Forecasting Elections Topic Of Oct. 8 Taft Lecture

One expert’s method for forecasting national elections will be on display, when Helmut Norpoth from SUNY-Stony Brook visits UC to deliver his lecture on "Forecasting the Presidential Vote in 2004: The Predictive Power of Primaries."

Norpoth’s talk, part of the Taft Lecture series and sponsored by UC’s political science department, runs from 2-3:30 p.m. on Friday, Oct. 8, in the Max Kade German Cultural Center, located in Room 733 of Old Chemistry building. It is free and open to the public.

Norpoth, a nationally known expert on elections and a long-time consultant for "The New York Times," has developed a statistical model for predicting elections that uses the outcome of presidential primaries as the key predictor. Norpoth’s model, which also controls for two other factors, has picked the winner of every presidential election but one since 1912.

For 2004, the model predicts a George W. Bush victory. According to Norpoth, the statistical error associated with his model’s forecast comes to 2.7 percent, which translates into a likelihood of about 95 in 100 that Bush will be the winner in November.

Within Norpoth’s model, any time a candidate of the party that controls the White House has won the primary battle without a significant challenge, the odds are overwhelmingly in favor of that candidate’s victory in the November election. This goes back as far as 1912, when presidential primaries were first used in large numbers.

The forecast for 2004 would have predicted a Kerry victory in November if the current president had repeated the primary performance of his father in 1992. The challenge by Pat Buchanan in the Republican primaries that year predicted a loss for President George H. W. Bush in the general election.

For additional information on Norpoth’s lecture, contact UC Professor of Political Science George Bishop at 513-556-5078.