Part of the popularity of our demonstration archive is that it is free for end users. We are happy to provide this service. It is a valuable resource for the academic community and it also publicizes the value of our SDA software. However, the flip side of providing this free service is that it does not generate any income to offset the cost of providing the infrastructure required. We receive no funding from GSS for hosting their datasets — which is often a surprise to our users. Almost all of our income comes from the fees provided by licensing the SDA software to other data archives (like ICPSR and IPUMS), and virtually all of that income goes to support the programming and technical support that we provide them. We obviously need some additional sources of revenue.
Long time readers know that it’s trivially easy to extract information from the GSS that political moderates and independents are not as intelligent as partisans and ideologues. New readers are not always familiar. A comment:
#8 Do you have something to back up the idea that independents are less intelligent? If anything, I would’ve expected the opposite- that independents are capable of thinking for themselves instead of following the party line.
First, a quick review of the data. I used two GSS variables, PARTYID and POLVIEWS, and limited the sample to non-Hispanic whites after the year 2000. I removed those of “Other party” as well. Finally, I crossed that against vocab score results, which correlates with intelligence with a value of 0.70. It is rather obvious that middle-of-the-roaders are not as bright:
In the post below I took the time out to link to the GSS, as well as posting my exact queries. As payment for this consideration the first comment was absolute drivel. I understand people have political opinions, but I’m not too interested in your opinions. You may be interested in your opinions, but I’d rather have more data. Most people don’t know enough for me to have interest in their opinions (most != all, many readers do have opinions in their specialties which I seek out).
I was trying to make a point that anger and even violence in reaction to actions which offend are actually comprehensible as the modal human response. The community reacts to punish those who violate taboos. The taboos may differ, but the response to the action of violation is normal and natural. A primary issue that needs to be considered is that taboos differ from society to society, so one is often not conscious of the act of violation (e.g., if you show the bottom of your shoes to people when you sit down, that’s an offensive act in some societies).
An implication here is that American norms of free speech near absolutism, enforced through the fiat of the courts because of their interpretation of the applicability of the Bill of Rights, are radically non-intuitive to most people. The only reason that they are intuitive to many Americans is that we are acculturated over time. This is clear when you look at differences of intelligence and education. In short, less intelligent and educated Americans are much more skeptical of allowing social deviants to speak. This is true even in cases where they are more likely to agree with the deviant in question (e.g., these groups have a more pro-military bent, and yet are more accepting of the concept of censure of pro-military opinions).
I have limited all the results to the year 2000 and later. Additionally, I classify those who score 0-4 on the WORDSUM vocab test as stupid, those who score 5-8 as average, and those who score 9-10 as smart. WORDSUM has been reported to have a 0.7 correlation with general intelligence. In this data set 20% of individuals scored 0-4, 69% 5-8, and 12% 9-10.
Obviously the news over the past week has been filled with the events in the Middle East, and the broader Muslim world, in reaction to an anti-Muslim film. I think the most eloquent commentary is from The Onion (NSFW!!!), No One Murdered Because Of This Image. That being said, there are some serious broader issues here. A friend of mine who lives in India (he is Indian American, though raised for several years in India, so not totally unfamiliar with the culture) has expressed to me his frustration with having to defend American liberalism in a society where American liberalism is an abstraction, rather than concrete. The frustration has to do with the fundamental divergence in basic values. For example, his interlocutors have argued to him (he is a practicing Christian of libertarian political orientation) that if someone committed an act of blasphemy against his faith of course he would react in anger and violence. And yet of course the clause “and” is false, though he is greeted with skepticism when he asserts he wouldn’t react violently. As a matter of fact I can attest to the reality that he wouldn’t react angrily necessarily, because in interactions where I’ve made casually blasphemous comments he’s only rolled his eyes. Just as Americans have a vague, even misleading, understanding of the broader historical forces which engender resentment of American hegemony in the broader world, so many non-Americans lack a proper awareness of the broader historical forces, and cultural reality, of the particular American radicalism and extremism in the domain of free expression.
It’s basically impossible to avoid hearing about Todd Akin right now. My Twitter and Facebook feeds are kind of swamped. But it did make me wonder: what percentage of Americans reject abortion in cases of rape and incest? The GSS has a handy variable, ABRAPE, which asks respondents about the possibility of abortion if a woman gets pregnant as a result of rape (let’s stipulate that it’s possible to get pregnant as a result of rape!). I also limited the sample to the year 2000 and later, and non-Hispanic whites (to clear out confounds). Demographic breakdowns below….
Long time readers know that one of my pet hobby-horses is to try and convince more pundits that they should use the GSS. Opinions based on opinions may be fun, but opinions based on facts may be useful. In general my appeals have fallen on deaf ears. But today I notice that Will Saletan is using GSS data to discussion the Todd Akin case. You may not agree with Saletan’s take on the results, but at least he bothered to generate some results.
Reihan Salam has a post up on the alignment of racism and political orientation. He begins:
Recently, Chris Hayes, host of MSNBC’s UP with Chris Hayes, made the following observation:
It is undeniably the case that racist Americans are almost entirely in one political coalition and not the other.
