Are survey responses to HBD questions honest?

In yesterday's post on the popularity of HBD, readers raised the reasonable concern that a person might be unwilling to tell an interviewer face-to-face (that's how the GSS is done) that he thinks that blacks are innately less intelligent than whites.

Fortunately, the GSS allows us to vary interview conditions in two ways: 1) was it a face-to-face interview or one done over the phone; and 2) was the interviewer white or black?

Here are the percentages who agreed that blacks are innately less intelligent:

Percent answering yes--white sample (N = 2,044)

White interviewer over the phone 3.4
White interviewer in home 8.5
Black interviewer over the phone 7.7
Black interviewer in home 4.4

Not much evidence here for the view that people are responding to the conditions of the interview. Whites should feel most comfortable giving the pro-HBD response to a white interviewer over the phone, but this condition gets the fewest affirmative responses. Myself, I would have the hardest time expressing my position to a black person, but the second-highest rate of positive response was given to black interviewers over the phone.

On the other hand, interviews over the phone are uncommon, so cell sizes are low. If we focus only on the in-home interviews, there is more support for the idea that people tend to give socially approved answers. 8.5 percent of respondents answered yes to a white interviewer at home, compared to only 4.4 percent of those responding to a black person. The difference, however, is not statistically significant. Whites might be responding to conditions, but keep in mind that only 9 percent of interviews were given by blacks.  

Race and substance abuse

According to self-report data, blacks use less alcohol and drugs than whites. Black students in my classes have argued that this is evidence that the criminal justice system is biased since blacks are arrested at high rates for drug-related offenses. (Let's leave aside the question of dealing, and that research suggests that blacks under-report deviant behavior). 

Treatment data, however, reveal that blacks are more likely to abuse alcohol and drugs. In 2008, 21.3% of Americans admitted for treatment of a drug or alcohol problem were black, although they are only 13% of Americans (and are less able to afford treatment). Black admissions are particularly high for crack and marijuana, but blacks are overrepresented even among those admitted for alcohol (16.8 percent of the total).

There is no Silent but Sensible HBD Majority

Steve Sailer wonders if there is a Silent but Sensible Majority out there when it comes to HBD-related issues like racial differences in academic performance. The GSS convinces me that the answer is definitely no.

Beginning in 1977, survey particpants were asked: "On the average, blacks have worse jobs, income, and housing than white people. Do you think these differences are because most blacks have less in-born ability to learn?"
 
Here are the percentages who answered yes by year:
 

There were quite a few race realists in the 70s, but the number has dropped to around 10 percent. Things have only gotten worse post-Bell Curve (1994).
 
Yes, but it's educated opinion that really matters. Perhaps they have more sense (the graphs shows 4-year college grads or higher):
 

Educated people didn't believe blacks differ innately three decades ago; even fewer do now.
 
Verdict: public opinion doesn't follow the data in the least. It is irrational; it follows fads.

Atheism not spreading

After months of therapy, I am finally able to return to John Derbyshire's We are Doomed (buy it, bitches) to test one of his ideas. He contends that 9/11 has unleased an atheist movement which some nominal believers have responded to affirmatively. 

So is the number of atheists growing? According to the GSS, the answer so far is no. Taking the six surveys between 1988 and 2000 that ask about belief in God (N = 8,027), 2.5 percent of Americans do not believe there is a God. For the two post-9/11 surveys--2006 and 2008 (N = 4,971)--2.5 percent are atheists. Holding steady.  

Gay flexibility

We all know that homosexuality is fixed, right?  The conventional wisdom is that homosexuals can no more have straight sex than we heteros can have gay sex. Myself, I couldn't have sex with a man if a gun were put to my head. Gay men are the same, right--just on the other side.

