Polling Cell Phone Users and Not Polling Them

Brian Schaffner yesterday examined the issue of pollsters who are finding a wide Obama margin of victory and pollsters who are reporting a narrower Obama margin of victory. He observes that the pollsters who show the greatest margin for Obama are the ones who include in their sample persons who exclusively use cell phones (that is, don't have landlines.) Schaffner notes:

Obama's support increases by almost 3% in the national trend that includes polls reaching cell phone only respondents while McCain's support decreases by about 1%.

He includes some great visuals in his analysis that makes clear the problem. I tried to embed them here, but it didn't work, so I recommend you click through and take a look at the differences between the two groups of pollsters.