Michael Barone on the polls

Michael Barone offers some of his deep experience and wisdom about the polls. He throws more cold water on the idea of a Bradley Effect inflating Obama's standing in the polls now.

There is so much uncertainty today in the polls. Pollsters really are having trouble figuring out how to indicate what the turnout will be. Gallup has resorted to offering two models, one based on their traditional turnout model and one based on increased turnout by people who haven't voted before and wouldn't previously make their "likely voter" screen.

Another caveat that Barone offers is about the exit polls. He notes that Obama has generally done better in the exit polls in the primaries than he did in the actual vote. It may be that Obama voters are more willing to talk to the exit pollster. There have been plenty of problems in recent elections with the exit polls.

David Paul Kuhn reported earlier this week that the TV networks were sweating out their use of the exit polls this year. They're going to isolate their analysts so that the exit polls won't leak before at least 5 pm. So we won't have the experience like we had in 2004 with Senator Kerry convinced that he had won simply because he had done better in the exit polls.

And the pollsters are trying to correct the problems they had in the past with these temporary workers not following the rules about asking voters according to the formula but rather gravitating to voters that seemed more congenial or eager to interview.

In theory, exit polls should match election results. But for all the care that goes into conducting accurate exit polls, errant results aren’t completely uncommon. Respected polling analyst Mark Blumenthal found that during the Democratic primaries this year, preliminary exit polls overestimated Obama's strength in 18 of 20 states, by an average error of 7 percentage points, based on leaked early results.

The reason? Obama’s supporters were younger, better educated and often more enthusiastic than Hillary Clinton's, meaning they were more likely to participate in exit polls.

Insurgent candidate Pat Buchanan’s support also was overstated in the 1992 New Hampshire Republican primary, a phenomenon attributed to the greater willingness of his impassioned supporters to participate in exit surveys.

More recently, in 2004, exit poll data that began circulating early in the afternoon led to short-lived Democratic elation and deep Republican anxiety. By evening, some of President George W. Bush’s key strategists were frantic, emailing reporters at polling organizations to better understand the gap between what they were finding on their own and what the leaked exit polls indicated.

As it turned out, preliminary exit polls overstated women’s turnout that year. This “programming error,” which affects the statistical method that pollsters use to match surveys to the electorate’s composition, was discovered by the third wave of exit polling.

By then, though, the premature polls had already been leaked online.

When you're depending on thousands of temporary hires to spend the day asking people questions outside of voting booths, there is a lot of room for error and bias. Add in the high numbers of early voters whom the pollsters are trying to phone poll, but that is going to be iffy given how many calls they'll have to make get a sufficient sample. They're working on trying to correct the mistakes from previous elections, but the fact that there was so many errors in the exit polls during the primaries this year when, according to Kuhn, preliminary exit polls overestimated Obama's vote in 18 of 20 states by an average of seven points.

But just because the exit polls might not be predicative of the actual results, doesn't mean that today's polls are wrong. The beautiful thing for pollsters is that no one will ever know. If the polls today don't match Obama's final tally, the pollsters can always claim that there was a last-minute swing away from Obama. We don't know today if the greater numbers of Democratic voters showing up in many of today's polls represent those who will really show up or are overestimating their numbers. And of course the national polls are really not what is important now, but the state polls. They don't look good for McCain now, so they're either all biased in the same direction, or McCain is headed for a drubbing.