Yesterday Michelle decided to put up a post with her own analysis of her ADMIXTURE results. With that in mind, I thought I’d revisit some results from my parents. After many runs of ADMIXTURE, both by myself and Zack, some consistent differences seem to crop up. To review, one of the big surprises from genotyping my parents is that both of them have about the same “East Asian” element of ancestry which is very distinctive from the conventional South Asian mix. Because both of my parents lack any oral history of recent admixture I posited that this element may be a uniform substrate common among eastern Bengalis, and that it was absorbed during the initial period of settlement and demographic expansion on the frontier in the period between 1000-1500 A.D. By analogy, low levels of Amerindian admixture persist across Brazilians, and African admixture among Mexicans, but because the admixture dates back several hundred years it does not seem to have percolated down to the present in oral history (though some old stock Brazilians of predominantly Portuguese origin have been able to infer Amerindian ancestry by looking at the church marriage records of their ancestors, and adducing that some women were natives due to common baptismal names given to such converts).

Since ADMIXTURE is sensitive to the genetic variance you throw into it in extracting out patterns, I created two pools with my parents in it. One was predominantly West Eurasian, and another was predominantly East Eurasian. In both samples my parents were Bengali A (father) and Bengali B (mother), and I included in the Gujarati_B and Pathan South Asian populations. Gujarati_B because it seems particularly South Asian, and therefore informative. The Pathan sample has less African admixture than the Sindhis or Makranis, and is not so isolated as the Kalash, Brahui, Burusho, and Baloch. For the East Eurasian sample I included Sardinians as the West Eurasian outgroup, while for the West Eurasians I included Japanese as the outgroup. Finally, I pruned the markers down to 65,000 SNPs. Below I report K = 6, as cross-validation determined that to be the optimal value for the number of populations.

(more…)

In my post earlier in the week I mentioned the possible relationship between attitudes toward evolution and the causes and likelihood of Global Warming. I haven’t seen any survey data myself relating the two, so naturally I wanted to poke into the General Social Survey. Two variables of interest showed up, both from 2006:

1) GWSCI, “Understanding of causes of Global Warming by environmental scientists.” A five-point scale, from understanding “Very well” (1) to “Not at all” (5).

2) SCIAGRGW, “Extent of agreement among environmental scientists.” A five-point scale, from “Near complete agreement” (1) to “No agreement at all” (5).

I paired these up against EVOLVED, which is a simple True vs. False answer in relation the question as to whether “Human beings developed from animals.”

Tables below.

(more…)

A quick follow-up to my previous post which points to the data that women tend to be more race-conscious in dating than men. There’s a variable in the GSS which asks if you support a ban on interracial marriage, RACMAR. Here’s the question itself:

Do you think there should be laws against marriages between (Negroes/Blacks/African-Americans) and whites?

There isn’t much surprising in the results for this variable. It was asked between 1972 and 2002, and support for a ban on interracial marriages dropped over time. Whites, old people, conservatives, and less educated people, tended to support these bans, as well as Southerners. But what about men vs. women? I’ve never actually looked at that. I limited the sample to whites; the number of blacks in the sample is small and wouldn’t alter the result, but I figured I’d control for race anyway. Support for such laws is in the 35-40% range for whites in 1972, before dropping off to 5-15% in 2002.

Here’s the trendline broken down by sex:

(more…)

I noticed that Jim DeMint has said some controversial things about the demographic criteria of teachers:

Sen. Jim DeMint (R-S.C.) says that even though “no one” came to his defense in 2004 after he said that gay people and unwed mothers should be banned from teaching, “everyone” quietly told him that he shouldn’t back down from his position.

…

The Spartanberg Herald-Journal described the comments this way: “DeMint said if someone is openly homosexual, they shouldn’t be teaching in the classroom and he holds the same position on an unmarried woman who’s sleeping with her boyfriend — she shouldn’t be in the classroom.” DeMint did not apparently state his position on whether sexually active unmarried male teachers should be similarly removed from classrooms.

My interest was piqued because there are questions in the GSS about allowing gays to teach. We can see how many people in the country agree with DeMint. How the proportion has changed over the years, and also the demographic correlates of variation in attitudes. Additionally, I wanted to compare attitudes to allowing homosexuals to teach with allowing anti-religionists and racists to teach. First, over time:

(more…)

Christine O’Donnell has said a lot of kooky things. Right now people are focusing on her Creationism. Though I’m obviously not a Creationist I think mocking someone for this belief in a political context is somewhat strange: the survey literature is pretty robust that Americans are split down the middle on opinions about evolution. More specifically most of the polling shows that around ~50% of Americans tend to reject the validity of evolutionary theory when asked. This is what I like to call a broad but shallow belief; for the vast majority of Americans attitudes about evolution are really just cultural markers, not stances of deep feeling or impact. One point of evidence for this conjecture is that polling on evolution is easy to massage through framing. Another is that Republican candidates for the presidency do not invariably hew to a Creationist line despite the likelihood that the majority of primary voters are Creationist. Politicians react to incentives, and my own hunch is that there isn’t a strong push from the Christian Right on evolution as there is on abortion or gay marriage.

