( PDF | ASCII text formats )
The following is mirrored from its source at:http://onlinejournal.com/evoting/111704DeHart/111704dehart.html.
Dr. Stephen Freeman is a University of Pennsylvania professor whose expertise includes research methodology. In a recent paper titled The Unexplained Exit Poll Discrepancy he reports that international foundations sponsored an exit poll in the former Soviet Republic of Georgia during their November 2003 parliamentary election and projected a victory for the main opposition party.  Exit poll data is considered so robust that when the sitting government counted the votes and announced that its own slate of candidates had won, supporters of the opposition stormed the Parliament, and the president, Eduard A. Shevardnadze, resigned his office under pressure from the United States and Russia. 
Contrast that event with what happened in the United States in the recent national election when in three battle ground states, Ohio, Pennsylvania and Florida, with data based on exit polls predicted an outcome in variance with the tallied vote outcomes. Major news organizations, including CNN changed their exit poll data to conform with the tallied outcomes, most of which came from paperless, electronic voting equipment. In each case the tallied outcomes favored the incumbent, George W. Bush. The odds for such an occurrence is one in 250 million for this to have occurred by chance. 
Does the phrase, "Something is rotten in Denmark" have any meaning for the media? In plain language the term refers to a line from the play Hamlet, when an officer of the palace guard, who after the ghost of the assassinated king appears, utters the immortal line, "Something is rotten in the state of Denmark." The term has universal meaning to describe corruption or a situation in which something is wrong. 
Professor Freeman's analyses of the data are compelling for a number of reasons. First, he was able to sample 2004 exit poll data that was not meant to be released directly to the public and was available through a computer glitch that allowed him to view "uncalibrated data that had not yet been corrected to conform to the announced counted vote tallies. These data remained on the CNN website until approximately 1:30 a.m. election night. At that time CNN substituted data `corrected' to conform to reported tallies." [1, p. 3]. Second, uncorrected exit poll data have been secreted in a black box and AP, Edison Media Research, Mitofsky International and the New York Times have ignored all requests for the raw data. In an open democratic system or any scientific inquiry the data would be open to inspection. The fact that it is not adds to the suspicion that widespread fraud occurred in vote tallies in the battleground states.
The integrity of the system is being questioned by citizens across the nation and internationally. The response of mainline media is a harsh attack on citizens and writers who dare raise questions about the data. Robert Parry  points out that The New York Times (NYT) has joined the Washington Post and other major news outlets in scouring the Internet to find and discredit Americans who have expressed suspicions that Bush's victory might not be entirely legitimate.
What the Freeman Data and Analysis Reveal
In the three battleground states, Ohio, Pennsylvania and Florida, exit polls differed significantly from the recorded vote tallies with Bush winning and thereby ascending to an electoral victory. Let us examine the exit poll predictions versus tallied votes in each of these battleground states combining the male and female vote, weighted for their percentage in the electorate by state. For example, the Ohio electorate data comprised 47 percent males and 53 percent females. This procedure was also followed in Florida and Pennsylvania [1, p. 4 & 5].
According to Professor Freeman,
"the likelihood of any two of these statistical anomalies occurring together is on the order of one-in-a-million. The odds against all three occurring together are 250 million to one. As much as we can say in social sciences that something is impossible, it is impossible that the discrepancies between predicted and actual vote counts in the three critical battleground states of the 2004 election could have been due to chance or random error." 
Given these discrepancies in the data and the probability that these events did not occur by chance, in order to document integrity of the process, it is crucial that the NYT, CNN and other media sources open their books for public inspection rather than provide questionable explanations about the discrepancies between exit poll and tallied vote data. While the NYT cites a report issued by pollsters that debunked the possibility that their exit polls are correct and the vote count wrong, they provide no data to support an error in exit polling data.
Multiple explanations provided about error in exit polling procedure crumble under careful scrutiny. For example, the predictions in the Utah presidential election were remarkably accurate. Exit polls predicted Bush would take 70.8 percent and Kerry 26.5 percent of the vote. The actual tallies recorded that Bush received 71.1 percent and Kerry 26.4 percent of the vote.
This was not the case in 11 key states (Colorado, Florida, Iowa, Michigan, Minnesota, Nevada, New Hampshire, New Mexico, Ohio, Pennsylvania and Wisconsin). In each of these states Bush's tallies were greater than expected, and in all but Wisconsin, Kerry's tallies were less than expected from exit polling. (See Professor Freeman's paper for tabulated data comparisons.)
Princeton economics professor Paul Krugman warned that there are many perils in electronic voting. He posits a scenario in which on election night the early returns suggest trouble for the incumbent. Then, mysteriously, the vote count stops and when it resumes, the incumbent pulls ahead.  What Krugman reported is not a paranoid fantasy. It is a true account of a recent election in Riverside County, California, reported by Andrew Gumbel of the British newspaper, The Independent. 
Analyses of available data by independent pollsters show some alarming trends from both Florida and Ohio. In Florida certain counties tallied votes for Bush that were far in excess of what one would expect based on Republican registrations. These were primarily counties that used optical scanning equipment to feed votes into precinct computers that were then sent to countywide databases. At any point after physical ballots became databases, the system is vulnerable to external hacking. Colin Shea reran preliminary CNN data and points out a number of disturbing trends that include counties where 88 percent of the voters are registered Democrats with Bush receiving nearly two-thirds of the vote. Other disturbing data reveal that "according to official statistics for Cuyahoga County [Ohio] they had a turnout well above the national average. In fact, their turnout was well over 100 percent of registered voters." 
Was November 2, 2004, the final act for what began in Florida in 2000, tested in various locales with electronic voting equipment in 2002 and finally played out in a disastrous final act that left the media simpering that this election was about moral issues? That may be true, but they depict the wrong moral issue. The untouchable topic is that election fraud rather than gay marriage turned this election on its ear. The media and politicians would be wise to listen to the voices of dissent and concern.
Professor Freeman concludes his paper with the following statement:
"Given that neither the pollsters nor their media clients have provided a solid explanation to the public, suspicion of fraud or among the less accusatory, `mistabulation' is running rampant and unchecked. That so many people suspect misplay undermines not only the legitimacy of the President, but faith in the foundations of democracy." 
Neither the people nor corporate media should accept the fact that networks altered exit poll results to fit the tallied vote numbers. This calls into question the integrity of other information these networks report. Or as Andrew Gumbel so aptly states, "As the world's most powerful democracy talks of exporting freedom to Iraq, it is at risk of becoming an object of international ridicule." 
For historical perspective, let us review what happened in the former Soviet Republic of Georgia when their November 2003 election results contrasted sharply with exit polls. Both the United States and Russia pressured the president, Eduard A. Shevardnadze, to resign. Compare that behavior to what has just happened in the United States. CNN changed its exit poll data to conform to counted vote numbers under the very eyes of Professor Freeman and other observers. Meanwhile the media do everything in their power to undermine the credibility of independent observers. Those who sound the alarm of voter fraud are summarily dismissed as conspiracy theorists and traitors of democracy.
Sara S. DeHart, MSN, PhD, Associate Professor Emeritus, University of Minnesota. Dr. DeHart is a freelance writer and democracy activist, living in the Seattle, Washington area. She may be contacted at firstname.lastname@example.org