Voting patterns of America’s whites, from the masses to the elites

by Andrew Gelman on March 23, 2012 · 33 comments

in Campaigns and elections

Within any education category, richer people vote more Republican. In contrast, the pattern of education and voting is nonlinear. High school graduates are more Republican than non-HS grads, but after that, the groups with more education tend to vote more Democratic. At the very highest education level tabulated in the survey, voters with post-graduate degrees lean toward the Democrats. Except for the rich post-graduates; they are split 50-50 between the parties.

What does this say about America’s elites? If you define elites as high-income non-Hispanic whites, the elites vote strongly Republican. If you define elites as college-educated high-income whites, they vote moderately Republican.

There is no plausible way based on these data in which elites can be considered a Democratic voting bloc. To create a group of strongly Democratic-leaning elite whites using these graphs, you would need to consider only postgraduates (no simple college grads included, even if they have achieved social and financial success), and you have to go down to the below-$75,000 level of family income, which hardly seems like the American elites to me.

The patterns are consistent for all three of the past presidential elections. (The differences in the higher-income low-education category should not be taken seriously, as the estimates are based on small samples, as can be seen from the large standard errors for those subgroups.)

Notes: These graphs show just (non-Hispanic) white voters because (a) most voters are white, (b) minorities tend to vote consistently for Democrats, so there’s less variation in their voting patterns, and© sample sizes are smaller for nonwhite groups so it’s hard to see clear patterns amid the noise. Data come from Annenberg pre-election polls for 2000 and 2004 and Pew pre-election polls for 2008; total number of non-Hispanic white respondents: 26161, 36476, and 15212. Graphs show intended vote in presidential election, including only those who expressed a preference for the Democratic or Republican candidate. Income categories are defined as family income less than $20K, $20-40K, $40-75K, $75-150K, 150K+. This work was done in collaboration with Yair Ghitza.

{ 33 comments }

Vance Maverick March 23, 2012 at 6:23 pm

There is no plausible way based on these data in which elites can be considered a Democratic voting bloc.

Unless we redefine “elite” with some form of special pleading. Mind you, I don’t think that would make sense, but that’s the way I’d predict a David Brooks to move.

Will Wilkinson March 23, 2012 at 8:19 pm

I think the right-wing idea of the elite is more about cultural influence than income. Media & professoriate–that sort of thing. That relatively high income makes you elite, or gives you influence, seems more a leftish notion.

wj March 28, 2012 at 3:42 pm

The true meaning of “elite” in current usage: “Anybody richer than me who does not pander to me.”

If you are in poverty, it is true, you are not part of “the elite.” But as long as you say all the right things about issues I care about (religion, race, the evil of government, etc.) you are still not part of “the elite” — even if you are making hundreds of millions of dollars a year, and doing so by moving jobs offshore.

I guess what I’m saying is that it not only isn’t a matter of income, it is also not really a matter of culture — at least, it’s not a matter of shared culture. It is a matter of failing to pretend to value a culture you do not belong to and do not agree with. As long as you are willing to pander, you are fine.

PBR March 23, 2012 at 9:55 pm

While I find this analysis really interesting, I agree with Will about the meaning of “elites.” The term seems to capture some level of social influence that professors, journalists, CEOS, and certain celebrities might have because they are highly public figures that many people look up to. By contrast, doctors, lawyers, and bankers may be high-income and well-educated, but they are not exactly viewed as opinion movers.

Andrew Gelman March 23, 2012 at 10:24 pm

Will, Pbr:

“Elite” can mean a lot of things, but I think income is part of it. You can think of this two ways:

First, among groups such as the news media, it is the richer ones who are elite. I don’t think you’ll get far claiming that some TV producer somewhere with a family income of $75K/year is elite.

