What are the polls really telling us about who will win this Tuesday?

In Sunday’s Review-Journal we will carry poll results from eight key battleground states produced by our longtime pollster Mason-Dixon Polling & Research. The polling was conducted for various media outlets around the country.

The states on which we will report are Nevada, Colorado, Florida, Missouri, North Carolina, Ohio, Pennsylvania and Virginia. Because all the media clients of Mason-Dixon agreed to release the results at the same time, I can’t reveal any specific numbers until Sunday morning.

I can tell you it is a tight race with Barack Obama leading in some states and John McCain in others, but with enough undecideds to determine the outcome. In the past I’ve generally assumed the undecideds would be fairly split at the close of voting on Election Day.

But the pundits are all atwitter over what has been labeled the Bradley effect. The phenomenon takes its name from the 1982 gubernatorial race in California in which the Democratic African-American mayor of Los Angeles, Tom Bradley, was well ahead in all the polls but lost the election. It is assumed that many whites told pollsters they were undecided rather than admit they were not voting for a black candidate.

Of course, some observers note that there was a gun control initiative on the ballot that attracted a large turnout by Republicans. And it was 26 years ago. Perhaps times and attitudes have changed.

Pollster Brad Coker, managing director of Mason-Dixon, points out other races with similar outcomes — African-American Doug Wilder’s 1989 Virginia governor’s race, the 1990 Senate bid in North Carolina by African-American Harvey Gantt against incumbent Jesse Helms and the 1990 contest for a Louisiana Senate seat between incumbent J. Bennett Johnston and Republican and former KKK leader David Duke.

In Virginia, Coker recalls, Wilder was leading by 4 points with 8 percent undecided. Of those, 96 percent were white. On Election Day, Wilder won by less than half a percentage point.

In North Carolina, Coker’s own Mason-Dixon poll showed Gantt 4 points ahead of Helms, also with 8 percent undecided, and 92 percent of those were white. Helms won by 6 points. Faulty polling or the white undecideds broke for the white guy?

In Louisiana, Coker’s poll showed Johnston leading Duke 53 percent to 26 percent with a third candidate getting some support and 13 percent undecided. Again, they were 90 percent white. Duke garnered 44 percent of the vote.


“So, the million dollar question is will there be a ‘Wilder/Gantt/Duke Effect’ in the 2008 presidential race?” Coker asks. “No matter what anyone theorizes, the answer today is that no one knows for sure. The three races I cited were all almost 20 years ago, and changes in the country may have mitigated many factors regarding race and voting. But, since this is the first time an African-American has been this close to the winning the presidency, it would be foolish to completely rule it out.”

Pollster Coker noted that in the eight battleground states the Review-Journal will be reporting on Sunday the vast majority of undecideds are white.