Political Scientists Forecast Bush Victory

[onedomino]

Can six of seven "forecasting models" be wrong?

Political Scientists Forecast Bush Victory

http://washingtontimes.com/upi-breaking/20041015-033338-3968r.htm

Washington, DC, Oct. 15 (UPI) -- Six of seven election forecasting models developed by political scientists predict a Bush victory, the American Political Science Association said Friday.
The APSA said nine non-partisan political scientists who developed the models predicted Bush would garner 53.8 percent of the two-party popular vote in the 2004 election when the seven forecasts were averaged together.
The seven forecasts employ a range of empirical and historical data-including economic indicators, public opinion polling, and factors reflecting the advantages of incumbency-to predict with probabilities ranging from 50-97 percent that Bush will win re-election over Democrat nominee Senator John Kerry.
"Each model predicts the share of the national two-party popular vote for the candidates of the major parties," said SUNY Buffalo Professor James E. Campbell, "so their evaluation should be based on their success in predicting the vote."
The forecasts will appear in an election-specific symposium moderated by Campbell -- who predicted a Bush win with 53.8 percent of the popular vote based on the incumbent's trial-heat Labor Day Gallup numbers and the real growth in the GDP in the second quarter of the election year -- in the October issue of the association's journal PS: Political Science and Politics.

Here is the current electoral vote count (click on a state to get polling data): http://www.electoral-vote.com/