Cellphone voters: Out of sight, should be on candidates' minds

I'm a menace to my own bank account. On more than one occasion, I have found myself betting a relative or friend that Dino Rossi will become the next governor of Washington state. A few close political polls, mixed with the overwhelming impact of the change message, made me do it.

But just as the prediction tumbles out, perhaps too confidently, I stop and think about a most perplexing group of voters out there — cellphone-only voters who some pollsters contact and some do not. We dismiss them at our own peril.

In the 2004 election, about 7 percent of registered voters were cellphone only. Today, national experts believe the number is closer to 14 percent. I didn't even consider investing in a landline for my two young voters now at college. They are cell-only creatures; both already have voted.

What about the thousands of students at the University of Washington, the ones eligible to vote in Washington state? How many have landlines? What about Western, Washington State and so on? These colleges are small cities and some of their residents are off the radar of many pollsters.

I believe no one can accurately predict the super-close gubernatorial election without better understanding these voters. Younger voters are more likely to be cellphone-only, more likely to be Democratic, or at least voting heavily for Barack Obama. Yes, there has to be a connection to our gubernatorial contest.

One poll that consistently foretells of an effectively tied Washington governor's race is SurveyUSA, which polls for KING-5. Several of their polls have shown Obama dominating John McCain in the presidential contest in Washington state — 16 points, according to the most recent tally. At the same time, this poll shows Rossi and incumbent Gov. Christine Gregoire statistically even, with 1 point between them.

SurveyUSA does not contact cellphone voters. Pollsters are prohibited from machine-dialing cellphones. So as good as this poll can be, and some years it is very good, it may understate Gregoire's support.

Another poll, the Elway Poll, which also does not contact cellphone-only voters, says Gregoire is ahead by 12 points, with 10 percent undecided.

Rossi's private pollster and Gregoire's pollster both say the race is very close but won't say which candidate has the advantage. Rossi's pollster says he does call cellphones, Gregoire's pollster does not.

Some years, political campaigns can't resist sharing their polls. They love to brag about how well their candidate is doing, presumably to stir fundraising and overall voter enthusiasm. Perhaps the campaigns won't say who is winning because they are more focused on turnout, or, as we say in the vote-by-mail era, participation. A campaign has to be careful not to make a candidate's supporters too confident. Gregoire only wins if there is huge turnout. Rossi only wins if he gets his voters out.

Don't forget the talking heads on 24-hour TV will not wait until 11 p.m., Eastern time, when our polls close, to start hyping results in the presidential race in Virginia, Florida, Pennsylvania and Ohio. Early reporting could suppress some of our late voting.

So who are the cellphone-only voters?

For a while, experts thought these voters were not statistically much different from landline voters, with a slight edge among them for Democrats and Obama.

But researchers at the nonpartisan Pew Research Center say understanding of these voters is evolving. Pew has used cellphone voters in five surveys since August and now believe cellphone-only voters are a bit more Democratic than first believed. For example Pew's latest poll puts Obama above McCain nationwide by 14 percent based on landline and cellphone contacts. If the poll were conducted only with landlines, the lead declines to 11 points, a three-point difference.

"Young people we reach on landlines are somewhat different than young people we reach on cellphone," says Scott Keeter, director of survey research for Pew. Landline voters are more likely to own their own home, be married and trend a bit more Republican.

Three points isn't that big a deal in some races but in Washington state where we had the closest governor's race in history four years ago, three points might be huge.

So those darn bets stand, but they feel a little funny. Rossi has the flagging economy working on his behalf. And Gregoire may benefit from the cellphone-only stealth voters who are out of sight but not to be cavalierly dismissed.

Joni Balter's column appears regularly on editorial pages of The Times. Her e-mail address is jbalter@seattletimes.com; for a podcast Q&A with the author, go to www.seattletimes.com/edcetera