I’ve posted plenty on how Creationists are more female, less intelligent, more conservative, more likely to be ethnic minorities, less educated, etc. Here I want to put the spotlight parameters which might shed some light on the O’Donnell race. Is her kooky opinion on evolution a particular liability in Mid-Atlantic Delaware? Are Creationists less likely to vote? And what are the regional breakdowns which might explain the bi-coastal shock and amusement at O’Donnell’s opinions?

(more…)

Sometimes I run into things in the GSS which just don’t fit expectations. On occasion the results are so weird or unexpected I check my coding over and over. Or, I have a suspicion that something was input incorrectly. This is one of those cases. As often happens a comment was made as to the acceptance of biological explanations for behavior, and their political correlates. I decided to poke around and confirm what I knew: that liberals are more environmentalist than conservatives, who are more hereditarian. This is not what I found!

The gene related questions have the following form:

… what percent of the person’s behavior you think is influenced by the genes they inherit, and what percent is influenced by their learning and experience. After each question, type the number of the box that comes closest to your answer. Remember, the higher the number, the more you think the behavior is influenced by learning and experience; the lower the number, the more you think it is influenced by genes

Each respondent could select from 21 values, from 1 to 21, with 1 = 100% genetic, 21 = 0% genetic, at 5% increments. So 3 = 90% genetic. This isn’t technically correct as an understanding of heritability, but I think it gets across the intuitions of heritability. All the questions were asked in 2004. They were:

- GENENVO1: Carol is a substantially overweight White woman. She has lost weight in the past but always gains it back again.

- GENENVO2: David is an Asian man who drinks enough alcohol to become drunk several times a week. Often he can’t remember what happened during these drinking episodes.

- GENENVO3: Felicia is a very kind Hispanic woman. She never has anything bad to say about anybody, and can be counted on to help others.

- GENENVO4: George is a Black man who’s a good all-around athlete. He was on the high school varsity swim team and still works out five times a week.

(more…)

I was having a touch of insomnia a few days ago, and wasn’t alert enough to do anything intellectually challenging, so I decided to poke around the General Social Survey. I found an interesting variable, POSTLIFE, which asks people if they believed in life after death. I decided to cross-check that against those who were atheists and agnostics, and specifically look at the distribution of WORDSUM scores of those who did, and didn’t, believe in life after death. My hunch before I checked was this: those who believe in life after death despite not believing in the existence of God are going to be less intelligent than those who don’t.

My reasoning was that it was close to philosophically incoherent to reject supernatural agents, but then accept some post-material existence. I know that this is actually not necessarily philosophically incoherent. Asian religious traditions have long had a strand which accepts both immortality of consciousness as well as agnosticism or atheism in relation to supernatural agents, gods. And, there are some secular Western philosophers who make an analytic case for the afterlife despite their lack of belief in the supernatural. But most people are not deeply involved in the philosophical literature on the afterlife, or, Jains.* Rather, I think those who are atheists or agnostics, and, who accept an afterlife, are relying on intuition and not following through deductively on the inferences from their avowed axioms. In other words, I believed they’d be likely to be less intelligent, and habitually make less use of analytic modes of thinking. To double-check on the thesis that the less intelligent are more likely to hold inconsistent views I also looked at self-identified liberals, conservatives, Democrats and Republicans, on specific policy issues and their relationship to intelligence.

Before we get to the tables, some methodology. I used WORDSUM, which has a 0.7 correlation with IQ. I recoded WORDSUM so that in terms of intelligence you have the following classes:

Low (0-3) – 11% of the sample, 6% non-Hispanic whites
Below Average (4-5) – 27%, 25% non-Hispanic whites
Average (6) – 22%, 21% non-Hispanic whites
Above Average (7-8) – 27%, 34% non-Hispanic whites
High (9-10), 13%, 15% non-Hispanic whites

For the religion related questions I used the whole GSS data set. For the politics related questions I limited to non-Hispanic whites, which also constrains the data set to the 2000s (politics tends to be racially polarized more than religion, so I wanted to remove the racial variable). The rows in each column below add up to 100%, so what you’re seeing are the intelligence distributions within each class. So, if you see 22% in the Low category, that means that the class has twice as many people in that category than the general population. Please note that ~20% of the population rejects an afterlife, a higher proportion than those who are irreligious, or atheists or agnostics. A substantial number of religious people don’t believe in an afterlife, just as a substantial proportion of atheists and agnostics do.