Chris is a good friend of mine, and we grew up in the same milieu. I can attest to the fact that the view he expressed is very widely held in the circles in which we both travel….
Salam then links to Alex Tabarrok, who uses party identification data to indicate that actually racism is split between the two groups, while John Sides suggests that there is a definite lean toward Republicans being more racist, using a few indicator variables. Overall I think Sides is about right, all things equal conservatives are more racist than liberals. At least in the modern context of the two ideologies.* I say conservative/liberal rather than Republican/Democrat, because my experience with the GSS data set is that ideology is a more powerful predictor of social views among whites. This holds true with the variables which Tabarrok and Sides query from what I can see; the gap between Democrats and Republicans is smaller than between liberals and conservatives. Why? There are still a non-trivial number of self-identified conservative Democrats in this country, as well as very well off socially moderate Republicans who vote their economic interests.
Prompted by a comment below I was curious as to the correlation between intelligence and income. To indicate intelligence I used the GSS’s WORDSUM variable, which has a ~0.70 correlation with IQ. For income, I used REALINC, which is indexed to 1986 values (so it is inflation adjusted) and aggregates the household income. Finally, I limited my sample to non-Hispanic whites over the age of 30 (for what it’s worth, this choice also limited the data set to respondents from the year 2000 and later).
The results don’t get at the commenter’s assertions, because 10 out of 10 on WORDSUM does not imply that you’re that smart really. But the trendline is suggestive. Note that aggregated 0-4 because the sample size at the lower values is small indeed.
A few years ago I put up a post, WORDSUM & IQ & the correlation, as a “reference” post. Basically if anyone objected to using WORDSUM, a variable in the General Social Survey, then I would point to that post and observe that the correlation between WORDSUM and general intelligence is 0.71. That makes sense, since WORDSUM is a vocabulary test, and verbal fluency is well correlated with intelligence.
But I realized over the years I’ve posted many posts using the GSS and WORDSUM, but never explicitly laid out the distribution of WORDSUM scores, which range from 0 (0 out of 10) to 10 (10 out of 10). I’ve used categories like “stupid, interval 0-4,” but often only mentioned the percentiles in the comments after prompting from a reader. This post is to fix that problem forever, and will serve as a reference for the future.
First, please keep in mind that I limited the sample to the year 2000 and later. The N is ~7,000, but far lower for some of variables crossed. Therefore, I invite you to replicate my results. After the charts I will list all the variables, so if you care you should be able to replicate displaying all the sample sizes in ~10 minutes. I am also going to attach a csv file with the raw table data. As for the charts, they are simple.
- The x-axis is a WORDSUM category, ranging from 0 to 10
- The y-axis is the percent of a given demographic class who received that score. I’ve labelled some of them where the chart doesn’t get too busy
All of the charts have a line which represents the total population in the sample (“All”).
Update: There was a major coding error. I’ve rerun the analysis. No qualitative change.
As is often the case a 10 minute post using the General Social Survey is getting a lot of attention. Apparently circa 1997 web interfaces are so intimidating to people that extracting a little data goes a long way. Instead of talking and commenting I thought as an exercise I would go further, and also be precise about my methodology so that people could replicate it (hint: this is a chance for readers to follow up and figure something out on their own, instead of tossing out an opinion I don’t care about).
A questioner below was curious if vocabulary test differences by ethnic and region persist across income. There’s a problem with this. First, the INCOME variable isn’t very fine-grained (there is a catchall $30,000 or greater category). Second, it doesn’t seem to control for inflation. But, there is a variable, DEGREE, which asks the highest level of education attained. I used this to create a “college” and “non-college” category (i.e., do you have a bachelor’s degree or not). Because of sample size considerations I removed some of the ethnic groups, but replicated the earlier analysis.
Below are two tables. One shows the mean vocab score for region and ethnicity (for whites) for those without college educations, and another shows those with college educations. I decided to generate a correlation over the two rows, even though it sure isn’t useful as a quantitative statistical measure because of the small number of data points. Rather, I just wanted a summary of the qualitative result. The short answer is that the average vocabulary difference seems to persist across educational levels (the exception here is the “German” ethnicity).
Mike the Mad Biologist has a post up, A Modest Proposal: Alabama Whites Are Genetically Inferior to Massachusetts Whites (FOR REALZ!). The post is obviously tongue-in-cheek, but it’s actually an interesting question: what’s the difference between whites in various regions of the United States? I’ve looked at this before, but I thought I’d revisit it for new readers.
First, I use the General Social Survey. Second, I use the WORDSUM variable, a 10 question vocabulary test which has a correlation of 0.70 with general intelligence. My curiosity is about differences across white ethnic groups by region. To do this I use the ETHNIC variable, which asks respondents where their ancestors came from by nation. I omitted some nations because of small sample size, and amalgamated others.
Here are my amalgamations:
German = Austria, Germany, Switzerland
French = French Canada, France
Eastern Europe = Lithuania, Poland, Hungary, Yugoslavia, Russia, Czechaslovakia (many were asked before 1992), Romania
Scandinavian = Denmark, Norway, Sweden, Finland (yes, I know that Finland is not part of Scandinavia, Jaakkeli!)
