"If you can judge a wise man by the color of his skin, then, mister, you're a better man than I."

I often listen to classic rock radio stations. Once in awhile I hear "Livin' on the Edge" by Aerosmith, and think about why they inserted the following lyrics: "If you can judge a wise man by the color of his skin, then, mister, you're a better man than I." The song came out in 1993, so this was prior to the Bell Curve brouhaha. Rushton was publishing articles on the issue, but I doubt Rock-n-Rollers would know anything about that. Too busy snorting coke and screwing groupies. Maybe the band was simply responding to the stereotype that blacks are unintelligent.

So, are they right about choosing wise men? In one way, I would answer yes. All races are made up of mostly unwise people. If you have two random people, one black and one white, and you guess the white one is the wise one, you're likely to be wrong. Chances are, they're both ordinary.

On the other hand in choosing the white guy, you'd be less likely to be wrong. If we were to set the cutoff for wiseness at an IQ of 115, for example, there is a 16% chance you'd be right with the white guy, but you've only got a 2.5% probability with the black guy. In other words, you're 6.4 times more likely to guess right by betting on the white person.

The Left calls us racists, but the truth seems to be that they are the people who think in absolutes and categories, and we're the guys who didn't sleep through stats class and so know a little something about probability.

Which racial groups think science does more harm than good?

While I am not a science worshipper, I do think it offers answers to many of our problems, at least eventually. Unfortunately, science doesn't happen magically. It requires very talented people and a society with the means and culture to support scientific work. 

Americans vary a great deal in their attitudes toward science. The General Social Survey (GSS) asked participants the following: "How much do you agree or disagree that, overall, modern science does more harm than good?" Answers ranged from "strongly agree" (1) to "strongly disagree" (5), so higher scores indicate greater belief in science. Here are the means by ethnic group (sample size = 6,226; I included religious affiliation as well):

Mean belief that science is good

Finnish  4.16

Jewish  4.08

Russian  3.99

Japanese  3.95

Lithuanian  3.87

Hungarian  3.86  

Buddhist  3.84

Austrian  3.83

Greek  3.83

English/Welsh  3.82

Danish  3.81
Scottish  3.81

Swedish  3.80

Norwegian  3.78

No religion 3.78

Italian  3.76

Irish  3.72

French  3.71

Hindu  3.67

Orthodox Christian  3.65

German  3.64
Swiss  3.64
Asian Indian  3.61

Dutch  3.61
Spanish  3.61

Total Sample  3.60

Muslim  3.60

Catholic  3.59

Chinese  3.53

Arab  3.50

Immigrants  3.37

Mexican  3.26

Black  3.20

American Indian  3.00

West Indian  2.82

The pattern is clear: whites groups have much more faith in science than non-whites. Japanese Americans, a smart, successful group, are an exception.  Among whites, Finns and Eastern/Central Europeans tend to be the most positive about science. Among religions, Jews come out on top. The gap between Finns at number one and West Indians on the bottom is well over one standard deviation--a very large difference. 

I included immigrants as a group: they are toward the bottom of the list. If we want to have a country fully in support of science, our immigration trends are counterproductive. 

It looks like some of the explanation for the pattern we see is IQ, but I wouldn't be surprised if non-whites have a tendency to see science as an evil white thing. Think Tuskegee. It's BS, but this might help explain negative attitudes. 

What about liking rap? Is it a racial or an IQ thing?

As a follow-up to my last post on IQ, race, and classical music, let's see if a fondness for rap music is more about race or intelligence.  I'll use the same statistical technique as last time (OLS regression) with General Social Survey (GSS) data (sample size = 926):

Liking Rap Music (standardized OLS regression coefficients)

IQ   -0.08*
Black   0.21***
Other non-whites   0.09**

*p < .05, **p < .01, ***p < .001

Smarter people are less likely to enjoy rap, and non-whites, especially blacks, are more likely than whites to be fans regardless of intelligence.

