11/02/2008 05:12 am ET | Updated May 25, 2011

Battle For The West: Cell Phones Missed By Polls


If Nevada continues to be one of the handful of pure toss-up states, why the dearth of polls in the state, asks UNLV political science professor David Damore in PolitickerNV. "Two of the primary issues associated with polling in Nevada are the sporadic nature by which the polls are conducted and virtually all of the polling is done on the cheap, yielding small sample sizes and large margins of error," Damore writes.

RealClearPolitics lists 15 presidential polls in Nevada since February; for comparison, fellow swing state Colorado has had 29 polls at last count.

The latest polls still show John McCain ahead, but on the ground Nevada's starting to feel blue. With Democrats leading Republicans in registration by more than 76,000, why is the race still so tight -- and McCain ahead -- even in the few polls available in the state?

"The folks who end up in these samples tend to lean towards the old and reliable. In this case of Nevada, this means that those more likely to be sampled are those who regularly vote, hold stable addresses, and rely on landlines (e.g., rural voters who overwhelmingly favor the GOP)," Damore explains. In contrast, many of Obama's most fervent supporters are "young, first time, cell phone only voters" who are difficult to reach by pollsters.

By the end of 2007, 15.8 percent of U.S. households were cell-only, more than triple the figure from just four years ago, according to government estimates. Though pollsters massage data with complicated weighting schemes, the rise of cell-only households poses a challenge. A 2007 report examining the cell-only problem by the Pew Center for the People & the Press sums up the challenge:

If people who can only be reached by cell phone were just like those with landlines, their absence from surveys would not create a problem for polling. But cell-only adults are very different. The National Health Interview Survey found them to be much younger, more likely to be African American or Hispanic, less likely to be married, and less likely to be a homeowner than adults with landline telephones. These demographic characteristics are correlated with a wide range of social and political behaviors.

And while a 2006 Pew survey concluded cell-only respondents answered similarly to landline users when asked about their party identification or if they'd vote for a generic Republican versus a generic Democrat, Pew's 2008 surveys show a stark difference. Landline respondents were split down the middle, with 45 percent each choosing McCain and Obama. But the cell-only respondents favored Obama 55 to 36 percent. The most recent report concludes:

Traditional landline surveys are typically weighted to compensate for age and other demographic differences, but the process depends on the assumption that the people reached over landlines are similar politically to their cell-only counterparts. These surveys suggest that this assumption is increasingly questionable, particularly among younger people... Pew's surveys this year suggest at least the possibility of a small bias in landline surveys. Such a bias could be consequential in an election that appears to be very competitive right now, especially if significant numbers of young people turn out to vote.

Read the rest of the Battle For The West Daily Digest here.