Donald Templer and J. Phillippe Rushton have recently published a study of the 50 U.S. states which finds strong correlations among the following varaibles: skin color, IQ, infant mortality, life expectancy, violent crime, HIV/AIDS, and income. This pattern mirrors what Templer has observed internatioanlly. Skin color tends to be more strongly correlated to the list of variables than does income, so the authors conclude that this points to biology rather than economic environment as the more important explanation of the pattern of correlations. They rely on Rushton's life history theory and Lynn's cold climate theory to make sense of why the variables hang together, but whatever the explanation, there seems to be a worldwide pattern of desirable (or undesirable traits) running together, and this is closely associated with skin color.
They review some interesting research by Ducrest et al. (2008) which found that in 20 wild vertebrate species, darker individuals are more aggressive, sexually active, and resistant to stress. They also tend to have a larger body mass and greater energy and physical activity. For example, darker maned male lions are reported to be more aggressive and sexually more active, and darker barn owls show stronger immune response. Darker coloration is associated with enhanced fertility, female sexual receptivity, male sexual motivation and performance, and higher levels of testosterone. Ducrest et al., however, cautioned that these findings might not apply to humans.
Tuesday, December 06, 2011
Skin color and desirable traits
Labels:
crime,
HIV/AIDS,
Infant Mortality,
IQ,
Life Expectancy,
Skin color
Subscribe to:
Post Comments (Atom)
39 comments:
we know the genetic architecture of pigmentation. that is, we know all the genes (~10, usually less than 6 in pairwise between population comparisons). skin color varies via a small number of large effect trait loci. in contrast, I.Q. varies by a huge number of small effect loci. so logically the correlation is obviously just a correlation. to give you an example, SLC45A2 explains 25-40% of the variance between africans and europeans.
long story short: it's stupid to keep repeating the correlation between skin color and I.Q. as if it's a novel genetic story. it's not. i hope don't have to keep repeating this for too many years.
I call Lewontin's Fallacy on Razib.
Razib,
So in the paper, the correlation is discussed in relation to differential K. Are you saying there could be no pigmentation-personality pleiotropy?
Ron,
I've been trying to track down data on immigrant IQ, specifically Black African IQ. 2nd generation Black Africans educationally perform so well, either a) hereditarianism is false or b) they perform well despite low IQs. (I went through the data and found that neither Affirm action or immigrant selection could account for the comparative success).
Anyways, before I ditch hereditarianism. I want to check out b) more. I found three data sources showing second generation Blacks with IQs of around 92. (Kao. 2004. "Parental influences on the education outcomes of immigrant youth"; Massey. 2007. “Black Immigrants and Black Natives Attending Selective Colleges and Universities in the United States”; Bennett and Lutz, 2009; How African American is the net black advantage? Differences in college attendance among immigrant blacks, native blacks, and whites.") If this holds, a hereditarian hypothesis is still tenable.
Anyways, I came across a paper, "Neighborhood and School Factors in the School Performance of Immigrants’ Children," which compared immigrant GPA using ADD Health data. They sorted by nationality and race and found 292 non-Hispanic Blacks. I was wondering if you could look to see if ADD Health contained IQ info on these - basically, if you could decompose your findings (http://inductivist.blogspot.com/2010/04/race-ethnicity-and-iq-according-to-add.html) into 2nd generation and 3rd + scores.
Thanks. Chuck.
Ok, I went through the data. I only found 10-16 kids who's parents were born in Africa, depending if I looked at place of birth for mother/father/or both. The mean IQs ranged from 100-105. Interestingly when I did a more inclusive research for Blacks who had mothers and or fathers born abroad, the mean (ranging from 93-96, with n's ranging from 26-62) was significantly different from the native white mean. I guess Black African are smart Blacks.
I call Lewontin's Fallacy on Razib.
you're a moron to call lewontin's fallacy on me as if it's awesomely impressive.
Are you saying there could be no pigmentation-personality pleiotropy?
did i say anything about personality in my comment? i specifically talked about the potential pleiotropy between pigmentation and I.Q. i told rushton years ago that we know enough about pigmentation to refute the connection pretty easily, but he doesn't care. so he's either stupid or ignorant. people who continue to push this after i point out the obvious logical problem are either stupid or ignorant.
Razib,
"did i say anything about personality in my comment?"
No. But that's why I asked. I trust your opinion on these matters.
