The European Tribune is a forum for thoughtful dialogue of European and international issues. You are invited to post comments and your own articles.
Please REGISTER to post.
National Election Archive Project - Home
Our statisticians analyzed Edison/Mitofsky's own explanation of their exit poll discrepancies, and found serious flaws in their argument. Exit polls have been used for years to detect corruption of official vote tallies - most recently in Ukraine.
If you do not want to read through the back and forth, Edison/Mitofsky's explanation is pretty much the shy republican voter (who does not want to admit their Bush preference). National Election Archive Project analysis shatters that assumption.
Sweden's finest (and perhaps only) collaborative, leftist e-newspaper Synapze.se
The position taken by the Edison/Mitofsky group is consistent with professional norms and practices. Election survey analysts ordinarily assume that official election results are the objective standard against which their own findings must be weighed, and perhaps found wanting. We admire Edison/Mitofsky's willingness to find fault with their methods and interview results. However, nothing in their report demonstrates that such errors could account for the gap between the exit polls and the election results.
Furthermore, I read into NEAP's actual arguments against the Edison/Mitofsky hypothesis for the exit poll - election result discrepancy. The hypothesis is, basically, that 56% of Kerry voters but only 50% of Bush voters were willing to respond to exit pollsters. There is no data to test this hypothesis directly. What NEAP did was to check the overall (e.g. Bush+Kerry+Other) rate of exit poll response as a function of the actual vote for Bush. The result was this graph:
However, note: (1) NEAP commits the sin of not conducting a trend significance analysis themselves, (2) something that should have been conducted by taking the different number of precints in the different bins into account, (3) there are a lot of possibilities for unaccounted-for systematic errors (a systematic error is a non-random bias, say correlation with race or settlement type), that is, factors that make people in general and Republicans especially more paranoid of pollsters in less red or blue areas.
One whose delusions are out of fashion.
by Cat - Mar 31 1 comment
by Frank Schnittger - Mar 22 3 comments
by Oui - Mar 27 23 comments
by gmoke - Mar 17
by Oui - Mar 16 22 comments
by Oui - Mar 15 5 comments
by Frank Schnittger - Mar 9 3 comments
by Frank Schnittger - Mar 14 14 comments
by Cat - Mar 311 comment
by Oui - Mar 31
by Oui - Mar 294 comments
by Oui - Mar 274 comments
by Oui - Mar 2723 comments
by Oui - Mar 22
by Oui - Mar 2211 comments
by Frank Schnittger - Mar 223 comments
by Oui - Mar 1988 comments
by Oui - Mar 1744 comments
by gmoke - Mar 17
by Oui - Mar 1622 comments
by Oui - Mar 1541 comments
by Oui - Mar 155 comments
by Frank Schnittger - Mar 1414 comments
by Oui - Mar 134 comments
by Oui - Mar 128 comments
by Oui - Mar 1112 comments
by Oui - Mar 1060 comments