Bryan Caplan  

Education, Ideology, and Awkward Weddings

Finally, Some Good Economic Ne... Theory, Explanation, and Polic...

Left-right ideology is by far the strongest predictor of party identification.  Education, in contrast, has very little connection to partisanship.*  However, when you look at opinions about specific issues, ideology and education are both extremely important.  For example, conservatives and the less educated are both sharply more hostile to immigration than liberals and the well-educated.  (See here for some simple regressions).

If you think about parties as coalitions of people with similar policy preferences, this is pretty weird.  Based purely on the data, we could just as easily have a "smart party" and a "stupid party" as a "liberal party" and a "conservative party."  But we don't.

OK, so maybe, per Robin Hanson, partisanship isn't really about policy.  But then what is it about?  The most natural alternative to me is that parties are coalitions of people who feel comfortable with one another.  Sure, educated liberals and educated conservatives have a lot of policy views in common.  But when they meet, there's a lot of social distance; people don't feel like they can "be themselves."

That's an interesting story.  But when I was watching the movie Rachel Getting Married, I started thinking: Which wedding would be more awkward?  A wedding where the groom's and bride's families differ sharply in left-right ideology?  Or a wedding where the groom's and bride's families differ sharply in education?  That seems like a pretty good measure of social distance.  And while I'm not sure, it seems like the mixed-education wedding would be more awkward than the mixed-ideology wedding.

What do you think?

* If you include both education and income in a regression, education seems to make people slightly more Democratic, and income seems to make people slightly more Republican.  But the key word is slightly.

Comments and Sharing

COMMENTS (18 to date)
Tom writes:

You might want to break it down between those who will have to compete against illegal immigrants and those who will not have to compete against illegal immigrants.

Those who are less educated are more likely to find themselves in competition with illegals for jobs. With more competition this will have a tendency to drive down wages for the less educated.

On the other hand, the more educated will not be in competition for jobs with illegals, but will be more likely to be buying goods and services at reduced prices from illegals or those in competition with illegals.

In general, those who benefit economically from illegal immigration are in favor of it and those who are harmed by illegal immigration oppose it.

El Presidente writes:

Great observation. I'm stumped.

jdo writes:

I find it baffling that well-educated people even vote. Seems like they should be offended by both Democratic and Republican policy.

ryan yin writes:

Something you have to keep in mind here is that education is arguably both a measure of human capital (and so conceivably part of the "political opinions are self-interested" hypothesis) and a demographic variable that might have an independent impact on political opinions (if no one thought this, would we really be arguing over public education?). In fact, if you regress immigration opinion on education and control for income, it's not clear why you'd be seeing a competition/self-interest story: why would high education/high wage people gain more from illegal immigration than low education/high wage people, and why would high education/low wage people be hurt less than low education/low wage people?

Les writes:

I think a better distinction is Thomas Sowell's differentiation of INTERESTS from VISIONS.

As Sowell explains it, there is the CONSTRAINED VISION (held by our founding fathers and libertarians) that people are imperfect, and that we need checks and balances to constrain bad behavior.

Then there is the UNCONSTRAINED VISION (held by liberals) that if only the right people are in charge, then everything will be perfect, be it Socialism, Communism or any other ism (except of course capitalism).

Tim writes:

It's certainly true that conservatives/libertarians as well as the well-educated tend to be in favor of free markets and free trade while left "liberals" as well as the low-educated tend to have an anti-market, make-work bias.

I'm not sure true small-government, libertarian-minded conservatives are more sceptical of high-potential (or even low-potential) legal immigration than left "liberals" - quite the opposite: The most pro-immigration president in the last decades was Ronald Reagan (who granted amnesty to more than 10 million immigrants and rightly so!), and GWB is a close second. Most opposition to the free movement of goods and labor comes from left-wing Democrats and the union movement. Some Republicans may be concerned that certain forms of illegal immigration may lead to the expansion of the welfare state but they are probably very much in favor of inviting high-level migrants from all over the world (who are likely to vote GOP).

In a perfect world we would abolish the welfare state as well as all forms of border controls - everybody would be free to come and live the dream of the pursuit of happiness. In a second best world scenario I would allow everybody to migrate as long as they opt out of all potential welfare programs. No true conservative or libertarian should be opposed to that.

Tim writes:

I've yet to see solid data but my hunch is that about a third of weddings are mixed-ideological weddings, but over time the worldviews of those who married more likely to converge than to diverge so over the years husband and wife become ever more likely to support the same party. Say, he is a persuasive eloquent Republican with strong libertarian/conservative convictions and a coherent worldview while she is a soft Democrat - a mixed-ideological wedding. Then it is somewhat likely that thanks to his persuasion she will have become a Republican after five or ten years. (The same vice versa.)

So we would expect a negative correlation between the probability that a given couple is mixed-ideological and the number of years they have been married.

