created Kerry's share of the with demo projection and the official count of the vote would be as
much as the final 3.4% version356 a virtual impossibility.357 As a matter of fact, there are broad
statistical variations of up to 9 percentage points between exit poll data and official results in
Ohio and other key states in the 2004 election.358 In state after state, Senator Kerry's advantage
in the exit poll results was lost by sizable margins.
The discrepancy between the exit polls and the official vote count must be due to an
inaccurate poll or an inaccurate vote. Either there was unintentional error in the exit poll or the
official vote count, willful manipulation of the exit poll or the official vote count, or other forms
of fraud, manipulation or irregularities occurred in the electoral process. Pollsters Mitofsky and
Lenski have intimated that their poll numbers deviated from the official results because a
disproportionate number of Bush supporters refused to participate in their polls.359 However,
Professor Freeman posits that part of the discrepancy is due to a miscount of the vote.360
As noted above, election polls are generally accurate and reliable. Pollsters are able to
categorize their sources of error and develop extensive methodologies to limit those errors with
each successive poll.361 Political scientist Ken Warren noted claims, ". . . exit polling has
become very sophisticated and reliable, not only because pollsters have embraced sound survey
research techniques, but because they have learned through experience to make valid critical
adjustment."362 In fact, prominent survey researchers, political scientists and journalists "concur
that exit polls are by far the most reliable" polls.363
Unfortunately, throughout American history various devices, schemes and legal
structures have been used to shape the outcome of an election. Elections at every level of
government have been skewed by tactics that deny voting rights, establish poll taxes, lose voter
registrations, disqualify voters and disqualify ballots to ensure a certain outcome. The Florida
356
Moss v. Bush, No. 04-208873.
357
See Freeman supra at 13.
358
See Freeman supra at 2.
359
See Steven F. Freeman, Hypotheses for Explaining the Exit Poll-Official Count
Discrepancy in the 2004 US Presidential Election (DRAFT), provided by author, Jan. 3, 2004, p.
3.
360
Id.
361
Id at 7.
362
Mr. Mitofsky has worked on almost 3000 elections in his career and he has confirmed
that the 2004 poll was conducted correctly. http://www.mitofskyinternational.com. See Freeman,
Hypotheses at 6.
363
Freeman, Hypotheses at 10.
76