Skin Color & Earning Studies=Woo?

[blutoski]

So you are an IQ fan. Do you also happen to know how IQ is measured in people who haven't learned to read and write, i.e. in (many places in) Africa?

I'm not an "IQ fan", but there are tests available for the illiterate. They use pictures and ask the participant to pattern-match, or arrange panels in the correct sequence or upright orientation.

 

[bpesta22]

So you are an IQ fan. Do you also happen to know how IQ is measured in people who haven't learned to read and write, i.e. in (many places in) Africa?

http://harcourtassessment.com/HAIWEB/Cultures/en-us/Productdetail.htm?Pid=015-8686-195

 

[dann]

Just a correction here. Bell Curve was 1994, Mismeasure, 1981, Gould was good, but not that good.

You should get hold of a more recent edition of the book, post bell curve:

The 20th-century obsession with numbers led to the institutionalization of IQ testing and subsequent assignment to work (and rewards) commensurate with the score, shown by Gould to be not simply misguided--for surely intelligence is multifactorial--but also regressive, creating a feedback loop rewarding the rich and powerful. The revised edition includes a scathing critique of Herrnstein and Murray's The Bell Curve, taking them to task for rehashing old arguments to exploit a new political wave of uncaring and belt tightening. It might not make you any smarter, but The Mismeasure of Man will certainly make you think.

http://www.amazon.com/Mismeasure-Ma..._bbs_sr_1/102-9360498-6692929?ie=UTF8&s=books

 

[blutoski]

Just a correction here. Bell Curve was 1994, Mismeasure, 1981, Gould was good, but not that good.

The "Revised and expanded edition" is from 1996, and specifically has chapters dedicated to refuting The Bell Curve.

In fact, the current (and probably final) edition is subtitled: "The definitive refutation to the argument of The Bell Curve"

 

[hammegk]

And back and forth we go ....

http://www.debunker.com/texts/jensen.html

http://www.lrainc.com/swtaboo/stalkers/jpr_gould_paid.html


and more, more, more ...

http://www.cse.emory.edu/sciencenet/mismeasure/http

 

[Dave1001]

That's a good question.

It really wouldn't surprise me, though, if there was a subtle racism in society that resulted in an economic disadvantage to darker skinned immigrants.

I'm sure it wouldn't surprise many of us. But that doesn't excuse cherry-picking, etc. to get a study to match up with popular expectations. And situational discrimination in a variety of settings is different then the sort of sweeping, universal discrimination model that Hersh claims the data unfalteringly supports. I'd like to see more details on this study and some good critical analysis by skeptics in her field.

Edited by Lisa Simpson: 

Inappropriate remark removed.

 

[andyandy]

edited to delete comment calling people who have derailed the thread effing winkers*

goodness, someone's cranky

threads evolve into questions that people are interested in discussing - it happens. It's hardly a big derail - if bpesta believes that skin colour and earning studies should have taken into account IQ data, then it's entirely relevant in his answer to your initial question. If people are interested in discussing an OP they will, if it's not especially interesting they won't.....
don't throw the toys out the pram just because your OP hasn't sparked the interest you'd hoped, in the direction you'd hoped....

*like Christiano Ronaldo

 

[bpesta22]

goodness, someone's cranky

threads evolve into questions that people are interested in discussing - it happens. It's hardly a big derail - if bpesta believes that skin colour and earning studies should have taken into account IQ data, then it's entirely relevant in his answer to your initial question. If people are interested in discussing an OP they will, if it's not especially interesting they won't.....
don't throw the toys out the pram just because your OP hasn't sparked the interest you'd hoped, in the direction you'd hoped....

*like Christiano Ronaldo

This is clearly off-topic you f'n f'er

 

[luchog]

Just a correction here. Bell Curve was 1994, Mismeasure, 1981, Gould was good, but not that good.
And I would not be too quick to call Pesta a racist. He's clearly not in Rushton's camp.

Nope, 1981 was the first version. Gould published a revised second edition in 1996, with new material included specifically to refute the claims made in The Bell Curve.

 

[luchog]

I see it as a levels of analysis issue. Look at the african countries. The represent basically one "race"-- africans. And, within that race alone, skin color clearly associates with IQ (I deleted the data from my stats package, but it would be interesting to go back and do the correlations just on "black" countries).

Nope, just more of your pseudoscientific BS. "African" incorporates a number of different distinctive ethnicities; including the caucasian Egyptians and Imazighen (berbers), both of which are among the oldest indigenous populations in Africa. Even the "black" ethnicities are not seperate races, since they are all members of the same Homo Sapiend Sapiens subspecies.

 

[Tanja]

I have to admit I did not read the whole thread, I just skimmed through it. However, I would like to comment on the table that was given on page 1, which lists various countries, their supposed average IQs, skin colour measurements etc. I find the average skin colour measurement quite bizzare.

For example, for European countries, they give a "1" to most west European countries, but they give a "1.3" for the Czech Republic, "1" for Slovenia, "2" for Croatia.

Now, even if you disregard the fact that most West-European countries had more non-white migration than Eastern-European countries (for example, if you disregard that the UK, with skin colour of "1", has nearly 10% of Asian and Afro-Carribean population, and you take into account only white Britons) it is in my view stupid to suggest that Britons are whiter than the Czech (if anything, I would think it is the opposite), or that Slovenians are whiter than Croats. Those figures look completely made up to me.

 

[bpesta22]

From the article:

A physical anthropology source was used to obtain data on skin color ( Biasutti, 1967). It should be noted, however, that physical anthropologists have traditionally assessed skin color inside the upper arm, which is affected only minimally by sun exposure. The source contains a map of the world with eight categories of skin color ranging from 1 (very light) to 8 (very dark). Because the map does not delineate the various countries of the world, three graduate students who were unaware of the purpose of our study independently determined the predominant skin color for each of the 129 countries. The word predominant was used because some countries had more than one skin color. The product–moment correlation coefficients between raters were 0.95, 0.95, and 0.93, suggesting very little subjectivity. For each country, the mean of the three skin color ratings was used.

Biasutti, 1967 Biasutti R., Le Razze e i Popoli Della Terra, (1967), Unione pipografiza-Editrice Torinese, Torino, Italy.


Tidbit from the main articles discussion section:

A possible reason why skin color correlated more highly than temperature with IQ is that it is a multigenerational reflection of climatic history. It takes thousands of years for skin color to change through evolution. For example, desert Amerindians are lighter than Negroid persons south of the Sahara Desert because they have not lived in a hot climate for as long. Mongoloid persons moved from Siberia to the Americas relatively recently. Nevertheless, it should be acknowledged that the present findings provide more support for the contention of Rushton and Lynn that higher IQs are found in colder climates than their postulated evolutionary processes. The formulation of any postulated evolutional process involves hundreds if not thousands of facts and relationships. The correlations of the present study are consistent with the evoluational postulates of Rushton and Lynn but do not provide definitive evidence.

 

[bpesta22]

They did also do correlations separate for the 41 black countries in the study; FWIW, the correlation just using black countries is much smaller, though significant.

The authors attribute the smaller correlation for black only countries to restriction of range.

 

[Tanja]

Bpesta, if the skin colour data was taken from a 1967 study, how come they have separate data on Slovenia and Croatia, which were not independent countries then? Is that a later addition?

Again, I am asking this out of curiosity rather than anything else as I don't have enough knowledge about the subject to debate it one way or another.

 

[Schneibster]

Derail accomplished, looks like to me. Another name for the ignore list, though, so it's not a total loss.

 

[Schneibster]

Interesting, luchog, didn't know that Gould had refuted that. I'll keep that in mind. Thanks for the information.

