Are Pollsters Undercounting Obama Voters if They Don't Call Cellphones?

Amid questions over how much public opinion polls on the race between John McCain and Barack Obama are being affected by racial prejudice and the unpredictability of new-voter turnout, a new study on the youth vote and cellphones suggests there could be another gap in traditional opinion polls—one that could mean Obama has more support than other surveys estimate.

The Pew Research Center for the People and the Press, in three separate surveys this summer, contacted users not only of traditional land lines but also of cellphones. Each time, Obama came out 2 to 3 percentage points higher than when Pew called land lines alone.

Opinion surveys are usually adjusted to account for age differences, so the discrepancy cannot be explained simply by the fact that people who use only mobile phones are younger. It's also, Pew found, that youths who are strictly wireless—nearly half of the 18-to-29 bloc—are 10 percentage points more supportive of Obama than their counterparts who retain land-line phones.

Why would this be? "Demographics," says Michael Dimock, Pew's associate director for research. Although Obama's plugged-in political campaigning could be a factor, it appears to come down to the kinds of voters he's attracted. Cell-only youths tend to be at the younger end of the group, more likely to be unmarried, and more mobile than those on land lines.

Not all polls call only land-line users. Gallup calls cellphones, as do New York Times/CBS pollsters. Meanwhile, another study on the issue last week by ABC News/Washington Post found that calling cellphones affects poll results only negligibly.

Dimock also cautions against overdramatizing Pew's findings. Two points, after all, isn't that much. "Any pollster would say, 'Well, you're really working in the margins here,'" Dimock says.

Still, even the margins could matter, given how close the election appears to be, particularly in several key swing states.