Second, when conservatives talk about elites, it’s not just about influence, it’s also about privilege. You con’t have to go all the way over to mouth-breathers [sorry, that was rude, I retract that; see comments below] like Michael Barone to get this. The oft-expressed conservative idea is that liberal elites are rich and out of touch. Sure, there are low-income and middle-income liberals, but I don’t see them hitting the “elite” category until they reach a higher income level. To put it another way, Paul Krugman is an elite, but some bearded Marxist dude adjunct-teaching at a community college is not an elite, even if he does have a Ph.D.

If you want to subdivide further and consider only people who live in cities, then you can get a group of elite Democrats. But then you have to consider the corresponding group of elite Republicans who live in suburbs. America is a land of multiple elite groups.

RobC March 24, 2012 at 11:27 am

Michael Barone a mouth-breather? Here’s a news flash for you: just because you disagree with someone doesn’t mean he’s dumb. Barone has degrees from Harvard College and Yale Law. At Yale Law, where I knew him, he was universally understood to be one of the standout intelligences–and it is, let me assure you, a very fast track. His long career in Washington has seen him move from liberal (Peter Hart Research and the Washington Post editorial board) to conservative, but the one constant has been his encyclopedic knowledge and sheer braininess.

Schoolyard taunts really are beneath you, Professor Gelman, but they’re especially pathetic when they are so utterly untrue. All they accomplish is to make you look arrogant and mean-spirited, traits you should seek to suppress or, at the very least, disguise.

Andrew Gelman March 24, 2012 at 5:25 pm

RobC:

I agree that Michael Barone knows a huge amount and has made huge contributions to our understanding of American politics. But on the particular topic of income and voting, he has some serious hangups, and his data sense seems to have deserted him. See, for example, here.

RobC March 24, 2012 at 6:03 pm

Interesting. In precisely seven years Barone went from “a smart person was led by ideology to overgeneralize” to a “mouth-breather.” This must be what people mean when they talk about the increasing lack of civility in public discourse. The other possibility is that the person who described him that way in 2005 was himself a mouth-breather, and we know that’s not true because, as we learn on the first page of his 25-page CV, he achieved an undergraduate GPA of 4.9/5.0, and at MIT noch.

Andrew Gelman March 24, 2012 at 6:22 pm

RobC:

You’re right. I would like to change the above “You don’t have to go all the way over to mouth-breathers like Michael Barone . . .” to “You don’t have to go all the way over to people such as Michael Barone who appear to be ideologically committed to this position . . .”.

P.S. Back when I was at MIT, I was a good student, but I’ve gone downhill since then.

Peter T March 24, 2012 at 1:07 am

Elite is a slippery notion – it’s something about wealth, but also something about being an opinion former or leader, and also also something about social status. And there are national elites, but also local elites.

In Deerhunting with Jesus Joe Baigent noted the influence of local opinion leaders (in small town Virginia, realtors, car dealers, bank managers and so on – not nationally wealthy, but well-off by local standards) – their ability to steer conversations, and the deference shown their opinions. The equivalent liberal local elite probably used to be schoolteachers, librarians, college teachers and similar – again not nationally wealthy, but reasonably well-off and, above all, secure. Not quite sure how you would capture this. The latter group seems to have suffered a decline in numbers, prestige, wealth and security relative to the first group, but it is probably the residual influence of the liberal group that those who talk about “liberal elites” have in mind.

Christian Collet March 24, 2012 at 10:24 am

“Elite,” in the conservative lexicon, is an umbrella term/epithet that captures the select institutions where conservatives have — or seem to believe they have — little social/political influence (e.g., The New York Times). Conversely, when liberals employ the term it seems to be more often applied to upper economic/corporate strata who wield influence over government. Nonetheless, what Andrew’s analysis reveals, ever so efficiently, is the consistency that manifests when it is examined at the electoral level. Any reasonable notion of an ‘elite’ within the country as a whole begins with the notion that it votes overwhelmingly Republican.