While we found in the previous post that smart people like classical music regardless of race, here we see that a taste for rap is quite racial.  It is best predicted by race: blacks (and other nonwhites) like it, whites do not. 

Does prenatal testosterone determine sexual orientation? A look at racial differences

The conventional wisdom is that sexual orientation, at least for men, is genetic, but the heritability estimates are actually rather weak.

Some researchers have theorized that homosexuality is caused by prenatal exposure to testosterone (T).  Baby boys develop into gay men if they are exposed to insufficient levels of T during a critical period of development, and baby girls develop into lesbians if they were exposed to high prenatal levels. The level of T isn't the only factor: individuals also differ in terms of how sensitive they are to T.

If such a theory were true, we should see racial differences in sexual orientation because blacks tend to be more sensitive to T than whites. Their androgen receptors genes located on the X chromosome tend to be more responsive to androgens. We should see a lower prevalence of black gay men and a higher percent of black lesbians.

The General Social Survey asks about sexual orientation.  Here are the responses by race (sample size = 26,227):

Percent homosexual
Blacks
Men  3.8 
Women  2.2

Whites 
Men  3.0 
Women  2.3

These numbers fail to support the theory for both men and women.  There should be more white gay men, but there are fewer.  There should be more black lesbians, but their prevalence is very similar to that of white women. 

How do black and white males differ in a way that would lead to more black gay men?  I'm not sure.  Greg Cochran theorizes that homosexuality is caused by some unidentified prenatal infection, and it might be the case that blacks live in conditions conducive to more infections.

Data: Which ethnic group has the most beautiful women?

Which ethnic group is the most beautiful?  General Social Survey interviewers -- predominately middle-aged white women -- were asked to rate the physical attractiveness of female participants (US residents) with a range from 'very unattractive' (1) to 'very attractive' (5). Here are the means for various groups (with at least ten participants, N = 1,152):

Mean Physical Attractiveness

Spanish  3.93
Puerto Rican  3.76
Muslim  3.64
East Asian  3.57
Mexican  3.56
Italian  3.55
Asian Indian  3.55
Jewish  3.48
Russian  3.47
French  3.43

Total Sample  3.42

German  3.38
Irish  3.38
Czech  3.36
English/Welsh  3.35
Black  3.34
Dutch  3.27
Polish  3.25
American Indian  3.23
French Canadian  3.23
Scottish  3.16
Swedish  2.92

Asians and and darker Caucasian and Hispanic groups have high means, while fairer whites (e.g., Swedish!), American Indians, and blacks have low means.  The gap between the highest group (Spanish) and the lowest (Swedish!) is over one standard deviation -- a huge difference.

This is somewhat consistent with my judgment -- I would give Spaniards a high score and blacks and American Indians low scores -- but north European women seem at least as beautiful to me as southern Europeans. I would give Greek women (with only four cases, too small include here), for example, a low score.  Eastern European women are beautiful, but Czech, Polish, and Russian women don't do very well here.

I thought ratings by interviewers would be more valid than self-ratings, but now I'm not so sure.  The tastes of middle-aged white women don't line up very well with this middle-aged white dude.

I imagine that many of these survey workers are liberals. Perhaps they penalize fair whites because they're too white.

UPDATE:  I discussed this with my wife, and we wondered if intrasexual competition is influencing the ratings. With the Barbie-type being the traditional American ideal, perhaps interviewers penalize fairer white women. It is weird that these interviewers are giving low scores to the women most likely to match the supposed cultural ideal.  (My wife says she is inclined to unfollow blondes who are just too perfect.)

Data: Sex and race differences in self-rated athletic ability

One study of athletic performance put heritability at .66.  In plain English, two-thirds of the differences in athletic ability are explained by genetic differences.

General Social Survey participants were asked to rate how athletic they are on a scale from 1 to 5.  I calculated the means for sex and race combinations (N = 2,373):

Black men rate themselves as most athletic with a mean of 3.57.  White women (2.58) and women of some other race (2.56) come in at the bottom.  The gap between black men and other-race women is nine-tenths of a standard deviation (SD) -- a big difference.

