by Nate Silver
December 30, 2011
I’m not always a big fan of dissecting individual polls — mostly because there are liable to be a plethora of them in Iowa and New Hampshire over the next several days and their errors will tend to be mitigated as more are added to the average.
Nevertheless, the new CNN poll of Iowa contains a methodological quirk that is worth bringing to your attention and which will probably result in the survey underestimating the support for Ron Paul.
The issue is that CNN’s Iowa poll was conducted by using a list of registered Republican voters and registered Republicans only:
Sample was drawn from the complete list of registered Republican voters provided by the Iowa Secretary of State.
What’s wrong with using a list of Republican voters for a Republican caucus poll? The answer is that it’s extremely easy for independent and Democratic voters to register or re-register as Republicans at the caucus site. Historically, a fair number of independent voters do this.
According to entrance polls in Iowa in 2008, for instance, about 15 percent of participants in the Republican caucus identified themselves as independents or Democrats on the way into the caucus site. Although the way that voters self-identify is not technically the same thing as which party they are officially registered with, this is probably a good proxy for what percentage of voters changed their registration to Republican when they signed in at the caucus location.