Have the raw EXIT polling data ever been released?
It'd be interesting to see how the exit polling looks this time since in 2004 it inexplicably skewed Democratic according to the Punditocracy.
Strangely, they'll gladly tell you how "women over 50 who drive red Fords" voted, but just try to find out if the exit polls matched the official results.
I can't find it.
The Edison-Mitofsky website just talks about how hard they worked. Nothing there of value that I could find.
Exit polling firms understand now that the general public's inability to understand the basic tenets of statistics make releasing that information a really bad idea. If they did, the losing side would always find an irrational reason to cry foul.
You can never know if exit polls are "well sampled". All exit pollsters weight their exit polls to match what they "think" represents the whole, but nobody really knows what that whole looks like. Thus the endless arguing over what the proper weighting should be. Everyone argrees that you have to weight the raw results (except for a few clueless people at DU), but what that weighting should be is extremely difficult to determine.
8. If it's so difficult, how can anyone ever have any confidence that the exit polls say anything?
If it is so difficult to weight the polls accurately, then why did the US insist that the exit polls in Georgia (the country, not the US state) indicated election fraud?
Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators
Important Notices: By participating on this discussion
board, visitors agree to abide by the rules outlined on our Rules
page. Messages posted on the Democratic Underground Discussion Forums are the
opinions of the individuals who post them, and do not necessarily represent
the opinions of Democratic Underground, LLC.