If we focus on just men, the black-white difference is one-third of an SD -- a small advantage for black men.

The biggest within-race gap is among blacks: the male advantage over females is three-quarters of an SD.  The graph doesn't show the overall sex difference: it's over half of an SD.

While it's difficult to rate oneself objectively, these self-ratings seem to have some validity.

A big black-white difference in a gene affecting the immune system

This new study looks at genetic differences in black and white women that affect immune system response. The top hit the researchers identified was a genotyped variant rs2814778, which is called the Duffy-null allele. 66% of black women carried the CC variant; 29% had the CT and 5% had the TT genotype. Compare this to white women: 100% of them have the TT genotype.

Now, the "experts" tell us race is not real. For example, they tell us that, say, whites differ among themselves much more than they differ from blacks. But look at this study: there is ZERO variation among whites, black women are fairly uniform too, and where they differ among themselves, it is due to having varied amounts of white ancestry.

It is widely thought that this geographic genetic diversity is the result of a strong positive selection for protection from malaria infection found in West Africa thousands of years ago. Blacks have faced different environments over tens of thousands of years, so what kind of researcher would claim that any genetic differences between blacks and other groups would be minimal or unimportant? Being able to survive malaria is pretty damn important, and the magnitude of the black-white difference in rs2814778 speaks for itself. The kind of researcher who would minimize and dismiss this is a dishonest researcher.

Interracial murder

The Trayvon Martin case is an opportunity to remind people of the reality of interracial crime in the United States.

For the purposes of this analysis, I'll use the term "whites" to refer to whites plus Hispanics since they get lumped together in the data. According to FBI data for 2010,  218 blacks were murdered by whites. Blacks murdered 447 whites. But this is comparing apples and oranges since whites are 6.4 times the size of the black population. If we calculate murders per 10 million in the race-specific population (interracial murder is rare), the rate is 8.82 for whites and 114.89 for blacks. So blacks kill whites at a rate 13 times higher than that of whites killing blacks.

Another way to look at it is multiply the black population by 6.4 to make it as large as whites. If it were as large, 2861 whites would have been murdered compared to the 218 blacks killed by whites.

Blacks, dopamine, and IQ

In this new study in Intelligence, Add Health Study data is used to compare the average level of dopamine-related alleles among students of 36 different middle and high schools. This measure is strongly associated with lower mean IQs (Beta = -.47). The relationship is weakened substantially (to Beta = -.35) when the percent of the school that is black is added as a control in the OLS regression model.

What's the interpretation of this in plain English?  It is this: Some schools have a higher percent of students with genes associated with a low IQ. These schools tend to have a low average IQ. So one important cause of low-IQ schools is having more kids with the bad genes. What the percent black predictor tells us is that: 1) schools with more blacks tend to have more students with the at-risk genes; 2) schools with a higher percent of blacks have lower intelligence levels; and 3) evidently an important reason why some schools are smarter than others is due to having fewer blacks--a group which has a higher prevalence of the bad genes.

This study contradicts the view that: 1) race is a social construction, and 2) the black-white gap in IQ has nothing to do with differences in genes.

Race and speaking up

MIDUS respondents were asked if they agreed with this: "I am not afraid to voice my opinions, even when they are in opposition to the opinions of most people." Here are the means by race/ethnicity:


Mean "speaking up" score


Black 4.87
Amerindian 4.67
Hispanic 4.63
White 4.55*
Asian 3.67


* significantly lower than blacks at 95% confidence level


The gap between blacks and Asians is eight-tenths of a standard deviation--a large difference.  Blacks on top and Asians on the bottom is consistent with my experience in the classroom.

Race and parental alcohol abuse

The MIDUS study asked 2,556 people if either of their parents drank so much that is caused  problems. Here are the estimates by race/ethnicity:

Percent
Mexican American 56.2
Amerindian 38.1
Black 28.7*
White 25.1*
Asian 14.3 

* significantly less than Mexican Americans at the 95% confidence level

The parents of Mex-Ams are four times as likely to have a drinking problem as Asian Americans.

