'Exit polls are the one tool we have when elections are in doubt.'

Steven FreemanPhoto credit: Candace diCarlo


On Election Night 2004, Steve Freeman found himself looking back and forth between the CNN web site and the TV news with disbelief. Why were the exit polls that had predicted a Kerry victory so wrong? The next morning, as he examined the exit poll data more closely, he was again astounded—this time by the fact that the data posted on the CNN site was completely different from what he had seen the night before.

He waited for news organizations to jump on the story. And waited. A few days later, he turned to the Internet for information, and found much speculation that the election was not exactly on the up-and-up.

“I never, ever, considered just the fact that there was this discrepancy as evidence that in fact, Kerry won,” says Freeman, a lecturer in Penn’s Center for Organizational Dynamics. “What we do is we look at the data and we say there are certain implications.”

In his new book, “Was the 2004 Presidential Election Stolen?” written with In These Times magazine editor Joel Bleifuss, Freeman examines cases of voter fraud and disenfranchisement and attempts to explain why the exit polls differed so much from the voter tallies. “Exit polls are the one tool we have when elections are in doubt, one large quantitative measure that you can use to ascertain voter intent,” Freeman says.

In anticipation of the Nov. 7 midterm election, Freeman is working with Ken Warren of St. Louis University to conduct a regional exit poll in precincts that may be vulnerable to fraud. Their Election Integrity web site (www.electionintegrity.org) contains instructions on how precinct captains and others can conduct their very own exit polls. “They may not be able to do it at a level that’s going to allow them to publish the results in a scientific journal,” says Freeman, “but they’ll get some sense of the veracity of the count.”

Q. Have you been dismissed as a ‘“wacko liberal” or “sore loser?”
A. I’m not very partisan. I’m really not. I get accused of being just a wacko liberal. I’ve been dismissed by one academic as some crazy [theorist] from the left.

Q. But others had put forth the idea of the election being stolen before you.
A. There was one article in particular by Thom Hartmann, who basically put forth the thesis that [the election] was stolen. It held together—not to say I believed it—but it seemed plausible. On the other side, I couldn’t find anything. I did as thorough a search as I could conceivably for explanations about the exit polls being off and there was so much nonsense. Electronic voting I did not know about. It’s astounding to any computer scientist that this could be the system used. I’ll come out and say it—there’s only one thing that the current voting system is good for and that’s to facilitate stealing elections.

Q. What makes exit polls so accurate?
A. They have tremendous advantages over the typical telephone poll. I liken the difference to trying to predict a storm several days in advance to just putting your beakers out and measuring the rainfall. You don’t have to decide who’s going to show up, you don’t have to worry about reaching the universe of voters, you don’t have to worry about people changing their mind, they’re all there.

Q. Many of the exit pollsters said the polls must be flawed.
A. Warren Mitofsky and Joe Lenski [National Election Poll officials] have not come forward with the data. But the truth of it is, if they are perceived as remotely partisan, they would never get another contract. I’m doing an independent exit poll for this election. I didn’t realize until I put the budget together for this, how profitable a contract they had. They had a $10 million dollar contract.

Q. If exit pollsters themselves say these polls aren’t meant to back up election results, what’s the point of them?
A. The point of them here, they say, is to allow the media to estimate who’s going to win various elections so they can accordingly plan their coverage and to provide them with data so as to analyze why and how voters voted the way they did. The media never mention the discrepancy. They all say the exit polls were no good, but yet they were all reporting how the exit polls showed that Americans voted the way they did because of moral values. That was so badly designed, that part of the questionnaire. It’s changed, by the way, for 2006.

Q. In what way?
A. They’re not going to give any data to the media clients until it is corrected. And that is to prevent the bloggers that allowed us to know about the exit poll discrepancy in 2004. By the way, whenever they say, “early,” they don’t mean early in the day like everyone imagines them to mean. Early means before they correct them.

Q. What do you mean by “correct?”
A. They change the numbers so that they match up with the official count. The early ones are the ones that are the direct polling results, what everybody imagines exit polls ought to be. Of course, everyone in the media misinterpreted it because it’s so misleading. Early in the day Kerry was ahead, but then the Bush voters must have come out late in the day. That is not correct, it’s one of many misleading terms used by the pollsters. One of the big improvements for the 2006 exit polls is that no data will be released ... until it is corrected.

Q. What do you expect in these midterm elections?
A. I expect a Republican victory no matter how unpopular they are. Never in a million years could I have imagined we would come up with overwhelming evidence, just mountains of evidence of a stolen presidential election and people say, well that’s not important, let’s look forward, that’s the important thing. But the fact is ... the perpetrators have been emboldened.

Q. Do you have faith in electronic voting?
A. Electronic voting has spread. I believe 90 percent of Americans in the upcoming election will vote on either direct record or optical electronic scan, both of which are highly vulnerable to fraud. Optical scans are better because there is a paper trail, but that paper trail only means something if it’s used and in fact in many states, there’s been increased hurdles to actually doing a manual count.

Q. What about the Democratic party’s response to this?
A. The Democratic response makes me sick inside. There’s this vicious circle of refusal to address the issue. It’s not like we’re trying to be partisan, we’re just trying to help them actually defend the right to vote. It’s not the Democrats we care about, but their constituencies. If people want to vote for them, against the other guys, they have a right to have those votes counted. The answer is not to not vote. Even though your vote may not be counted as cast and even though you may not be able to do anything about it, you can show up on Election Day and at least kind of affirm the importance of voting. One battle we’ve won on this is that there is almost a unanimous affirmation that we ought to have the right to vote and have that vote count. The truth of it is that voting has never been enough. It’s very naïve to think that a regime will allow ballots to really have the power to change. People have to do a little more than that—and always have had to by some degree of organization and awareness.

Q. Some say that the exit polls oversampled Kerry voters and had what you call “Reluctant Bush Respondents.” What do you say to that?
A. Basically, everything the exit pollsters say obfuscates the issue. They use misleading terms. In terms of election integrity here, there’s only two possibilities:The actual data was even more extreme than the election night data would have led us to believe. Nationwide, there was a seven percentage point disparity between how people said they voted that translates to nine million votes nationwide and this is incontrovertible. There’s only two possible explanations for this: Kerry voters were participating in the polls at far higher numbers than Bush voters. You have to underline the “far” there. Or, the count is off. So, what the pollsters say is well, what must have happened is Bush voters participated at far lower rates than Kerry voters because they never seriously consider fraud.

Originally published on November 2, 2006.

Originally published on November 2, 2006