(more…)

Yesterday I pointed to an OKCupid study on personals ads on their network. I didn’t highlight the fact that the analysis of the ads seem to suggest that secular people have more sophisticated prose, and that of those who claim a religious affiliation the less strongly committed tend to be more sophisticated. The blog Political Math is skeptical of the findings, and points to some evidence that the analytic technique might be biased against the type of things religious people are prone to say.

That’s fine, but there are two major issues with this sort of objection. First, there are clear confounds here which could be responsible for this correlation. Both blacks and Latinos seem less sophisticated, while Asians and Indians seem more so. The overwhelming majority of blacks are Protestants, and a majority of Latinos are Roman Catholic, and both of these religious groups rank low on sophistication according to OKCupid. Conversely, Hindus and Buddhists rank high. Though most Asian Americans are not Buddhists, around half of American Buddhists are Asian American. And the vast majority of American Hindus are Indian. As I noted yesterday the results correlate rather well with educational attainment by group in the USA.

(more…)

In my post on American fertility rates by racial group Mike Keesey asks: ‘It’d also be interesting to see what’s going on within “non-Hispanic whites”.’ One can explore this question in the GSS. Let’s look at ancestry group (e.g., German, French, etc.), religion, belief in God, political ideology, intelligence and education, for non-Hispanic whites. The data is limited to the 2000s, and I also constrained to those age 45 and up. Then I looked at the “CHILDS” variable, which asks the respondent how many children they have. Taking the mean of this value gives us a sense of the rank order in fertility. Note that this is not total fertility rate. That should be clear from the values being well above 2 in most cases. Additionally, I recombined some categories, so that “British” is the amalgamation of English, Scottish and Welsh ancestry. The “Irish” class almost certainly includes both Scotch-Irish (doing the regional and religious breakdown this seems obvious), and the Irish without modifiers. For intelligence I used “WORDSUM”. The variables I input into the GSS can be found at the bottom of the post so you can replicate.

(more…)

In the post below on the prolific nature of the Kennedy clan some commenters were curious as to the general socioeconomic slant of Irish Catholics. The GSS has a variable ETHNIC which asks which nation an individual’s ancestors came from. Combine that with RELIG, and you can figure out how Irish Catholics stack up nationally. While I was looking at Irish Catholics I thought I would look at whites from various nations. I decided to exclude Jews from the analysis because I think there’s a big difference between Polish Catholics and Polish Jews socioeconomically and we’d lose information aggregating. Further, I constrained the sample to non-Hispanic whites. To look at socioeconomic index I focused on the SEI variable. Here’s how SEI is calculated:

SEI scores were originally calculated by Otis Dudley Duncan based on NORC’s 1947 North-Hatt prestige study and the 1950 U.S. Census. Duncan regressed prestige scores for 45 occupational titles on education and income to produce weights that would predict prestige. This algorithm was then used to calculate SEI scores for all occupational categories employed in the 1950 Census classification of occupations. Similar procedures have been used to produce SEI scores based on later NORC prestige studies and censuses.

Here are some values for reference:

Jewish = 62
White non-Hispanic = 51
Hispanic = 43
Black = 42
Only High School Education = 43
Bachelor’s Degree = 63

I’ve crossed the ethnic groups with religion & region (RELIG & REGION):

(more…)

One of the trends that makes me less pessimistic about the inevitability of an idiocratic end-point to technological civilization is that it seems young Americans are more likely to accept evolution than earlier age cohorts.  The EVOLVED variable asks whether one believes that “Human beings, as we know them today, developed from earlier species of animal.” It was asked in 2004 and 2008, and its response is dichotomous between true and false. The favorable age trend I was aware of, but almost randomly I decided to control for some demographic variables, and I stumbled onto something which surprised me a bit, but in hindsight shouldn’t have: much of the greater acceptance of evolution among the youth has to do with a closing of the sex gap between men and women. Traditionally women have been more religious and Creationist in their inclinations, but far less so in Gen Y. Chart below of EVOLVED.

(more…)

Matthew Yglesias says:

I only wish the same level of scrutiny were applied to assertions about whether the public is “liberal” or “conservative” where I believe there’s strong circumstantial evidence that many people just don’t understand these terms in the way political and media professionals understand them. For example, when you break these things out by race you find that whites are more “liberal” than blacks, which simply doesn’t describe either voting behavior or views on issues correctly.

