Kerry'd Away

Friday, October 22, 2004

Pollsters Getting Kerry'd Away
The pulse-takers have decided that Kerry would win the electoral vote if the election was held today. See eg: [Citation Expired]. They might be right. But if they are, it is a matter of luck more than science. More than at any time in the past 50 years the surveys are all skewed in unknowable directions and their "samples" are no longer representative.

Pollsters never get a fair sample of the "likely to vote" 18 to 30 year olds because there is such a high percentage of them that have no phone except a cell phone. Indeed, the "cell phone only" crowd is completely absent from all telephone poll results. That age group is a lot more conservative than it used to be. But are they really most likely to be Bush voters or Kerry voters? No one knows and no one can tell you because no one can ask them.

Telephone pollsters usually don’t talk to people with unlisted numbers. Those folks are probably both more conservative and more likely to vote than the general population.

Telephone pollsters often don’t get to talk to people with caller ID because people often choose not to pick up if it is an unknown phone number or obviously a pollster. Same situation with people who screen calls with answering machines. Folks with caller ID or who use answering machines to screen calls are probably also more conservative and likely to vote. The pollsters generally don’t successfully call people at work, so they only reach people who are either home during the day or who get home in good time in the evenings or are hanging around at home on the weekends. In other words, the unemployed or underemployed ... a group that tends to be more liberal.

And polls based on the general population skew left because a higher percentage of Republicans actually vote than do Democrats.This isn’t earthshattering news. Not many people talk about it, but it is widely understood that polling is a lot less accurate than it used to be.

In this election, though, there is another reason for poll results to be suspect. I believe that more than in any election over the past many years ... a lot of people are lying about for whom they will cast their vote.

They lie to their friends, to their family and to pollsters. To a lesser degree, perhaps, but yet to some very real extent, I believe many people are still lying to themselves as well.

A lot of people simply can’t stomach the idea of re-electing Bush. There is a lot of fury at him out there. And people that are mad will tell pollsters or whoever else that will listen, that they are going to vote to fire him. Which means that they go down in the "Kerry" column in the polls. But once they are within the sanctity of the voting booth, I believe that a substantial percentage of these folks are going to grit their teeth and vote for Bush.

If I’m right ... not only will the pre-election polls be wrong ... but the "exit interview" polls will be as well. It could be interesting! :-)