British = England, Wales, Scotland
Northeast = New England, Middle Atlantic
Midwest = E North Central, W North Central
South = W S Central, E S Central, South Atlantic
West = Pacific, Mountain
The key method I used is to look for mean vocabulary test scores by ethnicity and religion. I also later broke down some of these ethnic groups by religion. Finally, all bar plots have 95 percent confidence intervals. This should give you a sense of the sample sizes for each combination.
First let’s break it down by race/ethnicity and compare it by region to get a reference:
Someone on twitter was curious about GOP attitudes toward astrology. I left the party breakdown out of the previous post because ideology accounts for most party differences. In other words, conservatives are more skeptical of astrology than liberals, and Republicans more than Democrats, but the second result just seems to emerge from the Republican’s greater conservatism.
|Astrology very scientific||Astrology somewhat scientific||Astrology not scientific|
Why are independents so gullible? It probably has to do with their lower average intelligence (this goes for moderates too). So I simply limited the sample to those with at least bachelor’s degrees to control for intelligence:
I just finished reading My Fertility Crisis, which is excerpted from a longer piece you can get on Kindle for $1.99. The author is a single woman in her early 40s who is going through IVF treatments, without success so far. She outlines the choices she made over her life which may have influenced her current situation.
After reading the piece I came back to an issue I’ve wrestled with before: it’s often really hard to find information on probability of pregnancy online in the form of charts. The reason is that there’s so much information, and much of it is skewed toward people who are undergoing treatment for infertility. But why look when you can generate your own visualization? I found a pregnancy probability calculator online which I cross-validated with some of the literature. Here is the best case scenario for probability of pregnancy if you are trying in the natural fashion (the probabilities exclude women who are clinically infertile, which is a rather slippery category strongly dependent on age, so the older cohorts are probably much larger overestimates than the younger ones):
The main focus is really the decade of the 30s for women. Here is a figure from Ovarian Aging: Mechanisms and Clinical Consequences which shows a finer-grain decline in fertility:
Lots of commentary below on my post about extramarital sex. I guess that’s fine, but I’m really not too interested your theories, I can do basic logic after introspection too. In fact, I can go down the street and ask a random person and I’m sure they could offer up after the fact rationales for the results I reported (people are always interested in sex and sharp about models to explain it). Instead, here’s the variable you need to use in the GSS: XMARSEX. I assume forms and graphical user interfaces worthy of 1997 are not too intimidating to readers of this weblog even if they perplex Matt Yglesias?
In any case, here’s some more results. First, I wanted to double check that there was in fact decreased tolerance of extramarital sex over the years. Let’s break it down by sex:
Some of you were curious about the demographic correlates of this behavior. Please note that all the following charts are limited to the year 2000 and later. The sample sizes for XMARSEX were rather large, so I saw no reason not to make it relevant to contemporary attitudes.
My girlfriend’s theory about this, which makes sense to me, is that as women’s labor market opportunities have improved their dependency on husbands for economic security has declined and, in turn, their willingness to put up with misbehavior has gone down. Looking at a gender breakdown of responses might shed some light on this, but I can’t figure out how to work the General Social Survey website.
He’s talking about a chart which shows decline in tolerance of extramarital sex by education:
I just replicated but broke it down by male and female:
People who believe in the Bible’s literal truth (BIBLE=1) are much more satisfied with their jobs than people who believe it’s just a book of fables (BIBLE=3)….
Relatively speaking, this is a huge effect. But what’s going on? It’s not just a disguised left-right effect; Biblical literalism crushes self-identified ideology in a multiple regression. And it’s not a disguised social support effect; Biblical literalism crushes church attendance, too. Marxists will no doubt claim vindication for their view that religion is the opium of the people. But you could just as easily conclude that traditional religion successfully teaches gratitude.
In my personal experience with the GSS the BIBLE variable, which asks rather awkwardly one’s stance toward the nature of the Bible, is one of the most powerful predictors of a whole host of social metrics. I suspect that scriptural literalism has very strong personality correlates.
One thing that’s rather amusing also is that Biblical fundamentalists are naturally skeptical of evolution, but they’re rather reproductively fit. Here’s the trend from the GSS:
In recent years, many scholars have explored the degree of polarization between red and blue states (red states are those carried by Republicans at the presidential level; blue states are those carried by Democrats). Some claim that red- and blue-state citizens are deeply polarized, while others disagree, arguing that there are only limited differences between the two groups. All previous work on this topic, however, simply uses difference-of-means tests to determine when these two groups are polarized. We show that this test alone cannot determine whether states are actually polarized. We remedy this shortcoming by introducing a new measure based on the degree of issue-position overlap between red- and blue-state citizens. Our findings demonstrate that there is only limited polarization—and a good deal of common ground—between red states and blue states. We discuss the implications of our work both for the study of polarization itself and for the broader study of American politics.
Generating a statistical construct of the distribution of liberalism and conservatism on social and economic issues the authors produced a set of plots which illustrate the differences between “red” (conservative) states and “blue” (liberal) states. In the figures below the blue line represents “blue states/regions” while the red dashed lines represents “red states/regions.”