On 4 Dec 2000 14:28:12 GMT, [EMAIL PROTECTED] (Radford Neal)
wrote:
< snip, some detail >
> Let me try one more time:
>
> 1) A certain number of ballots will have been filled out by people
> who HAD NO INTENTION OF VOTING FOR PRESIDENT. In a manual recount,
> a certain number of these will nevertheless be assigned to some
> presidential candidate.
- well, zero is a number. What makes you think it will be noticeably
above zero?
Oh, well, okay, if you are shuffling 100,000 ballots by hand. But if
you are looking through a few hundred doubtful ones as part of a team
doing a check, I think there is good chance of zero error, or close to
it. -- less error, probably, as you would have in feeding 100K cards
through a card-counting machine when that is not your daily
occupation. (You can jam the machine a few times, drop some cards,
mis-lay a pile of cards, ... What's 218 cards for Nassau County, but
the Margin of Machine Error? )
> If they are so assigned in an unbiased
> manner, they should split evenly. Furthermore, there is good reason
> to think that such spurious assignments of votes will be more frequent
> in the manual recount of ballots rejected by machine than in the machine
> count of ballots that were regarded by the machine as unambiguous.
- Bull. What are you going on about?
Okay, the machine count, when humans don't mess up
the bags of ballots (which happens too often) is not
something to fight about.
The ability of humans to determine most of the dented
ballots -- yea / nay / not determinable -- is not something
to fight about either, for the most part, if someone isn't busy
trying to preserve a thin, accidental margin.
By the way, as I understood the judge in Leon County,
I think he has a valid, winning argument, though he might
have botched some details:
The judge insists that there can't be a "two-tier" system,
that *all* votes (in some sense) have to be recounted.
I think he is wrong if he means "every county" or if he
means that every vote has to be hand-examined. BUT he
is right, if he means that fairness demands that if you are
going to search for undercount in punch-ballots, you have
to look at all the punch-ballots that weren't counted, across
the whole state.
- I could imagine that being parsed a little further, so that
any punch-ballot precinct (county?) with less than 0.5%
blank might not need a count, since that is at the limit of the
punch-technology. (That could also serve as a public target
and thus an inducement for improving local success rates.)
Since you ought to use the standard of "the whole state"
(and I think Judge Sauls said that more than once),
the Democrats failed to show that a recount might
overturn the original outcome.
-- It was never in the court case, but a newspaper
report asserted that about 56% of the Gore vote
was by punch-card, and also, Gore won about 56%
of all the punch-votes. So, leading perhaps 560-to-440,
Gore might expect to *gain* about 120 votes for
every 1000 under-count votes *recovered*, state-wide.
-- Would 5000 or so under-counted ballots be recovered,
in a statewide recount of punch-ballot-votes-rejected-by
machine? I would guess, "probably," but the Democrats
never thought to press this argument.
< ... Here is a bit more of Neal's curious fantasizing, with little
apparent connection to known processes ... >
> It would be interesting to know whether the recount is looking at
> ballots selected by some criterion that doesn't relate to whether it
> seemed like it might be a vote for Bush or Gore, rather than for some
> third-party candidate. If so, one could look at the vote counts for
> the third-party candidates in the manual recount. If the proportion
> of such third-part votes has gone up, that might give an indication of
< snip >
Wouldn't somebody have jumped on that scandal with both feet,
if there were any hint of it? Shoot, the Republicans have jumped
with both feet onto scandals where they had to invent 'em first....
Loose chad?
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