Tim writes:

If you include both education and income in a regression, education seems to make people slightly more Democratic, and income seems to make people slightly more Republican. But the key word is slightly.

That would imply that Republicans tend to have higher (financial) returns on their investment in education.

However, I'm not really sure about the accuracy of your data. According to many surveys, Democrats tend to get most support from those who are categorized as "high-school dropouts". CNN 2004 exit poll data show a slight positive correlation between education and GOP preference: Kerry leads among "No High School", Bush leads 52% to 46% among "College Graduate". And an NES survey shows that 35% of male Republicans have a 4-year college diploma, but only 22% of male Democrats, and similar results for women.

Data also show that there is a strong correlation between income and GOP preference: Kerry leads Bush 63% to 36% among those with income under 15.000 dollars, while Bush leads Kerry 63% to 35% among those with income above 200.000 dollars. This year, McCain outperformed among above-median earners in at least 48 states. There are tons of data in Andrew Gelman's new book "Red State, Blue State, Rich State, Poor State". It's a bit counterintuitive: Though Democrats tend to do well in rich, educated states, and Republicans tend to do well in poorer states, it's also true that rich, well educated inividuals tend to vote Republican.

PS: To be more precise, there seems to be a U-shaped relation: Democrats do well among those without high-school degree, poorly among those with college diploma, but well again among those with postgraduate degree. That's probably because among postgraduate degrees Democratic leaning fields such as teaching, humanities, liberal arts and law are heavily overrepresented, while fields that tend to be more Republican such as business, economics, mathematics, natural sciences, computer science, engineering are underrepresented. Here are some statistics on postgraduate degrees: 17.5 million adults in the US have postgraduate degrees: 3.8 million are in education, 2.1 million in humanities and liberal arts, and 1.3 million in law - that amounts to 7.2 million degrees in heavily Democratic fields. There are 2.5 million postgraduate degrees in business and 3 million in mathematics, natural sciences, computer science and engineering - that amounts to only 5.5 million degrees in Republican leaning fields. (There are 4.8 million degrees in other fields.)

So even though there are a lot of further subtleties, it is still a good first order approximation to say that there is a positive correlation between political preference for the GOP on the one hand and income, education and return on investment in education on the other hand.

Fabio Rojas writes:

A few quick notes:

1. In political science, there's a good tradition of analyzing parties as umbrellas for people who feel comfortable with each other. For ex, Blacks and Southern Whites have never been in the same party. Labor has never been in the same party as business. Urbanites are usually in a different party than ruralists. In Europe, linguistic minorities are often in different parties than nativists.

2. re: Weddings. The technical term in sociology is "educational homogamy." I don't think it's been measured against ideological homogamy, but same education level is a huge predictor of who gets married to whom. My bet is that only region, race and religion are bigger predictors.

On the issue of how education makes people similar, I'll point to Max Weber's discussion of status groups: people create status groups by engaging in the same lifestyle. Education is part of that story in modern societies. Aside from grouping people of similar cognitive levels, you also learn how to speak, what to read, consume, etc. And people usually bond over lifestyles more so than politics. Maybe you should read Bourdieu, he's got a whole book this stuff...

RL writes:


Interesting post.

I suspect wedding small talk is such that a group of educationally similar people is more comfortable than a group of ideologically similar people. But it's "unique" to the situation.

After all, we've all been raised to know that the sorts of things one may have in common with others ideologically are NOT the things one discusses at a wedding.

OTOH, if you had a group of people making small talk at a sporting event, or in a TV audience, I suspect people would feel more comfortable with an ideologically similar cohort than an educationally similar one.

Zac writes:

I agree with Bryan that the mixed-education wedding would be more awkward. Here's why: you can't really avoid the awkwardness that comes from having vastly different educational backgrounds. The differences will be very apparent - very different life stories, different lines of work, different day to day issues. They may find that they have nothing to chat about beyond the weather, and will probably have a very different vocabulary and way of talking. However, topics like religion and politics tend to be avoided in polite society events like weddings. At school and work people interact with people who have vastly different ideology and you may never know it.

I disagree that "parties are coalitions of people who feel comfortable with one another." I don't know what causes partisanship, but I think that story can be rejected out of hand, at least as a general explanation.. e.g., how comfortable would the educated, white, liberal democrats you know be with inner city blacks? I always think of Tommy Carcetti visiting a black church in The Wire while campaigning. You could cut the awkwardness out of the air with a dull spoon in that scene.

CJS writes:

We had a mixed ideology wedding in 2005; there was some slight ribbing, but mostly it went off without a hitch. The education levels were similar - although there were a couple less intelligent college grads who stood out like sore thumbs. Perhaps the biggest gap is intelligence, irrespective of education?

Jesse writes:

As Sowell explains it, there is the CONSTRAINED VISION (held by our founding fathers and libertarians) that people are imperfect, and that we need checks and balances to constrain bad behavior.

