Obama

NORTH PROVIDENCE –– These people say they have no need to lie to pollsters. They say they’re not ashamed of their views.

While waiting for bingo to begin, a handful of senior citizens inside the Salvatore Mancini Resource and Activity Center last week denied that race would influence their vote on Election Day, when the nation will decide whether to elect its first African-American president.

“We’re all God’s children –– the blacks, the whites, the Portuguese,” says Dolores Clemente, a 78-year-old retiree.

But she knows that everyone doesn’t share her tolerance. And as a lifelong Democrat, she fears her preferred candidate will become a victim of something political scientists refer to as the “Bradley effect.”

“People say they’re going to vote for him, but I don’t know if they’re going to,” Clemente says. “A lot of people might turn around and say, ‘We’re not going to vote for him because he’s black.’ ”

Clemente is not alone. A recent study cites Rhode Island as proof that the Bradley effect is alive and well.

While the Obama campaign dismisses such concerns, several studies suggest that voters have a tendency to lie to pollsters when they plan to vote for a white candidate instead of a black one. The phenomenon is named after Los Angeles Mayor Tom Bradley, an African-American, who narrowly lost the 1982 gubernatorial election to a white candidate after being favored by 7 percentage points just days before the election.

Bradley is just one example.

In 1989, African-American Virginia gubernatorial candidate Douglas Wilder was elected by less than 1 percentage point, despite leading by 9 points before the election. That same year, David Dinkins narrowly won the New York City’s mayoral race after leading by 18 points.

And in Rhode Island, polls the week before the presidential primary election suggested that Barack Obama trailed Hillary Clinton by between 5 and 9 percentage points. The former first lady ultimately defeated her African-American opponent by 18 points.

“Do I think racism is still alive in Rhode Island? Absolutely,” says state Rep. Joseph Almeida, who is black. “We have a long way to go.”

There is little doubt that racial attitudes have changed considerably over time. A December 2007 national Gallup poll found that 5 percent of white respondents said they would not vote for a black candidate for president. In 1958, Gallup determined that 58 percent of whites would not cast a ballot for a black presidential candidate.

But Rhode Island wasn’t the only possible example of the Bradley effect this year.

Pre-election polls also overstated primary election support for Obama in California and New Hampshire, according to University of Washington psychologist Anthony Greenwald, who recently analyzed data from 32 states, determining that race played an “unexpectedly powerful role in distorting pre-election poll findings.”

“People understand the Bradley effect in different ways,” Greenwald says.

Some believe that people deliberately lie to pollsters to hide inherent racist attitudes, he says. Others believe that people simply avoid participating in polls “if they don’t want to vote for the black.” Either way, the pre-election polls would be skewed.

Is he sure that the Bradley effect is to blame for Rhode Island’s discrepancy in March?

“It could be,” Greenwald says. “Let’s just say there is a racial effect going on.”

Brown University political science Prof. Jennifer Lawless, an author of one of the polls that overstated Obama’s Rhode Island support, thinks other things may be at play as well.

“While it’s plausible to conclude that the Bradley effect may have played some role, I don’t think it accounts entirely for how off many of the polls seem to have been,” she says. Other factors may include overestimated turnout among young people, she says, in addition to the Clinton campaign’s ability to get out the vote on Election Day.

“The Clintons have deep Rhode Island ties and most of the political establishment backed Hillary Clinton’s candidacy. It’s hard to quantify what effect this had, but it shouldn’t be underestimated in terms of last-minute get-out-the-vote efforts,” Lawless says.

Indeed, there is considerable debate about the impact of the Bradley effect, or whether it even exists. A Harvard University study released in August found evidence of the phenomenon in senatorial and gubernatorial races across the country through 1996, accounting for a 2.7 average discrepancy between poll numbers and voting patterns. But the study determined that the Bradley effect has since disappeared.

Pollster Andrew Smith, director of the University of New Hampshire Survey Center, is also skeptical. He says there’s no clear evidence that the Bradley effect explains why primary election polls there overestimated Obama’s support by more than 7 percent this year.

But Smith notes that a significant portion of New Hampshire’s electorate believes race will influence the election.

Almost one quarter of Granite State voters said they had heard a friend, family member, or coworker say they would not vote for Obama because he is black, according to a poll he conducted last month. Further, 9 percent say many people will not vote for Obama simply because he is black; 6 percent say many people will vote for Obama because he is black.

Indeed, Greenwald’s analysis found that there was, in at least 12 states, a “reverse Bradley effect” that actually underestimated support for Obama.

“Blacks understated their support for Obama and, even more surprising, whites did too. There also is some indication that this happened in such Republican states as Montana, South Dakota, Oklahoma, Missouri and Indiana,” Greenwald said. Overall, he believes the reverse Bradley effect more than compensates for the Bradley effect in places like Rhode Island and New Hampshire.

Whether it’s called the Bradley effect, or the reverse Bradley effect, pollsters acknowledge poll respondents have a natural tendency to tell the person on the other end of the phone what he or she wants to hear.

For example, Smith says, more respondents often say they are going to vote than actually do. And when asked, many downplay their use of alcohol or lie about their weight, even though the surveys are completely anonymous.

Smith expects the Bradley effect to play a greater role this week in key swing states such as Ohio, Indiana and Pennsylvania. Those are “areas with a large blue-collar white population that I think is more likely to have some racist attitudes,” he said, adding that he doesn’t expect race to be a deciding factor for most people. “My sense is that most people are going to vote for or against Obama because they are Republicans or Democrats.”

Meanwhile, back in the Salvatore Mancini Resource and Activity Center, 70-year-old James Johnson says he’s never voted for a Republican before. But he plans to this year.

“I’m not voting for [Obama]. I’m not speaking Arab. … If he wasn’t Muslim, I’d probably vote for him,” says Johnson, a retired industrial engineer.

Despite rumors on some conservative blogs, Obama is a Christian, not a Muslim. Still, concerns about his race and religion are very much alive in Rhode Island.

It’s unclear, however, whether such concerns will keep Obama from becoming the nation’s first African-American president.

Obama currently leads Republican candidate John McCain by 11 points, according to a CBS News/New York Times poll released Thursday. A separate Fox News poll released the same day gives Obama a lead of just 3 points.

In the latest Rhode Island poll last month from Opinion Factor, Obama led McCain by 22 points.

“We won’t know about the extent of the Bradley effect until the polls close on Tuesday night,” Lawless says. “But my hunch is that race will play a significant role in this election.”

speoples@projo.com