The University of Iowa
The University of Iowa News Services Home News Releases UI in the News Subscribe to UI News Contact Us


100 Old Public Library
Iowa City IA 52242
(319) 384-0011; fax (319) 384-0024

Release: Aug. 30, 2000

UI professor issues final election forecast, says there's no doubt Gore will win

IOWA CITY, Iowa -- A strong economy and a popular president will propel Al Gore into the White House in November, says University of Iowa Political Science Professor Michael Lewis-Beck. Using an election forecasting model with a track record for accuracy, Lewis-Beck--in his final forecast before the election--says the Democratic candidate will win 55.4 percent of the two-party popular vote.

Lewis-Beck's election 2000 forecast is based on the same political economic model he used in 1996, when he correctly forecast the exact percentage of the vote President Clinton would win. The model combines economic conditions, the popularity of the current president, and a public opinion index of which party will bring about greater peace and prosperity in the future.

Lewis-Beck and Charles Tien, a former UI graduate student who collaborates with him on election forecasting, have launched a Web site detailing the election 2000 forecast. The site,, explains the mathematics behind the forecast and provides links to the external economic and polling sources Lewis-Beck and Tien rely on. Tien is an assistant professor of political science at Hunter College in New York.

Lewis-Beck's current forecast updates the forecast he released in May, before the final figures he uses to make his prediction were available. Using the updated figures changed his earlier prediction by less than a percentage point.

"This is substantively the same forecast -- that is, it still shows a big win for Gore," Lewis-Beck said. "The public on balance thinks the Democrats will give them a better future. We would have to see huge shifts in the economy and in the popularity of the president for the forecast to be affected."

To change the forecast outcome, President Clinton's popularity rating would have to drop by at least 20 percentage-points, and the economy would have to come to a screeching halt, Lewis-Beck said. There is no indication that either will happen before November.

Lewis-Beck's forecast is different from political opinion polling in that it is finalized months before the election. Also, it stands apart from other forecasts because it is the only one that takes into account the "peace and prosperity" outlook, he said.

"The model gets very close to the actual results. In 1996, it was dead right," he said. This year's model has a margin of error of +/- 1.5 percent.

A panel of nine election forecasters from around the country will present their final forecasts Thursday, Aug. 31, at the American Political Science Association Annual Meeting in Washington, D.C. Lewis-Beck is unable to attend the meeting, but Tien will present their forecast.

To arrange an interview with Lewis-Beck, contact Mary Geraghty, University News Services, at (319) 384-0011 or