No. But that's why I asked. I trust your opinion on these matters.
i'd bet against it at this point. used to think there was a connection. but we know all the pig. genes. that means that we have candidates which should be robustly replicated. they haven't. what i'm saying is that we know with ~100% certainty which genes control pigment variance. 95% of the game is figuring out the loci which control a trait, so if there's a causal relationship it should be relatively easy. just like pigment genetics was easy. it hasn't been. if you positive that trait X is strongly correlated with pigmentation due to pleiotropy than the elucidation of the genetic architecture of the trait should b easy. if it isn't, then that implies that there isn't any real causal connection.
I've been trying to track down data on immigrant IQ, specifically Black African IQ. 2nd generation Black Africans educationally perform so well, either a) hereditarianism is false or b) they perform well despite low IQs.
Immigrants are not representative. Hereditarianism would predict 'typical' Africans developing 80-90 IQs in Western environments. Elite Africans should develop much higher IQs in the same context.
The GSS lists 'African' as an ethnicity, and asks if people were born in the country, as well as how many parents and grandparents were born in the country.
Most of those who list 'African' as their ethnicity are simply African-Americans, and many black immigrants might check 'African' ethnicity regardless of what country they are from. There is also no real generational data (all those born in the US have four US born grandparents as well). However there is some GSS generational data for 'blacks'.
IQ (sample size):
'African', foreign born: 87 (n 77)
'African', US born: 90 (n 1343)
'black', foreign born: 88 (n 182)
‘black’, US born, parents foreign born: 93 (n 27)
‘black’, parents US born, grandparents foreign born: 87 (n 7)
‘black’, grandparents US born: 89 (n 2197)
The sample sizes are small, but suggest a modest 1st generation spike, and subsequent decline.
Two more things:
A) well-educated immigrants with foreign credentials perform at a much lower level than educated immigrants with US credentials. Foreign degrees are often awarded to people with low cognitive ability and their earning potential here tracks their IQ much closer than their credentials.
B) immigrants often have a number of economically beneficial traits and punch well above their IQ weight.
The truth is that blacks and whites matched for IQ perform about the same, but that's about where the comparison ends-- IQ tests are "biased" against just about every other group. Asian Americans perform like a group with an IQ 20 points higher. The disparity between Jewish IQ and Jewish achievement might be even larger.
I call Lewontin's Fallacy on Razib.
Good call. And Razib's poor retort is even less impressive.
Good call. And Razib's poor retort is even less impressive.
hey retard, my blog is on the first page of google's search for "lewontin's fallacy."
"Asian Americans perform like a group with an IQ 20 points higher."
Asian Americans, and not Asians in Asia?
you're a moron to call lewontin's fallacy on me as if it's awesomely impressive.
Not awesomely impressive. Just correct.
hey retard, my blog is on the first page of google's search for "lewontin's fallacy."
Too bad you forgot what you wrote. Also too bad that your response to criticism is to call people morons and retards, but hey, that's what I expected.
hey retard, my blog is on the first page of google's search for "lewontin's fallacy."
It's too bad that this fact didn't help you to really understand all of its implications. Pathetic.
"Immigrants are not representative. Hereditarianism would predict 'typical' Africans developing 80-90 IQs in Western environments. Elite Africans should develop much higher IQs in the same context. "
Jason, thanks for the info.
Per your comment, Elite Africans are only so elite. Let's say they represent the upper third of the population in terms of IQ. (This might be a generous assumption, given the numbers of Africans coming through the "diversity lottery," but let's go with it. ) Given a normal distribution, this puts them 1.1 SD above the phenotypic mean. Given regression, their offspring, then, would be .66 SD above. So if their offspring had IQ's of 100 when raised in the US, we could infer a genotypic population mean of 90. But, as you note, the typical hereditarian hypothesis preposes a mean between 80-90 -- and usually the estimate lands below 85. So, for such a hypothesis to be tenable, the IQ of the second generation needs to be below 100, maybe around 92.
Now, in terms of tests, everything that I've seen points to an IQ below 100. (The ADD health data that I mentioned was relative to a white IQ of 105). But we have to take into account the "totality of the evidence." And that includes educational success, which points to intellectual parity. To put it another way, taken by itself, the 2nd generation educational performance suggests an African geneotypic IQ of no less than 90.
