ALEX CHADWICK, host:
A poll released late last month by Stanford University for the Associated Press and Yahoo News concludes that racial prejudice could cost the senator six points on election day. Charles Henry is a professor of African-American studies at the University of California, Berkeley.
More than 20 years ago, he first identified what's called the Bradley Effect. Tom Bradley, the black mayor of Los Angeles in 1982 running for governor in California, led in all the polls but lost the election. Professor Henry, welcome to Day to Day. And the Bradley Effect actually describes not how people vote, but what they tell pollsters, is that right?
Dr. CHARLES HENRY (African American Studies, University of California, Berkeley): Yes, that's what it's come to be called. Now, of course, at the time, we were trying to find the factors that might have resulted in Bradley's defeat. And my study in '83 tried to control for all those extraneous issues on the ballot, like handgun initiative, a nuclear freeze etcetera, and focus on race, and that was the most likely factor we found in Bradley's defeat. You just - a minute before the polls closed, the field poll indicated he would win by 10 percentage points, and, of course, he lost by 93,000 votes.
CHADWICK: Well, there is a kind of a subtext to this election, people writing in many places that, in fact, Senator Obama needs to be leading in the polls by eight, nine, 10 points if he really hopes to win because there may be this lingering Bradley Effect. Others say, is it really still valid? Because a lot of elections in the last 10 years, polling seems to predict pretty accurately how black candidates are going to do.
Dr. HENRY: It's certainly true that times have changed. We have a black governor in Massachusetts now. We had Doug Wilder in Virginia. But some of the research done on the primary votes indicates that, in 18 primary cases, it was the percentage of the population in the state that either over-predicted or under-predicted how Obama would do. That is that the polling results in 18 cases exceeded the margin of error for these polls. And the racial population in the state was the best indicator of what direction that error went.
CHADWICK: This is a delicate issue for the campaigns of both senators, Obama and McCain. How do you think they're doing with that?
Dr. HENRY: Well, I think that...
(Soundbite of laughter)
Dr. HENRY: As Senator McCain's folks have said, the gloves have come off. And what I'm hearing in Sarah Palin's speeches yesterday about he doesn't represent us. He doesn't understand our viewpoints, the small town values. He's an elitist. He doesn't think like you. He doesn't have your experience. He's not as positive about America. Those can be implicit cues that sort of trigger a negative reaction in terms of race. You know, it's been called playing the race card. I think that's probably more famous than the Bradley Effect. So those implicit appeals, I think, we're going to see more and more of in the McCain campaign.
CHADWICK: And what about Senator Obama's campaign? How is he handling all this?
Dr. HENRY: Well, I think the strategy is the correct one right now, and that is to keep trying to break the - bring the campaign back to the issues. And, of course, the issue that does best for him is the economy. And I think he will do well to sort of get away from character assassination, not sort of get down and sling the mud with McCain, but continually try to bring it back to - while pointing out that, indeed, this is what the McCain campaign is doing trying to do, as he said, swift boat tactics and using some of the same people that were involved in swift boat. I think he's got to try to bring it back to economic issues, where that's gained - where he's gained his momentum in the last couple of weeks.
CHADWICK: Charles Henry, a professor of African-American studies at UC Berkeley. He identified the so-called Bradley Effect 25 years ago. Professor Henry, thank you.
Dr. HENRY: Thank you. It's good to be with you.