I am sympathetic to Matthew Yglesias’ point, but looking at the General Social Survey he does seem incorrect in regards to his assertion that whites are more liberal than blacks. Rather, blacks seem a bit more liberal than whites, though not nearly as liberal as their voting for the Democratic party would suggest. But most humans not are very intelligent, and Americans are humans, so there’s a good chance that they don’t really know what liberal and conservative mean (here my liberal readers will observe that there’s been nearly two generations of usage of the term “liberal” as an insult in American politics to the point where the public may be unaware of what liberalism even entails aside from sex, drugs and rock ‘n’ roll).

To see if there is a strong relationship between avowed political ideology and particular policy positions I decided to look at the GSS. I limited the sample to the last 10 years of the GSS, 1998-2008. I focused on the POLVIEWS variable, which asks people their own political ideology, from a spectrum of very liberal to very conservative. I want to focus on those who admit to being rather ideological, so I’ll focus on the very liberal and conservative, and those who assert that they’re just liberal or conservative without qualifiers. This excludes those who are slightly liberal or conservative, as well as moderates. The sample size then is ~2,000 for liberals, and ~3,000 for conservatives at most most (some of the cross-tabs have way smaller N’s as only a subset were asked both sets of questions).

(more…)

I’ve mentioned this before, but I thought it would be useful to repeat again. Many of my social science related posts use Berkeley’s web interface with the General Social Survey. Regularly people ask me in the comments details as to the variables, or a more explicit elaboration of the methods. First, this is a weblog, not a venue for me to publish scholarly papers. Most of the GSS related posts are meant to be “quick & dirty,” and stimulate further exploration by readers. Unfortunately follow ups rarely happen. One can speculate why, but that’s how it is. Nevertheless, I thought I would repeat really quickly how to use the GSS in a basic fashion.

First, here’s the URL:
http://sda.berkeley.edu/cgi-bin/hsda?harcsda+gss08

This is the database from 1972 to 2008. You’ll meet a screen like this:

(more…)

A comment below about intelligent people who believe in dumb ideas made me want to revisit the Creationism demographics in the GSS. More on point I wanted to look at the relationship between IQ and Creationism crossed with demographic variables. I used the WORDSUM variable as a proxy for IQ (the correlation is ~0.70). WORDSUM scores range from 0 to 10; 10 being a perfect and 0 being not so perfect. To get a sense of the range, here are mean WORDSUM scores by highest degree attained, constrained for the years 2004 and later:

I decided to limit the year to 2004 and later because to explore Creationism I want to use the variable EVOLVED, which was asked in 2004 and 2008. I selected EVOLVED because the sample size was not that small, nearly 1,500, and, the response is dichotomous. Here’s what EVOLVED asked:

Human beings, as we know them today, developed from earlier species of animals. (Is that true or false?)

Querying Americans about human descent from animals primes them to be a bit on the Creationist side. True and false come at at about 50:50 for the above question. Below is a table where the columns have mean WORDSUM scores for non-Creationists and Creationists, and the rows indicate the particular demographic. I have put in bold those variables where the horizontally adjacent cells are outside each other’s 95% confidence interval. Additionally I constrained the sample to non-Hispanic whites (so the N is closer to 1,350).

(more…)

In the post below on Bryan Caplan’s arguments for why one should have more children there was an “interesting” comment:

As if we’re harmless little creatures at one with our environment and put no toll on the balance of nature around us. Funny how we humans act like mindless rabbits and lemmings and put the sole unintelligent directive of our DNA as the mouth of god. Men most interestingly in power or self described intellectuals after sitting around picking belly lint and jerking off in praise of their penises find clever monkey justifications (patriarchal religions mostly) for more more more babies and women must be subservient to male sexual needs and demands of more babies. See a huge male god said so.

Funny how women mostly never jump on the soapbox bandwagon of wanting to pop out tons of kids, just male spermatozoa fed rants formed by the human male organism to insist his natural inclination is the word of gawd. If you can’t use holy massive penised Jehovah to instill this dreck then dream up socio-biological propaganda for the atheist hip guys needing a good shagging with their female cohorts.

Ignoring the weirdness of much the comment, is it true that men are more pro-natalist than women? I have shown that there seems to be a trend within the last 10 years of preference for larger families. What’s the sex breakdown for this?

The correlation between men and women is 0.65 year-to-year in their mean for ideal number of children. About 43% of the variance of the trend over the years can be predicted from one sex to the other. Is there is a systematic difference? Here’s a chart:

(more…)