On 5 Jan 2001 08:15:56 -0800, [EMAIL PROTECTED] (Simon, Steve, PhD)
wrote:
>Data quality: What caused the two premature calls for Florida in the media
>on election night? Are exit polls a useful source of information?
>
>Data quality again: What is the impact of having different standards for
>counting votes in different counties?
>
Another corollary issue is the media calling Florida for a candidate
based on common exit poll data prior to the closing of the polling
stations due to differing time zones within the state. Needless to
say, some citizens from the western part of the state may not have
bothered voting once the electronic media has consecrated a "winner."
Exit poll methodology should be explained to the public. The public
has the right to be aware of these techniques and pollster statistical
extrapolation prior to the announcement of the victors. Essentially
announcing this is our "best guess" would have been more apropos.
Exit polling as a "science" took a hit when one is confronted with a
raw vote indicating one thing and the media exit poll results
extolling just the opposite. Talk about a rush to judgement(s)! Both
Bush and Gore have a right to wonder about this boondoggle.
Differing balloting standards in Florida was a serious issue. Some
asked why is it a ballot card counted in one county is considered a
non-vote in another. Consequently, where a voter votes is an
additional dimension. Statisticians can get it right if given a
decent platform or methodology. It sort of reminds me of graduate
students coming in with all their data collected and announcing, "now
help me make some statistical sense for my dissertation."
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
