In the darkness of shutdown, brinksmanship, and default, NBC News and Esquire think they’ve found the last great hope for American politics — The Center.
The independent voter is a myth, at least as far as the science is concerned.
Everything we know about politics is wrong and America isn’t hopelessly divided is the gushingly optimistic conclusion of their widely publicized (and criticized) poll released Tuesday. “Emanating strongly from this rich and complex set of data from which the most complete and useful portrait of the new American Center has emerged comes this theme, expressed in a dozen different ways: a demand for the classic American notion of fairness,” according to Esquire.
The survey draws a circle around four groups of Americans, out of eight, it has determined to be in the middle (they are the vaguely titled “minivan moderates,” “MBA middle,” “Pick-up Populists,” and the “Whatever Man”). When joining these groups together, the poll finds majorities or near majorities of Americans agree on fuzzy concepts such as “the political system is broken” and “the economy is bad,” as well as on more specific issues, such as supporting gay marriage and legalizing pot.
But under scrutiny, the center doesn’t hold. These four center groups don’t agree uniformly. What does it mean when only a plurality within four of NBC/Esquire‘s eight political groups agree on something? For instance, there is a 16-point spread between center groups on the role of government in individual lives — perhaps the most fundamental question possible.
But nitpicking at the methodology of the poll and its premise misses a bigger point: Of course there is a center — ideas a simple majority will agree to — but that doesn’t mean America isn’t firmly sorted into political camps.
And here’s why: The independent voter is a myth, at least as far as the science is concerned.
“The folks who we see as independent, we think of them as closet partisans who act in almost indistinguishable ways to those who identify as partisans,” Brendan Nyhan, a political scientist at Dartmouth, told me a few weeks ago.
Recent research in the Personality and Social Psychology Bulletin illustrates this idea. In the study, the researchers employed what’s called an “Implicit Association Test,” which manipulates respondents to indicating what they really believe, or are otherwise unwilling to admit due to social pressures. For example, a smoker might not admit how many packs he smokes a day because smoking a lot isn’t socially desirable. You have to probe their brain a bit deeper.
How this works is kind of complicated, but basically implicit association tests whether you have a basic negative or positive attitude toward a statement. For instance, a phrase like “Full Medicaid Coverage” will be presented with another concept like “good” or “agony.” If “medicaid” and “good” are an association in your mind, you’ll respond to the prompt slightly faster, because that association is easier for your mind to process. Crazy stuff.
Those who were implicitly Democrat tended to side with Democratic issues. Same goes for the implicit Republicans. “In their voting patterns, independent leaners are almost indistinguishable from their respective partisan blocs, even though they decline to identify as party members,” the authors write.
So most of us, whether we admit it or not, have a political preference. And here’s another reason why the NBC/Esquire premise is flawed: What matters most, in terms of predicting political behavior, is how people view themselves — and not what they actually believe on individual issues.
For instance, researchers at the University of North Carolina found that college-age participants overstate their conservativeness. That is, when asked about their political orientation they indicate one thing, but then on specific policy questions, they move to the left. “Self-reported political orientation was significantly more conservative than political orientation scores assigned to subjects using a more objective process,” the paper concludes. The research suggests there is stronger in-group pressure for conservatives to stick with the conservative label, even when their beliefs would lead them elsewhere.
So, yes, these respondents are in the center (leaning toward the middle while maintaining a foot on a conservative base). And yes, that does give credence to the idea that Americans aren’t so divided. But here’s the crucial part: That doesn’t mean they’ll change their political behavior. “Biased self-perception,” the authors write, “predicted voting behavior in the 2012 presidential election even after controlling for objective political-orientation scores.” So it doesn’t really matter what they believe, the authors are concluding. Sorting people by shared belief is a failed exercise.
There are some voters who are truly in the middle, Nyhan says, but they are usually the least engaged in the political system, and are usually the least knowledgeable about it. “So the knowledgeable folks, the ones who follow politics the most closely, end up acquiring a set of beliefs and come to support one side or the other.”