[statistical content included]
The design of butterfly ballot still seems to be generating the most
press. What's more interesting to me is the request for recount by
hand. A few years ago here in MA, we had a hotly contested congressional
race involving punch cards. William Delahunt lost in the initial tally.
He went to court and argued that voters were disinfranchised because
many of the punches did not go through and therefore were not registered
by the machines. *The judges* examined each questionable card
individually and depending on how much of an impression was made decided
whether a vote would count (irrespective of who the vote might go to, of
course). The election was reversed, Delahunt was awarded the seat, and
punch cards are now illegal in MA.
Statistical content: The hand recounts will be taking place in areas
where Gore already won handily. Unless there's some factor causing
those whose punches failed to go through to favor Bush, it is likely
that newly admitted votes will heavily favor Gore. No wonder there are
reports that the group that is unlikely to benefit from a hand recount
would try to block it.
Comment on the FL vote threads: The statistical insights are valuable.
I'm surprised by the level of the off-topic, heated political debate
that same have tried to introduced. There's a process to be followed.
It's being followed. The system explicity provides for judicial input.
It seems inappropriate to criticize either side for using the process to
seek a resolution that is valid under the law. If one side wants
something, let them ask. If another side wants to block something, let
them ask.
Practicing statisticians deal with this all the time (can I throw out
this observeration? can I collect a few more? what data cleaning
procedures should we use? can I now remeasure this subject whose
original value is questionable?). All we're doing in Florida is moving
toward a finalized, closed data set by using the established
procedures.It's a very important data set, but just another data set.
It was a complicated protocol. Sometimes we've forgotten what the
protocol specified (rules for ballot design, procedures to deal with
voters who acknowledge an error at or after leaving the polls).
Sometimes we're not sure how the protocol should be interpreted (so we
ask the judge). But, if we treat the votes like any other dataset and go
about our business, it'll all get taken care of.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