Then there is the UNCONSTRAINED VISION (held by liberals) that if only the right people are in charge, then everything will be perfect, be it Socialism, Communism or any other ism (except of course capitalism).

This is a great point. Why do liberals support greater government regulation of business, for example? This is a perfect example of an UNCONSTRAINED VISION held by idiot utopians. Why don't liberals realize that the world would be a better place if individuals and businesses are as free as possible to promote their own interests? Is it because they don't realize the need for checks and balances? It all makes sense to me now.

Jacob Oost writes:

Mixed-ideology wedding sounds like the one that would *tend* to be the most awkward, as many people who are highly educated do not come off as elitists or whatever (take GWB, who went to Harvard and Yale and comes from blue blood, or Conan O'Brien).

As to immigration, one mistake I notice a lot of libertarians making (calm down, I'm pretty econotarian) is that they do not understand that people who complain about illegal immigration are actually TWO different groups, with different viewpoints. There is the paleoconservative, keep-em-away-from-our-jobs-and-our-women-folk viewpoint, which is the viewpoint that those categorically in favor of amnesty (btw, amnesty from what? Mexican drug gangs?) project onto anybody making an argument against amnesty, and then there is the more practical and utilitarian viewpoint which opposes illegal immigration for the simple reason that IT IS ILLEGAL, and a country needs laws which are to be followed and enforced by the government in order to function. The Road to Serfdom can tell you all about this.

If thirty million legal immigrants were here, it wouldn't bother me a bit, but thirty million illegal immigrants, benefiting from but not paying for government programs (schools, welfare, etc.), bankrupting hospitals, acting like they own the place and that they, as people who broke into this country, have the same rights as people who came here by the book? No, these are not two equivalent hypothetical groups but for one difference. They aren't undocumented Americans. They aren't unofficial citizens. Countries need laws and those laws need to be enforced. If our existing immigration laws need reform, then reform them, don't ignore them. Otherwise we are relying on the arbitrary rulings of individuals, rather than law, to govern society.

One of the most basic functions a government can perform is to distinguish between citizens and non-citizens. Maybe one day, if and when all nations have free economies and national borders seem as transparent to us all as state borders do now, then unfettered immigration would work. Until now, I want every single immigrant to have documentation. I want to know when they come and when they leave. We could have prevented 9/11 if we'd actually followed our immigration laws.

I find that many people who are in favor of blanket amnesty have virtually no experience with illegal immigrants. I'm not a professional, I've worked side by side with Mexicans, Venezuelans, Somalis, etc. Let me tell you all something. These aren't the America-loving immigrants you see in John Ford movies. More often than not they hate this country, they show open disdain for its laws, culture, and people. They are here to make money, live in ethnic enclaves, and bad mouth America and Americans, and then go organize protests whenever we complain that they aren't paying their fair share in taxes (because they get paid under the table). How does a country make itself better off by importing loads of people who hate it?

Babinich writes:

"education seems to make people slightly more Democratic"


Instead of "education" maybe the statement should read: a college degree seems to make people slightly more Democratic.

I agree with Jacob: a ideological wedding where the families are one hundred and eighty degrees apart (liberal - conservative) seems to me to be the relationship that would tend to be most awkward.

nicole writes:

I think the mixed-education wedding would be much, much worse. My boyfriend and I have a similar level of education, but I am the first in my family to go to college and his parents are both lawyers and professors. Both our families are very progressive, he's progressive, and I'm something like a libertarian. Our families have never met, and I would like to keep it that way, because while they might agree on everything political they would share nothing else in common--as mentioned above, life stories, paths taken, jobs, day to day life. I can't imagine them relating to each other at all.

Politically, I am the outcast, but it's not that awkward because it's much easier to ignore or not bring up.

Steve Roth writes:

Bryan, you might look at the "authoritarian" character type. Seems to be a major predictor of both "conservative" and "Republican." (Note that there are authoritarian "followers" and authoritarian "leaders.")

John Dean ranted about this quite a bit in his book, Conservatives Without Conscience.

Steve Roth writes:

But what I really want to ask--have been waiting for a somewhat on-topic post to do so:

Based on SAEE, do libs or cons, dems or pubs, think more like economists? (IOW, who's more "rational"?)

You skirt close to the question in a few of your writings. i.e. here:

p. 11 n. 13 "There are also two equations where Othideol is significant. In both cases people who do not think in liberal-conservative terms think less like economists."

And the Steven Miller paper (that you sort of co-authored?), tries to tease this out by combining General Social Survey results with SAEE results. It addresses this very question, but relies on two separate and not-necessarily-comparable surveys, and that in a statistically unsophisticated manner.

Odd, because presumably the answer is immediately to hand in your SAEE analyses.

The SAEE data doesn't seem to be published on the web, and in any case you have the statistical apparatus in place. What's the answer? Is the result significant (statistically and/or politically)?

Comments for this entry have been closed
Return to top