Whatever the case, the important point is that we should be able to rule out various hereditarian hypotheses on the basis of immigrant data. Immigrant selection can be estimated by comparing immigrant IQs to national IQs. Genotypic differences can be estimated by comparing offspring IQs, after adjusting for parental immigrant selection.
i specifically talked about the potential pleiotropy between pigmentation and I.Q.
Who said anything about pleiotropy? You're attacking a strawman.
For the benefit of those who don't know, pleiotropy is defined as the control of more than one phenotypic characteristic by a single gene or set of genes.
Nobody is claiming that one gene or set of genes controls both skin pigmentation and IQ.
i told rushton years ago that we know enough about pigmentation to refute the connection pretty easily, but he doesn't care. so he's either stupid or ignorant. people who continue to push this after i point out the obvious logical problem are either stupid or ignorant.
This is clearly a pretty emotional topic for Razib, he being a dark skinned guy.
if you positive that trait X is strongly correlated with pigmentation due to pleiotropy than the elucidation of the genetic architecture of the trait should b easy. if it isn't, then that implies that there isn't any real causal connection.
If you posit that trait X is correlated with pigmentation, period, then elucidation of the genetic architecture of the trait should not be so easy.
It's very probable that skin color and IQ are not directly linked, in the sense of both sharing the same gene or set of genes.
OTOH, it is probable that skin color and IQ are linked, even if less directly than by sharing the same genes. What we find in practice is that dark-skinned people are less intelligent, on average, than are lighter-skinned people, on average. The boundary lines between the populations are blurry though, and some dark skinned people are quite intelligent while some light skinned people are quite unintelligent.
This is clearly a pretty emotional topic for Razib, he being a dark skinned guy.
Well, if it is true that there is strong pleiotropy for IQ on pigmentation (and yeah, I think Razib's analysis is basically correct, i.e. there isn't) then his F2 generation with a White chick could be either pretty awesome (all the stuff that makes Razib, brown guy awesome, and which maybe Whites don't have plus pigment boost) or totally suck (none of that makes Razib, brown guy awesome, and which maybe Whites don't have and no pigment boost).
But I can't imagine him getting too het up by that if it were the case - a chance of a very smart kid is probably worth more to him than the chance of a very dumb kid, given that most people (most of us included) probably full beneath the threshold for "retards" that he regards as a much of a muchness ("mindreading" from his blog here).
Those differences are all within population of course.
In the case of separate populations, populations which are darker and more cryptic could moderate the non-melanin related components down, even to the point where, for example a darker population might have lower levels. E.g. darker Filipinos vs lighter Europeans or darker Indians vs lighter Europeans (I think its at least arguable that Indians are more or equally sexually abstemious and restrained than Europeans- I don't think there's much in it)*.
With a mixed population, it might be more hard to avoid.
*This has the interesting implication that F2 hybrids in between these populations would have a chance of having a more extreme profile, in both directions.
Who said anything about pleiotropy? You're attacking a strawman.
oh christ, the retard can't read. from rushton's paper: The hypothesis that skin color is a genetic orrelate of IQ was endorsed by Jensen (2006) who suggested that pleiotropy
(genes having more than one effect) may underlie the relationship.
i told rushton several times to stop bringing up the hypothesis unless he outlines why it is trivially false.
also: OTOH, it is probable that skin color and IQ are linked, even if less directly than by sharing the same genes.
what i'm trying to explain to you morons is this: there are about 6 genes which control almost pigment variation between any two pairs of populations (the 6 vary, so the total number of genes is closer to 12 in the aggregate). there are at a minimum *hundreds* of genes which control IQ variation (probably thousands of genes, though fewer QTLs). to repeat myself to you retards: by definition pigment genes can't have much effect on IQ variance, there aren't enough of them. the only other alternative is that there are large effect IQ QTLs hanging out right at those pigment genes which decades of research happens to have missed, and which we have no evidence of in the genetic architecture of IQ. i guess i could draw something in crayon for some of you, but have i made it clear to you morons? science is awesome, and we've discovered a few things in the past few years. bother to pay attention please.
(and don't repeat lewontin's fallacy as if it's going to be an awesome retort to someone who is a contributor to a blog which helped familiarize much of the HBDsphere with the concept)
and i'm pissed because there are plenty of you morons who thinl your world views are awesomely rooted in Science. but the reality is that you don't give a shit about Science. this is a great time to keep up on human genomics, but a lot of you idiots treat it like a sugary supplement.
