tazzygirl
Posts: 37833
Joined: 10/12/2007 Status: offline
|
quote:
ORIGINAL: DarkSteven quote:
ORIGINAL: tazzygirl Ken Bickers from CU-Boulder and Michael Berry from CU-Denver, the two political science professors who devised the prediction model, say that it has correctly forecast every winner of the electoral race since 1980. tazzy, I need to correct you on a fine point which is very important to data crunching guys like myself. The model did NOT predict races since 1980. The model AGREED WITH the outcomes. Had the model been in existence since 1980 and correctly predicted, I would be in awe. But it actually was built to reflect results from previous elections. That's using hindsight, which we all know is flawless. The problems with the CU model are: 1. It uses the economy as the sole independent variable. No other factors. Including anything about the incumbent's opponent. Karl Rove proved that even if you are a lousy incumbent, you can win if you make your opponent look bad enough. 2. The model assumes that Obama is viewed as responsible for the economy. The ownership is viewed as split between him and Bush. He's also trying to tie Romney's Bain activities to the sort of thing that leads to middle class unemployment. So the bad economy will hurt Obama, but not as much as if 100% of the electorate held him responsible. 3. Saying that the model correctly reflects all election results since 1980 implies that there are a lot of results prior to 1980 that won't work. grins.. I just offered the data, DS. I was thinking along the same lines as well.
_____________________________
Telling me to take Midol wont help your butthurt. RIP, my demon-child 5-16-11 Duchess of Dissent 1 Dont judge me because I sin differently than you. If you want it sugar coated, dont ask me what i think! It would violate TOS.
|