Then why are the numbers so unequal? Using GSS data, we can define homosexuals as men who say they only have sex with men, and straight men defined as those who say they have sex exclusively with women. Only 3.6% of heterosexual men so defined have ever had sex with another men. It sounds like a gun to the head situation. Not so for gay men. Fifty percent of homosexuals have had female partners. Here's the breakdown (sample size = 244):

Number of female sex partners since age 18--percent:

Zero 50.0
One 12.7
Two 7.0
Three 7.8
Four 2.0
Five or more 20.5

I could imagine a gay guy, feeling society's pressure, trying to force himself to have sex with a woman once, then realizing that the idea was simply impossible. But more than 37 percent of gay men have had more than one female sex partner. More than 20 percent have had sex with five or more women! The median number for straight men is only 6 or 7 women.

I can hear you responding that these men have sex with women because society pressures them to. But you are simply making my point. The whole country could gather around me and bark orders to have sex with a man, or else. I'd have to take the "or else." Fifty percent of gay men choose sex with a woman. There's substantial flexibility there, contrary to the PC version of homosexuality.  

Waste of time

In the late 80s and early 90s, GSS participants were asked if they have ever served in the military. Slightly more than 20 percent of gay men (of all ages) said they had. This is a much greater estimate than we would get for young gay men now, but let's stick with it. According to CNN, only one-sixth of gay enlisted men plan to come out now that the "don't ask, don't tell" policy will be ended. So multiplying the two percentages gives us 3 percent of gay men who will ever make use of the new policy. According to research, around 3 percent of all men are gay, so if we multiply 3 percent by 3 percent, we get basically 1 out of 1,000 men who will ever be affected by the change in policy. The number would be probably be even smaller for women since there are half as many lesbians as gay men (according to the GSS, 21.6% of lesbians say they have served). 

I'm pleased that the Democrats feel so good about themselves after getting the policy changed, because otherwise it kinda seems like a waste of time. 

Personally, I wish lawmakers would focus on issues slightly more important; for instance, whether the hundreds of thousands of embryos and fetuses that are killed every year are immature humans or not.      

Conservatives give more to non-religious charities

Lawrence O'Donnell responds here to Ann Coulter's empirical claim that conservatives give more in charity than liberals by praising the generosity of his audience and asking Ann to buy some school desks for kids in Malawi.

It's pretty common knowledge that conservatives give more (O'Donnell didn't dispute the claim) but maybe it's due to religious charitable giving? 

In 1998, the GSS asked people how much they gave in non-religious charity last year. Here are the means in dollars by choice for President in 1996 (sample size = 787):

Mean dollars donated

Dole 399
Clinton 387

Even setting aside money donated to churches, conservatives give more.

I, for one, look forward to the end of "don't ask, don't tell" for conservatives in higher education.

How to reduce teen pregnancy

Want to reduce teen pregnancy? Then make it hard to get an abortion. Using state-level data, Marshall Medoff finds that raising the cost of abortions, Medicaid funding restrictions, and informed consent laws lower rates of teenage pregnancy. The author concludes that youths are more careful about sex if abortions are harder to get.  

Ways to reduce abortion

This study shows that policies short of prohibition of abortion can lower rates. In a state-level analysis, the author found that parental involvement laws, regulation of abortion providers, and Medicaid funding restrictions in particular reduce abortion levels.  

Men with more friends at work have more kids

In his book Men in Groups, Lionel Tiger (damn, I wish I could have cool animals for my first and last names--Tiger Woods, eat your heart out) hypothesized that men who are attached to male groups will have greater reproductive success than isolated men. Even if a guy is not dominant, being a part of a group will enhance his attractiveness since he is perceived to have greater access to resources and protection.

Men who participated in the GSS were asked: 1) the number of close friends they have at work, and 2) the number of children they have (n = 136). The correlation between the two for guys who are old enough to have completed their families but are not yet retirement age (45-64) is .19. It is small but statistically significant. Men with more friends at work do tend to have more kids. (Work is not generally an all-male group, but it does tend to be dominated by men and is thus "male" in that sense). 

UPDATE: I thought men with many friends at work might earn more due to connections which, in turn, would lead them to have more kids since they could better afford them, but the impact of friends on family size is not reduced when income is entered into an OLS regression model.  

