July 18, 2008

Political Polling

Political Animal raises an interesting question for political poll watchers: are poll results skewed because pollsters do not call very many cell phones? This may not be a big issue because cell phones, while pervasive, are still seen by most as a supplement to and not a substitute for land lines. But no pollster is going to call me, ever, because I do not have a land line. Now, I have developed a somewhat peculiar set of political beliefs and attitudes and I don’t think that I fit nicely into a single box anymore, but that’s not the point – the point is that my opinions, whatever they are, are unlikely to ever be counted in one of those polls intended to test the political winds on things.

This is a more distressing notion for Republicans because Republican voters tend to be older, and cell phone users tend to be younger. So if there is a cell phone omission bias inherent in political polls, it will result in an underreporting of Democratic respondents. Mitigating this is the fact that younger voters are lower-propensity voters, so while they might say they support a Democratic candidate or a liberal/progressive cause, such a respondent is less likely to actually vote for the candidate in question.

No comments: