Study: 'Bradley effect' didn't skew polls

ANN ARBOR, Mich., March 31 (UPI) -- Poor methodology, not the "Bradley effect," undermined the accuracy of polls that wrongly predicted U.S. presidential primary winners, a study indicates.

In what is being called the most comprehensive analysis conducted of presidential primary polls, University of Michigan researchers said the Bradley effect wasn't to blame for polls that failed to predict the victory of U.S. Sen. Hillary Clinton, D-N.Y., over Sen. Barack Obama, D-Ill., in the 2008 New Hampshire Democratic Party presidential primary, the university said in a release Tuesday.

The Bradley effect, named after former Los Angeles Mayor Tom Bradley, posits that people tell pollsters they support a black candidate in order to appear unbiased but cast ballots for a white candidate in the privacy of the voting booth.

"When Clinton won, some people pointed to latent racism as the reason," said researcher Michael Traugott. "But in the data we have from a wide variety of New Hampshire pre-election and exit polls, we found no evidence that whites over-represented their support for Obama."

Instead, the New Hampshire poll results were skewed in part because the election occurred only five days after the Iowa caucuses, truncating the polling period, the study found.