When You Omit Cellphone Users From Surveys, Do You Bias the Results Toward Republicans? - NYTimes.com

/via @creepysleepy:

Predictions of a solid Republican Mandate being delivered on November 2 may be a bit overblown. It appears that pollsters don’t call people with cellphones, and traditional demographic weighting methods have not been tested on this particular problem. We shall see just how off these polls are on November 3, but the 50 seat gain roundly predicted by most polling models may be quite overinflated. If it is just 10 seats off (reflecting even a modest 2 percent bias against Dems in landline only polls) Democrats will narrowly hold onto the House, and Republicans will probably scream robbery, because apparently the only legitimate elections are the ones where they win.

Here is an excerpt;

On Wednesday, Pew Research issued a study suggesting that the failure to include cellphones in a survey sample — and most pollsters don’t include them — may bias the results against Democrats. Pew has addressed this subject a number of times before, and in their view, the problem seems to be worsening. Indeed, this is about what you might expect, since the fraction of voters who rely on cellphones is steadily increasing: about 25 percent of the adult population now has no landline phone installed at all.

Clearly, this is a major problem in survey research — and one that, sooner or later, every polling firm is going to have to wrestle with. What isn’t as clear is how much of a problem it is right now.

Read more in the 538 blog, which apparently is now controlled by NYT.