I first heard about this around a week ago, but could not confirm it, and in fact it suspected it was wrong, but it turns out that Gallup, for one, is not calling mobile phones for their polls, according to an article at Wired. Zogby started talking about this a while back; I thought it was market hype for his company, since I figured random dialing (which at least used to be common, to the best of my knowledge) would catch mobile phones with no problem. But Steve Hanway of Gallup confirms in the article that they are not calling those phones, based on anticipated antipathy from mobile phone users.
While this is sort of big news, the article claims that only 3% of persons in the country are using only mobile phones. If that number is right, it should have fairly little impact on the polls. I don't know the precise impact it will have; it will mostly be an issue for how confident one can be about the poll results. But 3% is not going to make a big dent in that. The article does mention that this 3% is not random, it is skewed toward a younger demographic, but, still, at only 3% (and as they again point out, among a group with lower voter turnout), it won't have a huge impact. I would be interested in seeing some numbers, or maybe if I suddenly invent some free time modeling it myself (which seems ridiculously optimistic regarding my statistical abilities).
It will be an increasing problem, though. But, I think for now it need not be a great concern, and I don't think any of need worry any more than usual about the numbers coming out of the current polls.
Want to comment? You'll need to create an account first.