 

[bpesta22]

Tough crowd on the derail issue-- I'm betting this thread would long be dead without the derails.

I still perceive the IQ stuff to be marginally relevant to the OP, and considering this is an internet discussion thing, I'm gonna post again at the risk of making other people's ignore lists.

Tanja:

From the article, it seems like the 1967 source was from a physical anthropologist's study. That study included a map of the world's geography, without identifying any countries.

The map had 8 different shades of color showing the average skin colors of people who lived in that area.

I guess it makes sense to me to do it this way, as just crossing a neighboring country's border likely does not result in a radical change in skin color.

Given that countries weren't identified in the '67 source, the IQ authors enlisted the help of grad students. They coded skin color by country using their eyeballs, with apparently reliabilities in the .90s

***

Just my take on the above mess. It wouldn't surprise me if skin color-- even within "races"-- correlated with IQ.

It also wouldn't surprise me if it did not, nor do I think Spearman's hypothesis predicts one way or another whether skin color should relate to IQ within races.

This study alone seems informative, but definately not definitive in addressing whether skin color is "additive" re its effects on IQ.

The thing that struck me was the linear / additive relationship between skin color and IQ in the scatter plot of countries by skin color.

The plot would look a lot different if only skin color confounded with race predicted IQ (i.e., the graph would show discrete groupings of white and black countries, etc, with little variance within race-- whereas the actual graph showed data points all across the skin color variable).

So, I'm not sure what to make of it all, and perhaps I should apologize now for dragging this in, though, I still think it relevant and my intent was not to troll or stir *****.

 

[hammegk]

Interesting, luchog, didn't know that Gould had refuted that. I'll keep that in mind. Thanks for the information.

Gould didn't. What's closer to being refuted is Gould's Mismeasure. (But of course you won't see this.)

 

[Roboramma]

Plus, I am well aware that people within the same race show individual differences. Inidividual differences within a race don't preclude mean differences across two race groups.

Wait, where did he make this argument? Did you read his post, or are you just arguing with what you expected him to say?

The point of the data in the table I posted was to test a hypothesis for why skin color might correlate with IQ. The aticle asserts that distance from the equator is directly correlated with intelligence-- life's easier in warmer climates; requiring less g to survive. Life's harder where it snows, which selected for smarter people. All this being confounded with skin color.

It's an interesting idea, but at this stage little more than that. Certainly the evidence given doesn't show causation.

I don't disagree with what you've said (except your claim that culture influences score on an IQ test), I just think it's irrelevant to anything I've claimed above.

Well, considering that his whole post seems to be about the fact that you've just assumed that culture doesn't influence scores on an IQ test, it seems like you are disgreeing.

 

[bpesta22]

Wait, where did he make this argument? Did you read his post, or are you just arguing with what you expected him to say?

It's an interesting idea, but at this stage little more than that. Certainly the evidence given doesn't show causation.

Well, considering that his whole post seems to be about the fact that you've just assumed that culture doesn't influence scores on an IQ test, it seems like you are disgreeing.

Not sure I'm following your arguments.

I thought my claim that: "individual differences with a race don't preclude mean differences across races" summarized my take on the points he was making.

I don't quarrel with the idea that within the african race, there's huge variability on probably any variable you want to measure. Again, though, this brings me back to the point above.

In the gif here, substitute perceptual effect for IQ. The left distribution might be Africans; the right distribution whites.

The graph shows millions of africans scoring higher on IQ tests than 10s of millions of whites (the point where the black distrbution crosses over into the white distribution and beyond).

That said, there's still a mean difference favoring whites.

This is my claim; the mean difference favoring whites. I'm not claiming that all blacks are the same, or that all blacks score lower than all whites, or even that the mean difference is caused by genes. I think the best conclusion for what causes the black/white difference is still: We don't know.

...

I don't know what culture causes re any other variable, I do believe culture has little to no affect on Spearman's g.