at all.

Posted above re my discussion of Wade's book. Would be interested in those studies suggesting otherwise (though it's possible SG linked these already; too busy replying here for now...).