Razib, will you stop that childish crap please? You are not doing your reputation any favours by behaving like a buffoon.
by definition pigment genes can't have much effect on IQ variance, there aren't enough of them. the only other alternative is that there are large effect IQ QTLs hanging out right at those pigment genes which decades of research happens to have missed, and which we have no evidence of in the genetic architecture of IQ.
So your point boils down to "it would have been found if it were so". It's hard to see why you are so sure. Quite obviously, a cluster of genes involved in pigmentation can in principle have a large effect on IQ (or anything else). If it does exist, it will manifest itself only in between-populations comparisons and will be nearly undetectable in within-population search. Please cite a GWAS that had a chance of detecting something like that with regard to IQ. I contend that it does not exist.
Razib, will you stop that childish crap please? You are not doing your reputation any favours by behaving like a buffoon.
shut the fuck up.
Quite obviously, a cluster of genes involved in pigmentation can in principle have a large effect on IQ (or anything else).
your only candidate is SLC24A5. everything else segregates in appreciable minor frequencies among europeans (e.g., the "darker" versions of HERC2-OCA2, SLC45A2 and the KITLG loci). in which case you'd predict that middle easterners, who are nearly fixed for that locus, would cluster with europeans in IQ. they don't. i know this shit, you don't. so shut up.
"appreciable minor frequencies among "
note: in many european populations the "darker" version is not really even at a "minor" frequency.
e.g.:
http://hgdp.uchicago.edu/cgi-bin/alfreqs.cgi?pos=46213776&chr=chr15&rs=rs1426654&imp=true
oh,and if SLC24A5 is responsible for between a lot of pop diff. in IQ, the uyghurs, who are a relatively recent east asian-european mix, should be geniuses. they have east asian ancestry + a lot of european SLC24A5 from their west eurasian ancestors. naturally on expects that they traditionally dominate the chinese civil service, etc. they don't. you should also have massive between sibling differences in IQ above expectation in groups where the two SLC24A5 variants are segregating at ~50 percent (e.g., south indians). now that we really know pigmentation genetics it's trivial to do thought experiments in terms of what you might predict.
i know this shit, you don't.
No, you don't. You can't even comprehend what you read, much less think about it. All you are good for is parroting someone else. Hint: One has to be be a complete mindless moron to discount epistasis as completely and categorically as you do. But you spewing garbage is becoming tiresome, so I am not going to waste any time on educating you.
Hint: One has to be be a complete mindless moron to discount epistasis as completely and categorically as you do.
this is kind of an interesting issue, because i'm really into the potential of statistical epistasis at the tails of the distribution. but the reality is that you're either a moron or mendacious (in the latter case trying to impress the retards with a buzzword). people have been studying inheritance patterns of quantitative traits in populations where the putative between population loci have been segregating for decades. regular non-linearities would have been noted. retard.
"becoming tiresome"
good point though. need to go back to my blog where i have better control of the inevitable influx of the moron masses. bye-bye knuckle-draggers :0)
Razib, you put up quite a performance, calling your critics morons and retards, telling them to shut the fuck up, asserting that only you know what your talking about. That's science, Razib-style!
And yet your emotionality backfires. Only someone unsure of himself or so emotionally invested in the topic would talk that way. But you clearly don't get that.
need to go back to my blog where i have better control of the inevitable influx of the moron masses.
Translation: "Everybody must bow down before Mighty MEEEEEE!"
It takes a certain amount of ego to be a blogger, but you are verging on being mentally disordered.
Rushton and Templer just use the black % of a state's population for their skin color measurement. That seems pretty crude, and misses the potential differences among Amerindians, Asians, and Europeans of various ethnicities (ie, southern Italian v. scandanavian).
As someone who is merely observing the conversation in this thread, I've been surprised by the sheer amount of hostility shown over this topic, particularly from Razib. I always knew he hated idiots, but his over-the-top attempts to tear them down seem a bit childish.
AE: Agreed.
Interesting thread. Razib's point seems straightforward enough.
In terms of crime, I think Rushton & Templar might find it more productive to look into the MAO-A research that the blog 'Unsilenced Science' discusses.
Temple Grandin says that white-skinned domestic animals tend to be more neurotic and troublesome.
Razib's hostility didn't seem that different from his Discover blog. It's just that he bans people earlier so there aren't extended arguments.
For what it's worth, he actually provided some scientific citations while folks here complained about his attitude. I've got as big a preference for the classic style in these discussions as anyone, but the presence of facts trumps that.
Post a Comment