Interesting column by Jimmy Breslin yesterday. He notes that none of the polling companies actually calls people on cell phones.
Now, Breslin being Breslin, he takes what might be a barely interesting angle on statistical sampling and turns it into a diatribe against the polling companies and those media outlets that use polls. Breslin notes that there are nearly 170m cell phone owners in the US and none of them has been polled.
Breslin quotes John Zogby who says he doesn't use telephone polls anymore because they're outdated. Unfortunately, he doesn't reveal what method Zogby does use. I'm curious.
Breslin doesn't deal with a few obvious follow-on questions such as (a) what percentage of cell phone owners are of voting age? (if it's anything like Ireland, I'd say about half) and (b) what percentage of cell phone owners don't have a land-line?
And, what about the Do Not Call Registry? Polls "with no commercial purpose" are exempt from the Registry, but is it possible that people on the Registry's list are not as receptive to polls now? Or is it possible that some polling organizations are shying away from calling people listed with the Registry?
Undeniably, Breslin raises an interesting point about the polls. Are we getting skewed data? Are the polling companies selling us short because addressing these concerns would be too costly and too involved?
Obviously, we won't know until after the votes are counted, but these sorts of questions should be causing both campaigns at least some concern.