ScottA March 24, 2012 at 10:52 am

Building on PeterT – Charles Murray’s recent book defined elites as opinion leaders, high SES, high prestige professions. There’s some co-variance with income, but it’s not perfect. He might argue that what you’re seeing is self-selection based on ideology; more conservatives go into business for the same reason most social scientists are liberals; it’s not really bias, they just choose industry careers / graduate education at higher levels, respectively. And industry pays off better in terms of raw income. In terms of influence, I’m not so sure. Anyways, I think the income component is misleading you; there are plenty of non-wealthy elites. How much does David Brooks make (perhaps a bad example, but he was influential before he was rich, if he is rich, in any case)?

Andrew Gelman March 24, 2012 at 5:28 pm

ScottA:

I’m pretty sure that David Brooks has a family income of more than $150K/year, placing him in the top category above. According to Wikipedia, he seems to have a college degree but no postgraduate degree, putting him in education category #4 above.

J March 24, 2012 at 12:50 pm

Hey Andrew, I’d be interested in seeing this graph where the x axis was actually weighted to fit the distribution of income. The x axis distorts might distort the impact of high income voters visa vi middle class voters. Though to be honest if political participation among the lower income groups is low then this may be more accurate than I am giving it credit for.

TXG1112 March 28, 2012 at 3:47 pm

I think you’re trying to do a technical analysis on what is merely a partisan epithet. That the wealthy mostly vote Republican isn’t really news. I would suggest that when conservatives refer to liberal elites, what they really mean is liberal elitists which is not the same thing.

Andrew Gelman March 28, 2012 at 8:29 pm

Txg:

It’s not really news to you, but apparently it is news to Tucker Carlson, Michael Barone, and many others.

TXG1112 March 29, 2012 at 10:50 am

I’m sure I’m at risk of sounding overly snarky, but that isn’t news either.

Andrew Gelman March 29, 2012 at 1:35 pm

Yes, but I’m hoping that by plotting this sort of graph, I’ll get the message out to some people who have been confused on the topic. It’s part of the service aspect of my job.

ezra abrams March 29, 2012 at 5:00 pm

this is a really nice example of a trellis plot; I feel a little churlish, but my graphics juices are flowing – how would you add to the trellis something that indicates % of electorate in high med low income groups ?

also, “elite” means diff things to diff people; in rushlimbaugh speak, as I degrade myself every now and then to hear, “elite” means grad school educated liberals, it doesn’t mean what normal people think of when they think of elite

Andrew Gelman March 29, 2012 at 9:30 pm

Ezra:

You could indicate group sizes via gray circles at each point.

Jim March 29, 2012 at 6:28 pm

Spare me. How about an article with graphs and statistics describing how

Black People

vote? Are they ill-informed as well?? Are there financial variables that impact the voting demographics of how

Black People

Vote?

What totally asinine racist sh t stirring. F this ‘report’ and F the authors of this blog.

Andrew Gelman March 29, 2012 at 9:31 pm

Jim:

As I wrote, these graphs show just (non-Hispanic) white voters because (a) most voters are white, (b) minorities tend to vote consistently for Democrats, so there’s less variation in their voting patterns, and (c) sample sizes are smaller for nonwhite groups so it’s hard to see clear patterns amid the noise.

Carl Weetabix March 29, 2012 at 7:00 pm

I think it’s useful to frame this out even if I agree they don’t mean “elite” in the same way when they talk about liberals. On the other hand, I don’t think they really mean academics, lawyers, Hollywood personalities either. What they mean is something roughly equivalent to the n-word. That is, it doesn’t have any true meaning other than to dehumanize the target. They could substitute say, “something beneath contempt” for “liberal elite” and it would satisfy the same function.

Frankly they say the word “liberal” with nearly the same vitriol, but “elite” adds an extra special twist.

Carl Weetabix March 29, 2012 at 7:05 pm

Another way to describe “liberal elite” is, “those who are not us who we don’t want to be like”. It really doesn’t matter what target thing it is “like” – that really exists in some sort of xenophobic unimagined otherness that isn’t important. The point is, it’s something you don’t want to be.