Predictors of infidelity

What predicts infidelity?  Here are the results listed for men and women separately (GSS data):

Logistic regression coefficients--men (sample size = 2,379)

Age .01
Black .77
Other race .33
Education .01
Conservatism -.12
Church attendance -.08
IQ .09
Income -.01

All relationships are statistically significant except for Other Race and Education. Male cheaters tend to be: older, black, liberal, not religious, smart, and low-income.


Logistic regression coefficients--women (sample size = 3,315)


Age -.01
Black .77
Other race .09
Education -.05
Conservatism -.12
Church attendance -.11
IQ .19
Income -.01

Other Race is the only independent variable that is unrelated to cheating. Female cheaters are likely to be: young, black, less educated, liberal, not religious, smart, and low-income. The main difference between men and women is age: older men have higher rates of cheating than younger men, but younger women have a higher probability of infidelity than older women. Also--education predicts a faithful wife, but is unrelated to fidelity among husbands.

What's the underlying theme of these predictors? Unconventionality, perhaps? You might want to think twice about committing to an unconventional person.

Elites are ruining the country

Just when I begin to develop some hope that America might be able to survive the ethnic diversification of the country, some Lefty moron throws cold water. This Asian UCLA professor tells the crowd of DREAM Act activists that they will someday replace the old white men in Congress. So did the crowd, in defense of a color blind America, boo him? On the contrary: the quiescent audience burst into cheers as the words "old white men" were uttered.

My approach on this blog is limited because surveys capture the attitudes of the average person. But history is driven by elites, and all too many of them want whites to take their rightful place at the bottom of American society. All this talk of equality and justice for all is just a smokescreen. It's about power and domination for these people, pure and simple.

Many ordinary people might have benign intentions, but they are easily manipulated by demogogues. Race-baiting and racial politics as methods to advance politically are such powerful temptations, I'm afraid the country is going to become increasingly dysfunctional. 

Predictors of patriotism

I was interested in predicting patriotism. GSS participants were asked how close they feel to America. I recoded the variable so that "close" and "very close" equal 1 and "not very close" and "not close at all" equal 0.  Here are the logistic regression coefficients:

Logistic regression coefficients (sample size = 2,441)

Male .49
Age .03
Black -.86
Other race -.29
Born in America .37
Jewish 1.45
Church attendance .09
Mid-Atlantic resident -.32

All of the relationships are statistically significant except for those of some other race. Keep in mind that the coefficients are unstandardized so they cannot be compared for size. Age, race, gender, and church attendance have the strongest impact. The profile of a person who loves America goes like this: white, male, older, native-born, Jewish, religious, and does not live in the Mid-Atlantic region (NY, NJ, PA). (Education, income, IQ, job prestige, and political orientation were found to be unrelated to feeling close to the country).

Homosexuals love America as much as straights

We learned recently that the servicemen who gave Wikileaks a zillion secret documents is gay. Ann Coulter suggests that homosexuality might be a risk factor for treason. I guess the idea behind it is that gays are more likely to hate America because of its homophobia, and are thus more likely to betray it. 

The General Social Survey asked people how close they feel to America. The percent of heterosexuals who answered "not very close" or "not close at all" is 16.2. For homosexuals and bisexuals of both sexes (n = 52) it's 19.2 percent. Basically the same.

Gallup: Little evidence of an American desire to soak the rich

Here's a new Gallup poll on attitudes toward extending tax cuts to all Americans (and unemployment benefits). Where's the evidence that ordinary Americans are itching to stick it to the rich?  Sixty-six percent of all Americans wants the tax cuts extended for all Americans; a majority (52 percent) of Democrats favor the extension.

The country is split (44% vs. 40%) when respondents are given explicitly the choice to raise taxes on the wealthy. Obama is wrong to claim that the people are on the Dems' side on this issue. Based on my analysis of GSS data, many of those who favor it are black and Hispanic.  

