The Demographics of Election Year Polling

The proliferation of news media in the United States has been accompanied by a proliferation of public opinion polls, which are all now trying to tell us how the presidential election is going to turn out. Indeed, Nate Silver, surely one of the smartest people tracking elections, complained recently that 

Following the polls on Wednesday reminded me of the aphorism: “If you don’t like the weather in Chicago, wait five minutes.” When there are twenty or more polls published in day, as there were on Wednesday, there are necessarily going to be some stronger or weaker ones for either candidate.

A real problem with polling is that the world has gotten very complicated. It is not as easy as it used to be get people on the phone, and the country has more demographic differences than there used to be. For this reason, pollsters get ahold of whomever they can and then they weight the results by various demographic categories, including age, sex, race/ethnicity, and in some cases political affiliation. The results then reflect not only how good a polling firm is in getting people to answer their phone (and several firms now have cell phones in their database), but also on the way in which they weight those demographic factors. This became an issue this week because on Fox News, Dick Morris (who advised Bill Clinton when he was President, but now has turned on Democrats) claimed that Mitt Romney was actually ahead in the polls, rather than losing, because undecided voters never vote for the incumbent. As David Weigel noted on Slate, this was a a knee-slapper:

According to Morris, literally 100 percent of "undecided" voters will eventually vote against Barack Obama. Leaving aside how the theory ignores spillover to write-ins and third-party candidates, this is one of the stupidest things I've ever heard. Nate Silver, bless him, has already explained why the "incumbent rule" doesn't actually exist.

The lesson in all of this is that polling results are not easy to disentangle on our own, so we really need some professional guidance. In my opinion, the best guide continues to be Nate Silver, whose NY Times blog continues its geography of the election series, which I mentioned previously, when the series began this summer.