It’s the same reason that 5th graders call other 5th graders “fags”. They don’t really know what a “fag” is and it doesn’t really matter in the end, they just know it isn’t something you want to be. “liberal elite” has the same significance.

Carl Weetabix March 29, 2012 at 7:43 pm

Or perhaps an even better comparison – calling someone “liberal elite” is like calling someone a “cheap jew”. Sure we know that both “cheap” and “jew” in their literal definitions are effectively “unloaded” terms, but when put together in this particular combination it gains a far, far uglier connotation. In that case the emphasis truly is on “jew” not “cheap” and while “jew” by definition is simply a follower of Judaism, in the “cheap jew” form the “jew” becomes some nefarious undefined “other”. It doesn’t mean anything that you can really define – it means some sort of icky thing you don’t want to be or be with. A “boogeyman”.

If the word “cheap” in this case was truly significant then you could go “cheap protestant” with the same effect, but you can’t.

The same is true with “liberal elite”. “liberal” and “elite” both have relatively unloaded definitions, but spun together they become something else, with the emphasis being on the badness of “liberal”. Again, like “cheap”, “elite” isn’t really all that important. If it was we could go “conservative elite” and it would be similarly painfully derogatory, but it isn’t.

The point of all this being, “liberal elite” comes from the same dark roots as “cheap jew”, the n-word, or “fag”. The literal definitions are lost in the (evil) connotations and any literal definition has no real importance other than its dehumanizing effect.

The making of “liberal” and “liberal elite” into defamatory meanings is no accident, just as “socialist” and “communist” has become lingua-non-grata. A lot of careful propaganda (as in Edward Bernays) has been put into the effort of demonizing these words. Addressing them as if they actually mean something is probably jousting at windmills.

frederico perry March 29, 2012 at 10:17 pm

> In contrast, the pattern of education and voting is nonlinear.

Not that it affects your argument in the least, but the word you want is non-monotone, not nonlinear.

Andrew Gelman March 29, 2012 at 11:13 pm

Frederico:

Yes, I actually intentionally used the term “nonlinear” to mean “non-monotone” because it is my impression that this is how the word is commonly used in these contexts. “Non-monotone” is more precise but to me sounded too jargony. But I’m glad someone noticed!

Arms Merchant April 9, 2012 at 1:14 am

You and your commenters all dance around the central issue, which is not semantics, but that liberals control Hollywood, Education, traditional Media, and the permanent Federal bureaucracy. Wealth has something, but not nearly everything, to do with who runs the country.

That’s what conservatives and libertarians mean by “The Liberal Elite.”

Andrew Gelman July 20, 2012 at 3:48 am

As we discuss (briefly) in our book and elsewhere, different industries and sectors of the economy have different political inclinations. Given that Republicans are 50% of the voters and most of the rich people, it is unsurprising that they are influential in many aspects of our economy and society. For example, military officers and big business executives are much more likely to favor the Republican party. Democrats might be in charge of Children’s Television Workshop, but conservative Republicans such as Grover Norquist have a lot of power too! And, indeed, to some extent we have social policies that would be acceptable to the creators of Sesame Street and economic policies that make Grover Norquist happy.

cynthia curran July 20, 2012 at 2:32 am

Republicans in Texas were found to be in counties over 100,000 but we get this blue collar sterotype. In fact I seen a graph where lower income whites in Texas vote Dem and high income vote Republican.

Matthias September 14, 2012 at 1:37 pm

That’s pretty much every state – I mean, it’s true of the country as a whole, as Gelman points out, but within states the covariance of income and voting is even stronger. The lower association nationally is because poorer states are more conservative.

Duh September 23, 2012 at 3:31 pm

Educated white people vote democratic. Minorities vote democratic.
Rich white people vote Republican. Uneducated white people vote Republican.
Educated Rich white people votes are split.

You can see why the races are usually so close.

Kate October 2, 2012 at 11:30 pm

I’d be interested to see a geographic distribution of the demographics presented here.

Comments on this entry are closed.

Previous post:

Next post: