Reply #41: Of course the ignore button is useful [View All]

on occasions, but it does remove your right to make any inference as to whether an ignored poster has, or has not, covered an issue "exhaustively".

I believe OTOH shares my view (and that of Walter Mebane) that the data unambigously confirm that Democratic precincts in Franklin county were systematically allocated fewer voting machines per active voter than Republican precincts. The only area of disagreement, I believe, is the extent to which this can be inferred to have been deliberate on the part of who.

As for exit polls, you are of course entitled to respect the opinions of Baiman and Freeman. OTOH and I are also entitled to point out, from positions, it has to be said, of considerable expertise, that they have made fundamental statistical errors in their analyses. Ditto with TIA.

If you have OTOH on ignore, you will have to be content with my critiques only. And if you have both of us on ignore - well, I post this for the sake of other posters who may benefit from the critique:

• Baiman's conclusion that the exit polls for Ohio provide "virtually irrefutable" evidence of miscounts is faulty on so many levels, it is hard to know where to start, but the bottom line is that the only slightly valid finding is not statistically significant, and the statistically significant findings are based on faulty assumptions regarding the data.
  
  
• Freeman's conclusions that the state-level exit poll data indicate widespread fraud are based on a misunderstanding of the nature of the data, a misunderstanding of the error variance in the data, on at least two simple mathematical fallacies, and cannot account for the major finding that redshift was not correlated with benefit to Bush, a finding that, ironically could only be theoretically reconciled with massive fraud at the cost the evidence he finds to support it.

Mod Mom - I respect your on-the-ground knowledge of the abuses in Ohio. I would like to see them centre-stage. I resent that they are upstaged by the hyperbolic claims based on fallacious inferences from the exit polls. There is no evidence of multi-million vote digital theft in 2004.

If the election was stolen, it was not stolen by methods that resulted in the exit poll discrepancy.*

*edited to add: which pretty well rules out widespread, massive, digital fraud