Mean IQ of Americans with advanced degrees has dropped almost nine points

When GSS data revealed that the mean IQs of college graduates dropped almost nine points over the past four decades, I assumed that grad schools, by contrast, had not lowered standards so much. I was wrong. Here are the means by decade for Americans in their thirties with an advanced degree:

Mean IQ of Americans with advanced degrees (sample size = 446)

1970s 115.7
1980s 112.7
1990s 110.0
2000s 107.1*

* significantly lower than 1970s

The mean IQ dropped 8.6 points over the four decades--just as much as for undergraduates. College degrees across the board do not mean what they used to.

Trailer Park America

Take a guess:  What is the average IQ of a white dude who lives in a trailer? According to the GSS, it's 93.1. It's 85.5 if you're black and 88.6 if you're Hispanic.

Whites who live in mobile homes are not average whites in terms of IQ. They are lower- or working-class people. Blacks and Hispanics, on the other hand, who live in trailers have IQs typical for their respective groups.

This helps give us an image of the kind of America we are becoming as the country Latinizes. Instead of a traditional America which has been middle-class, we're headed toward Trailer Park America.

Real punishment

I was looking forward to Pfc. Bradley Manning spending a good long time behind bars. Turns out he's gay, so prison is gonna be one big party. For him to really feel his punishment, the warden needs to ban all Lady Gaga CDs.

While I'm at it, let's remember today the America-loving men who, as my grandma would have put it, eventually kicked Jap ass. My maternal grandfather was one of those men, God rest his soul.

A bright side to AIDS

According to this study of U.S. states, the perceived threat of HIV lowered the rate of unwanted pregnancies and abortions.  The PC strategy of scaring the hell out of heterosexuals turned out to be a good thing. 

Closeness to ovulation predicted voting for Obama

In a recent study of young white women, researchers found that subjects were more likely to plan to vote for Obama for president if asked around the time when they were ovulating. This was only true, however, if the women perceived Obama to be lighter-skinned than he actually is. If they perceived him to be darker than he is, closeness to ovulation lowered the chance of voting for him.

The authors interpreted this to mean that women become sensitive to men as potential mates around the time when they are most fertile; if a man seems dangerous (in this study blackness is thought to be associated with dangerousness) they are repelled by him out of fear of sexual assault; if a man seems powerful, attractive, and safe, they will be attracted to him. The study also found that these tendencies were particularly strong among conservative women (liberal women planned to vote for Obama, regardless). 

The authors also suggest that, since Obama won the election, many whites do not perceive him as black. They imply that people look at more than skin color when determining race, and Obama's education, career, demeanor, Waspishness, and restraint communicate whiteness.

The message here for Republicans for 2012 is that a few percentage points might be gained among women if the nominee is a powerful, attractive yet safe candidate. Of the current prospective men, I suspect that Romney best fits the description, which is not saying much. Most of the candidates seem weak or unattractive: politics attracts many weasels. 

On the morality of remittances

A good point by Hanson on the morality of remittances:

It is deemed noble to send billions of dollars back to families and friends struggling in Latin America. But how is such a considerable loss of income made up? Are American taxpayers supposed to step in to subsidize increased social services so that illegal immigrants can afford to send billions of dollars back across the border? What is the morality of that equation in times of recession? Shouldn't illegal immigrants at least try to buy health insurance before sending cash back to Mexico?

Almost half of Muslims say terrorism may be justified

World Values Survey respondents in 13 European countries and the United States were asked the following question: "Terrorism is everyday news. In principle, most people are against it, but there is still room for differences of opinion. Which of these two statements do you tend to agree with?  A) There may be certain circumstances where terrorism is justified; B) Terrorism for whatever motive must always be condemned."

Of those who answered the question, here are the percentages who responded "A" by religion:

Percent saying terrorism may be justified (sample size = 14,566) 

Muslims 46.2
Hindu 26.3
Jewish 16.6
Protestant 15.3

Total sample 14.9

Catholic 14.7
Non-denominational 12.5
Other 11.1

Almost half of Muslims feel that terrorism can be acceptable--a rate that is more than three times that of the total sample. Sympathy for terrorists appears to be common among Muslims living in Western countries. (Warning: there are only 21 Muslims in the sample).