Sun Apr 25, 2004

Red and Blue America

Via Kevin Drum comes this comment from political scientist Hans Noel, quoted in the Washington Post:

“Most people say they are ‘moderate,’ but in fact the country is polarized around strong conservative and liberal positions.” [Noel said, and the article continues] … As it becomes more difficult to reach across the party line, campaigns are devoting more energy to firing up their hard-core supporters. For voters in the middle, this election may aggravate their feeling that politics no longer speaks to them, that it has become a dialogue of the deaf, a rant of uncompromising extremes.

Noel is pushing the attractive idea that polarization feeds on and reinforces itself. (Attractive from the point of view of elegant social mechanisms, I mean.) And Kevin can’t see a way to break the cycle. Red and Blue America is the latest version of the Culture Wars thesis. However, while it’s clear that the chattering classes—at least their representatives in the media— have become more polarized over time, I’m not sure I believe that everyone else has.

My main evidence for this comes from a 1996 paper by Paul DiMaggio, Bethany Bryson and John Evans called “Have Americans’ social attitudes become more polarized?” (JSTOR link102:3%3C690:HASABM%3E2.0.CO;2-M), institutional subscription required).[1] They used a long time-series of General Social Survey opinion data and measured how skewed the distribution of public opinion on a wide range of questions was, and whether that changed over time. Respondents to opinion questions are generally given a statement and asked to choose a place on a five- or sometimes seven-point scale ranging from “Strongly Agree” to “Strongly Disagree.” If polarization was happening, you would see more and more people at the extreme ends of the scales and fewer in the middle. But DiMaggio et al found that, with the exception of questions about abortion, the distribution of opinion had not become more skewed. Across a wide range of issues, there were about as many people in the middle in the early 1990s as there had been in the early 1970s. I don’t know of sample-based research that rebuts this finding. At the same time, as an an update by John Evans demonstrates, more recent data suggests that such polarization as does exist is being driven by the political system: “it seems clear that members of the public who are involved in politics are becoming polarized on moral issues.”

fn1. Full disclosure: Paul was my Ph.D advisor and John and Bethany are friends of mine.