[EM] advocacy by means of exit polls

[RLSuter at aol.com]

<< There are well-known cases where exit polls get the margins wrong
 (e.g. Phillipines 2004) or even the results wrong (e.g. USA 2004).  It
 is extremely hard to correct for sampling error in exit polling,
 making the overall polling accuracy comparable to certain alternatives
 such as phone polls.  For electoral results, both of those are
 significantly more accurate than, for example, random polling in other
 public places.
 
 Michael Poole >>

It is far from certain that 2004 US exit polls got the results of
the presidential election wrong. Many well-qualified polling
experts and statisticians who have studied the polls and the
official election results believe there is a strong likelihood that
the exit poll results were correct and the official results were
incorrect because of widespread fraud or at least unintended
error. None of the efforts to explain the polling results in terms
of sampling and other kinds of errors, including efforts by the
people who designed and conducted the polls, have been
very persuasive. Unfortunately, the kinds of post-election
investigations that could have shed much needed light on
this question were not done or were greatly delayed because
the losing candidate, John Kerry, conceded the election less
than 24 hours after the polls were closed without demanding
that serious investigations be conducted and without doing
any himself, despite having something like $20 million of
unspent campaign contributions on hand (twice the amount
spent on the exit polls themselves) that he could have used
to keep his earlier promise to make sure that every vote
would be counted. Reportedly, even Kerry's running mate,
John Edwards, was upset with Kerry's quick concession,
which also had the effect of discouraging reporters from
seriously investigating the election results and the possibility
of widespread fraud. Kerry's quick concession had the additional
effect of enabling news organizations and pundits to dismiss
people who questioned the election results as irresponsible
"conspiracy theorists." It now appears that the most irresponsible
people of all (except for those who may have engaged in
election fraud) were Kerry and news media people and that
the hypothesis that the election was stolen as a result of fraud
is a very reasonable one that is supported by at least as much
evidence as the theory that the exit polls were incorrect.

-Ralph Suter