> If approximately 2/3 of the tallied votes were for Gore, by what leap
of
> logic do you conclude that 1/2 of the untallied votes would have been
> for Bush?
The GOP lawyers in their cross of the voting expert argued that people
may have dimpled their chads by handling the ballots, e.g., running
their fingernails across the surface while waiting in line to vote.
Such random handling might be expected to have a 50:50 Bush:Gore ratio
(but who knows). Hengartner's analysis indicates that a higher
percentage of these hand-counted undervotes were for Bush (34:66) than
even a random draw from the overall 31:69 ratio of the machine-counted
ballots. This is inconsistent with the random handling hypothesis.
Although the Yale statistician Hengartner didn't note this, There may
be some indication of too many Bush votes in these hand counts. A
two-tailed test of the hypothesis might have been significant in favor
of the alternative of too many Bush votes. Perhaps having the Governor
of Montant, Marc Racicot intently inspecting the choices and then
declaring to every Sunday talk show and to the press on dozens of
occasions that votes were being manufactured by the board led to the
election board selecting too many Bush votes relative to Gore votes.
Sent via Deja.com http://www.deja.com/
Before you buy.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
