Campaign Diaries

Some Democrats have long been worried that Barack Obama’s race will complicate his electoral prospects, but it is important to not conflate two different phenomenons. On the one hand, there is the question of how much Obama is underperforming because of the reluctance of white or Hispanic voters to cast a ballot for a black man; the AP generated a lot of talk last week by positing that number to be 6% based on a poll they conducted. On the other hand, there is the question of the Bradley effect: how much is Obama’s support overrepresented in the polls?

The former question was an interesting one to debate a few months ago, when Democrats were trying to decide which of their candidates was more electable; but it does not impact the way we should read today’s polls. If a white voter who would have voted for another Democrat but not for Obama is willing to tell that to a pollster, his reluctance is already incorporated in the polls and in no way hints at the likely existence of a Bradley effect.

It is important to keep that in mind when reading about that AP poll or about stories like this one in the New York Times. The piece documents the reluctance of some union workers’ to embrace Obama, and labor organizers are saying that some of their members openly admit they will not vote for a black man; others use sentences like “I don’t know him” which some see as racially charged code words. But in both cases, what is striking is these voters’ openness to admit they are not voting for Obama. Their voting for McCain would not cause a gap between Obama’s Election Day performance and his level in polls.

In a sense, it is reassuring for Democrats that Obama is leading by healthy margins while a number of voters who might have voted for another Democrat are admitting they will not cast a ballot for him. It suggests Obama has already taken the blunt of the AP’s 6% and is still in the lead.

The question that we should answer to determine whether a Bradley effect is likely is whether any white voters are lying to journalists, canvassers and pollsters and saying they are supporting Obama when they have settled on McCain because of racial prejudice.

This is possible, of course. While a recent academic study shows that the Bradley effect has disappeared since the mid-1990s, a presidential election is a whole other ball game and might make some voters uncomfortable. But, as I said already, the question isn’t whether racial prejudice exists but whether the voters are likely to feel ashamed enough they are not voting for Obama to lie about it. If anything, the more such voters have other excuses to give (there might be less social clues against admitting you aren’t voting for a Muslim, for instance, than against admitting you aren’t voting for an African-American), the less they are likely to feel like they ought to deceive pollsters.

In short, whether Obama’s lead is smaller than it could be because of race is one question - and the AP poll and stories like the New York Times suggest it would; whether Obama’s support will be lower in final results than it looks today is another.

Meanwhile, in news from battleground states:

• Ed Koch campaigns for Obama in Florida, in what could be a golden opportunity for Obama to address his vulnerabilities among older Jewish voters.

• Jon Ralston of the Las Vegas Sun is surprised Nevada polls are still finding a toss-up.

• The Miami Herald reports on Florida’s Hispanic community and finds that the GOP’s hold is still strong.

• A bizarre story out of Indiana has the Obama campaign scrambling to deny they made a deal with GOP Governor Mitch Daniels.

• Ballot questions are rarely covered, but the New York Times explores the possibility that Massachusetts voters might

• Limbaugh attacks the Obama campaign for having enlisted prosecutors in his Missouri truth squad (though the McCain campaign has done the same), creating complications for a Missouri reporter as tens of thousands of callers call in to complain about Obama’s plan to arrest old ladies.

And then there is everything related to early voting - starting with Ohio, where the state’s first early voting program starts in 24 hours! This is really make or break time for both campaigns in one of the most crucial state’s of the election, and there is no room for error. The Columbus Dispatch reports on the frenzied activity both campaigns are undertaking, and on the fact that demands for absentee voting ballots are already hitting records in some counties. One potential wild card: The GOP is suing to stop voters from registering and immediately voting (as the Secretary of State ruled they could).

Meanwhile, we are getting some of the first breakdowns of early voters by party, and they have Colorado Republicans worrying: Absentee balloting traditionally favors Republicans in this state by heavy margins, and Democrats rely on Election Day voting to close the gap, but many more Democrats than usual are requesting ballots this year. In Arapahoe County, for instance, Republicans cast 48% of mail-in-ballots in 2004, versus 30% for Democrats; as of last week, however, 40% of mail-in-requests had come from Republicans versus only 37% from Democrats.