Thus it was written in the epistle of Stigers, Greg [And],
>
> But has this issue of repeatability been addressed at all, at least for the
> machine counts? What happens when the same decks are ran and the results
> compared with previous runs? I would very much want an explanation of any
> discrepancies in multiple runs, and only at that point do hand counts even
> begin to make sense, preferably after the error-prone decks are divided down
> into smaller units to better isolate the error-prone cards.

Pretty much any method of counting, machine or otherwise, is prone to some
error.  In general this error is small enough that it makes no difference.
In fact, to have any confidence in the results of the election, this error
must be several orders of magnitude less than the difference between the top
two candidates.  I don't know what the machine- or hand-count errors are in
this case, but it seems unlikely to me that they are several orders of
magnitude below 1/6th of 1 tenth of 1 percent.  It is on that basis, not on
the "butterfly ballot" issue or the "premarked ballot" issue that Florida
should either step aside when Dec 18 rolls around or else hold a run-off
election now.  The difference between the votes for the two candidates is, it
seems to me, not statistically significant.

> It is ironic. If in life, I manage to make a mistake paying a bill, that's
> just too bad, more often than not.

Perhaps so, though I have been pleasantly surprised the few times when a
mistake on my part caused a payment to arrive late at how willing to work with
me the companies involved were.

>                                    I certainly cannot think of a time when
> anyone showed me an ambiguous mark I had made on a scantron form or any
> other test and asked for my clarification.

I know that on the teacher evaluations we do here, any ambiguous marks are
reviewed by the person scanning the forms and if it is fairly clear that a
person had intended to choose a given answer, that answer is the one used.

> However, if voters were not careful to vote correctly, they should not
> expect the state to jump through hopes to divine their original intents.

I know that members of this list have been pretty adamant about the
reasonableness of throwing out double-marked ballots, and while I can
understand the lack of patience with someone who mis-marks a ballot, I would
still like to know the real story behind those ballots (something we likely
will never know :-/).  I believe it was NPR this morning which was reporting
that some voters had been handed ballots prepunched for Bush.  Those who
noticed and complained were given new ballots but one wonders if any either
did not notice or did not complain.  Such a ballot would, of course, become a
double-marked ballot upon the voter's voting for anyone else.

I am again reminded of how elusive the truth often is.

Ted
--
Ted Ashton ([log in to unmask]), Info Sys, Southern Adventist University
          ==========================================================
Let us grant that the pursuit of mathematics is a divine madness of the
human spirit, a refuge from the goading urgency of contingent happenings.
                         -- Whitehead, Alfred North
          ==========================================================
         Deep thoughts to be found at http://www.southern.edu/~ashted