Race, ethnicity, and jail

Participants in the Midlife Development in the United States Study (2004-06) were asked if they were ever in jail (N = 2,167). I calculated the percentages by ethnic background:

Percent
Black 11.9
Amerindian 11.4
Hispanic 10.0
Irish 7.5
Russian 5.9
German 5.4
Polish 5.2
English 4.9
French 4.5
Italian 4.2
Jewish 3.3
Norwegian 2.5
Swedish 2.5
Scottish 2.3

GSS data indicates more Italian arrests, but the ranking is otherwise similar--NAMs on top, and Jews and Scandinavians on the bottom. The gap is wide: 5 times as many blacks as Americans of Scottish descent report having been in jail.

Among high school seniors, Hispanics are least likely to have a job

We saw in a recent post that Hispanics in high school skip classes more than others. Well, it's because they're working, right, which might not be ideal, but hey, at least it's not hanging out on the street corner or something. The only problem is that Hispanic teens attending school are less likely to have a job than whites.

The graph (N = 11,758, 2008 Monitoring the Future) shows that 73 percent of white seniors have some kind of part-time work. This is compared to only 61 percent of Latinos (and 62 percent of blacks). Whites get better grades, even though they spend more time working.

Fatal gun accidents by race

Using self-report data, we haven't gotten very far in identifying racial differences in personality. (Maybe there aren't many differences, but I doubt it.)  Let's try to infer traits from behavior. Fatal gun accidents (FGA) are an interesting first step. People usually kill themselves or others accidentally with guns when they are fooling around. A typical scenario is one drunk guy trying to shoot a beer can off the head of another drunk guy. Rates are highest for teenage males which supports the claim that the behavior measures recklessness.

The graph for the United States 1999-2006 shows that rates of fatal gun accidents vary across racial groups (raw data is available here). Amerindians and blacks have the highest numbers. Whites are in the middle, while rates for Asians are very low.

The incidence of FGAs should be affected by gun ownership prevalence and the popularity of hunting among groups. Amerindian numbers might be elevated a bit, but according to the GSS, they are not really much more likely to own guns or to hunt than whites. Blacks are less likely to hunt and possess firearms, so their high rates of FGAs must be due to more improper handling. According to the GSS, few Asians hunt or have guns, so this might lower their rates. (High rates of drunkenness among Indians might also raise their numbers, but self-report surveys indicate that blacks drink less or no more than whites).

I take this as evidence that, on average, Amerindians and blacks are more reckless than whites and Asians.


UPDATE: I checked, and Hispanics have rates similar to whites, but they are less likely to hunt and own guns (GSS)).

Question of the day

From this knowledge there has arisen that fine dictum of morality so much bandied about by the philosophical crowd that men are everywhere the same, and that having everywhere the same passions and the same vices, it is rather useless to attempt to characterize the different races; which is just about as reasonable as if one were to say that one could not distinguish Peter from James, because each of them has a nose, a mouth, and eyes.

Okay, who wrote this? And no google cheating.

Race differences in personality: I found an interesting British study (Journal of Occup. and Organizational Psych. 2002. 75, 255-76) that examined racial differences in personality. The sample included university students: 324 whites, 56 blacks (Caribbean, African and other), 47 Asians/Indians (Pakistani, Bengladeshi, Indian), and 30 Chinese. Three personality tests were used. They are all self-reports:



Let's just hit the larger differences. Compared to whites, blacks have more managerial potential and are: less sociable, more prudent, and more service-oriented. Compared to whites, Chinese folks have more clerical and managerial potential and are more ambitious, prudent, service-oriented, and geared toward school success. Compared to whites, Asians have more managerial potential and are more ambitious, prudent, and oriented toward school success.






