(This article was originally published at Statistical Modeling, Causal Inference, and Social Science, and syndicated at StatsBlogs.)
Here’s some psychology research that’s relevant to yesterday’s discussion on working-class voting. In a paper to appear in the journal Cognitive Science, Andrei Cimpian, Amanda Brandone, and Susan Gelman write:
Generic statements (e.g., “Birds lay eggs”) express generalizations about categories. In this paper, we hypothesized that there is a paradoxical asymmetry at the core of generic meaning, such that these sentences have extremely strong implications but require little evidence to be judged true. Four experiments confirmed the hypothesized asymmetry: Participants interpreted novel generics such as “Lorches have purple feathers” as referring to nearly all lorches, but they judged the same novel generics to be true given a wide range of prevalence levels (e.g., even when only 10% or 30% of lorches had purple feathers). A second hypothesis, also confirmed by the results, was that novel generic sentences about dangerous or distinctive properties would be more acceptable than generic sentences that were similar but did not have these connotations. In addition to clarifying important aspects of generics’ meaning, these findings are applicable to a range of real-world processes such as stereotyping and political discourse.
Jonathan Haidt’s generic statement was, “working-class people vote conservative.” Actually most working-class people don’t vote conservative, but as Cimpian et al. point out, such a statement “requires little evidence to be judged true.” At the same time, “these sentences have extremely strong implications,” leading Haidt to chase down all sorts of explanations for political patterns he is not describing accurately.
Consider this statement from Haidt:
When working-class people vote conservative, as most do in the US, they are not voting against their self-interest; they are voting for their moral interest.
Wrong on three counts. First, most working-class people in the U.S. don’t vote conservative. Second, working-class people are diverse and make their vote decisions for different reasons. First, there’s lots of evidence that people vote based on what they think is good for the country, not on their interests. Many Americans of all classes and income levels are conservative in their economic ideology (that is, they support low taxes and limited government).
What happened to Haidt? I suspect the psychologist got caught in a psychological trap: He interpreted a novel generic such as “Working class people vote conservative” as referring to nearly all working class people, but they he judged the same novel generics to be true given a wide range of prevalence levels (e.g., when only 40% of lorches working class people voted conservative.
Haidt is a respected researcher, and he might want to think carefully about how he made such a basic mistake about American politics as to write—not just to write, but to base an entire column on—the false statement that most working-class Americans vote conservative. Sure, various casual readers of Thomas Frank drew this conclusion too, but Thomas Frank is an activist. Haidt is a scientist, he should know better. Findings from cognitive psychology such as in the above-linked paper might give Haidt some insight how he can be fooling himself. (And, yes, I’m sure I’m subject to these sorts of mistakes too, especially when I’m not focusing on them.)
P.S. This research by John Huber and Piero Stanig is relevant to Haidt’s original discussion of lower-income conservatives. You can also check out the chapter on religion and voting in Red State Blue State.
Please comment on the article here: Statistical Modeling, Causal Inference, and Social Science
