September 27, 2008


The Bradley Effect (WILLIAM SAFIRE, 9/28/08, NY Times Magazine)

The root of the phrase is in the campaign for the governorship of California in 1982. Surveys up to and including exit polls reported that Tom Bradley, the first black mayor of Los Angeles, was well ahead of George Deukmejian, the Republican. But the popular mayor lost by 1.2 points. How could that happen? Speculation ranged from inaccurate sampling, to last-minute mind-changes, to latent racism, to freely lying voters, to the reluctance of those being polled to admitting a preference that may be socially unacceptable — anti-black — in talking to interviewers.

Those impressed with the Bradley effect (put “so-called” in front if you dispute it) point to a series of polling surprises in races between candidates of different races. In 1989, David Dinkins won the New York mayoralty with a two-point margin after polls gave him a double-digit lead; on the same day, Douglas Wilder, who had been ahead by 15 points in the pre-election weeks, squeaked through to win the governorship of Virginia by less than 7,000 votes.

“My lesson was learned in 1978,” says Tully Plesser, the veteran Republican pollster telling me about that year’s Massachusetts race for senator. “Late polling for the Republican incumbent Senator Edward Brooke had him nearly 8 points ahead of Democrat Paul Tsongas, and Tsongas was elected by a 10-point margin.”

Is there such a thing as “the so-called Bradley effect”? Some on the right think so: Wesley Pruden in his column in The Washington Times defines it as “the phenomenon of black candidates to register significantly better in public-opinion polls than on Election Day.” Others differ. Daniel Hopkins, a postdoctoral fellow at Harvard, issued a lengthy scholarly study last month of what he calls the “Wilder effect” and concludes that it was significant “only through the early 1990s” and “disappeared swiftly at about the time that welfare reform silenced one critical, racialized issue.”

Wait — there’s another linguistic wrinkle. Martin Peretz argues in The New Republic that all the talk about the Bradley effect has an impact on the race, which he has named “the Bradley-Effect Effect,” which “actually benefits Obama. Is it so crazy to think working-class voters will react to the racism charge by going out of their way to prove it false?”

Yes, it actually is crazy to think that a Kansan will act like a Beltway liberal, even assuming that Mr. Peretz won't pull the R lever in the privacy of the booth.

Posted by Orrin Judd at September 27, 2008 8:40 AM
blog comments powered by Disqus