Compared to whites, black are more indecisive, relaxed, emotionally controlled, and more likely to give socially desirable answers. Chinese are more influential, emotionally controlled, active, data rational, achieving, methodical, and are less decisive, relaxed and sociable. Asians are more influential, empathic, innovative, data rational, achieving, and methodical.







Compared to whites, blacks are warmer but have less stamina. Chinese people are more competitive, limelight-seeking, and have less stamina. Asians are more risk-taking.

Some of the results for Chinese and Asian students seem plausible, but many black-white differences don't square with other types of data; in fact, they seem just the opposite. I'd like to see personalities assessed by others--perhaps by spouses, close friends, or experts, with a mixed-race population as the reference group. Or measurement through experimental manipulation would be interesting. Keep in mind that the sample is university students, so the range of personalities is restricted. Also--it would be nice to know how many students are immigrants.

UPDATE: In connection with this post, it is interesting that black Americans rate themselves as the smartest in self-assessments of intelligence, while Asian Americans rate themselves as least intelligent.

Extraversion and lifetime sex partners: Here are the median lifetime sexual partners for different demographic groups ages 30-59 by level of extraversion:


Extraverts
White men 8.0
Black men 10.0
White women 4.0
Black women 3.0


Introverts
White men 5.0
Black men 10.0
White women 3.0
Black women 3.0


Extraversion is associated with more partners for both white men and women, but not for blacks. It seems odd that introverted black men have the same median partners as extraverted black men.

And you can see the typical gender disparity: men are probably exaggerating while women are under-reporting their sexual experience. (Prostitutes probably up the male numbers as well). The gender gap is larger for blacks.

Race, extraversion, and crime: Using last post's measure of extraversion, I wonder if it could help explain racial differences in crime. Ever getting arrested is the dependent variable. Here are the logistic regression coefficients, and I'm comparing two groups at a time:


Logistic regression coefficients

Whites and Asians (N = 3,087)

White 0.64
Extraversion 0.14*


Blacks and Asians (N = 468)

Black 1.46
Extraversion 0.12*

* p < .01, two-tail test


Compared to Asians, whites are more likely to get arrested, but the difference becomes non-significant when extraversion is entered into the equation. The exact same thing happens when we compare blacks and Asians. So the evidence here suggests that extraversion helps explain the racial gap in criminality.

But why, you ask, would extraversion raise the risk of breaking the law? Compared to, say, the guy who likes to spend his time blogging alone, people who love to be out doing things with other people are more likely to find themselves in situations conducive to crime. And they would be less likely to back away from criminal opportunities.

Another look at extraversion: I took another approach to measuring racial and ethnic differences in extraversion by looking for GSS behavioral questions. Respondents were asked how often they: 1) visit neighbors, 2) visit friends outside of the neighborhood, and 3) go to a bar. Answers varied from "almost every night" (1) to "never" (7).

These items were all positively correlated with each other, but Cronbach's alpha is only .45. I summed them together to create an introversion score and limited the analysis to those ages 30-49 out of concern that the groups might have different average ages and that going out, especially to bars, is influenced by age. Fortunately, the sample size is large (N = 9,528).


Introversion score

White 13.80
Black 13.91
Mexican 14.84*
Chinese 14.79*
Filipino 14.57
American Indian 15.04*

*p < .05, two-tail test, compared with whites.


Consistent with the earlier analysis, blacks and white do not differ, but Mexican Americans are significantly more introverted. The large sample size allowed me to look at other minority groups, and you can see that Chinese Americans and American Indians are less outgoing. The difference between whites and Filipinos fell just short of statistical significance. I went ahead and calculated the effect sizes too:


Effect size

Mexican .30
Chinese .29
Filipino .22
American Indian .36


These numbers indicate small to moderate ethnic/racial differences. Keep in mind that broad heritability for extraversion is around 0.5.

Race differences in IQ: For those of you who don't have the time to read Rushton and Jensen's 50 page review of Nisbitt's Intelligence and How to Get It, I outline the points I found interesting below. It's funny how the research literature is this rich and advanced (compared to, say, educational research work in the area) but the typical mainstream position is that IQ research is way too crude and undeveloped to inform policy. One frickin' doll study helped decide Brown v. Board of Education, but a century of research by psychometricians shouldn't even give us pause.


1. World IQ is 90: East Asian cluster is 105, Europeans are 100, Inuit-Eskimos 91, SE Asians 87, Native Americans 87, Pacific Islanders 85, South Asians and North Africans 84, sub-Saharan Africans 67, Australian aborigines 62, and Kalahari Bushmen and Congo Pygmies 54.

2. Flynn effect: IQ has increased in the developed world in the last 50 years, and the black-white gap has narrowed a bit. The gain is not on g which is in large part genetic.

3. Research was cited suggesting that the IQ gap is larger than 15 points because a disproportionate number of low IQ blacks do not make it into samples.

4. Research shows that the Flynn effect is qualitatively different from black-white differences in regards to the g factor.

5. A test's g loading is the best predictor of biological measures of heritability coefficients determined from twin studies, inbreeding depression scores calculated in children of cousin-marriages, brain evoked potentials, brain pH levels, brain glucose metabolism, as well as nerve conduction velocity and reaction time measures. The correlations argue strongly for the biological, as opposed to the mere statistical reality of g.

6. Across 149 psychometric tests, the correlation between g-loading score and black-white difference is .62. This contradicts the culture-only theory.

7. Research indicates that there is no factor lowering IQs that is specific to blacks.

8. The correlation between several measures of reaction times and IQ is .6. Asians are faster than whites, and whites are faster than blacks. The differences are not due to motivation: while Asians and whites react faster, blacks move faster.

9. Heritability is lowest early in life and rises to 60-80% by adulthood. Shared environment falls to 0% by adulthood. Adoption can raise the IQ of a child quite a bit, but g remains like that of the biological parents.

10. The model that best fits data from a sample of black and white twins is 50/50 genes and environment.

11. Non-hereditarians like Robert Sternberg get results like those reported by Lynn when they administer IQ tests to sub-Saharan Africans. Even after training kids how to solve problems like those on the test, means were under 80. Teaching to the test denudes the test of its g loading. Studies of Africans who have won math and science competitions put mean IQ at 100. African IQs predict academic and job performance as well as for non-Africans. With a white mean set at a mental age of 16, an African mean of 70 implies a mental age of 11. Eleven year olds are not retarded. (I babysat and had a paper route at age 10 and started driving and hunting when I was 13.)

12. In a MRI study, Posthuma et al. found heritabilities of 82 percent for whole-brain gray matter volume, 87 percent for whole-brain white matter volume, and 86 percent for IQ.

13. Smaller black brain sizes do not appear to be due to prenatal or perinatal conditions. In a study of fetuses, Schultz found that from the 9th week of intrauterine life, 165 Black fetuses averaged a smaller brain case than did 455 White fetuses. The difference became more prominent over the course of fetal development.

14. In meta-analyses of general population samples on the Progressive Matrices, Lynn and Irwing found no sex difference in IQ among children aged 6- to 14-years but a male advantage from 15-years through old age, with the male advantage by adulthood being equivalent to about 4.6 IQ points.

15. Winick et al. studied 141 Korean children malnourished-in-infancy and then adopted as infants by American families. They found that by 10 years of age the children exceeded the national average in IQ and achievement scores: A severely-malnourished group obtained a mean IQ of 102; a moderately-nourished group obtained a mean IQ of 106; and an adequately-nourished group obtained a mean IQ of 112.

16. Regression to the mean effects support hereditarian theory. The IQs of the siblings of blacks with 120 IQs is 100; for their white counterparts it is 110. It is the same on the low end: when blacks and whites are matched at an IQ of 70, black siblings average a 78; the white siblings, an 85. This helps to explain why Black children born to well-educated, affluent, parents have test scores 2 to 4 points lower than do White children born to poorly-educated, impoverished parents.