Are Cell Phones Screwing Up Polls?

You've surely heard about the wide discrepancy in some of the recent surveys for the Presidential campaign, with some showing a dead heat between Bush and Kerry, while others give Bush a comfortable lead. One intersting explanation is that pollsters are missing out on a significant bloc of voters: those who have cell phones but no land line. FCC rules prohibit pollsters from making unsolicited calls to mobile phones. However, the number of people--young adults, mostly--who opt for a mobile phone instead of a land line is growing. New rules allowing cell phone customers to keep their numbers will likely exacerbate this trend, making phone-based random surveys less and less reliable.

This issue confounds pollsters in another way. Because they must omit any number that might be a cell phone, pollsters are less able to use automated dialing systems, returning to manual dialing. This was my first job out of college... and believe me, it was no fun!

So what's the solution? Some have proposed a national opt-in list that would give pollsters permission to call cell phones. Others are suggesting a return to door-to-door and "man on the street" polling. I can't imagine any of these solutions being successful this day and age.

Pollsters, by the way, are exempt from restrictions imposed by the federal Do Not Call Registry.