I would second Ray's comments in the following way. When you are
dealing with a vote share of less than 1% of the total voters, it's
pretty clear that a variety of factors not taken into account in any
given model can swing the result towards accepting or rejecting any
given hypothesis. Thus, it's going to be difficult to make the data
speak on this one no matter how hard we torture it.
It is also going to be difficult to get other unbiased sources of
data. It's amusing, for example, that the plaintiffs in the Palm Beach
Case are not doddering seniors as the media implies but in actual fact
are Democratic politico's. For example, there was a profile of Andre
Fladell, one of the three plaintiffs, in the New Times Broward-Palm
Beach in December of *1999* which had this to say about him:
"Next to volleyball Fladell's favorite sport is politics, and he plays
it the same way he has since leading antiwar protests in the late 1960s
- with a ruthless drive to win. When he's not realigning patients'
spines and stomping volleyball opponents, *he's advising candidates and
elected officials in Palm Beach County, mostly Democrats, on how to
crush their election foes and steer government decisions their way.*"
Does this sound like someone who got confused over a ballot!!! I
think not.
Of course there is then the issue of what to do if we did in fact
conclude that Buchanan's vote share was unusually large - Hmmm, if the
data indicates this and we are willing to have a new election based on
what we expect to happen then what is the point of the election!
Moreover, if there were a second election then we obviously have no way
of knowing that people would vote the way they truly wanted to vote
before the results of the election were known.
Returning to a more Armchairish point note that there has been an
awful lot of discussion about the "will of the people." But we know
from Arrow's theorem that the very notion of "the" will of the people is
incoherent. The truth is that what we care about, in this case
especially, is not that the will of the people be done but that
something be done to bring an end to the uncertainty. My view is that
we should toss a coin. (Yes, I am serious. An interesting question is
whether there are probabilities for a coin toss that both sides would
agree to - Robin any thoughts?)
Alex
--
Dr. Alexander Tabarrok
Vice President and Director of Research
The Independent Institute
100 Swan Way
Oakland, CA, 94621-1428
Tel. 510-632-1366, FAX: 510-568-6040
Email: [EMAIL PROTECTED]
