Vote Fraud Theorists Battle Over Plausibility
Study Gets Blog Love, But Comes Short of Proof

By Terry M. Neal
Washingtonpost.com Staff Writer
Sunday, April 24, 2005; 11:27 PM

After my recent column on President Bush's popularity woes, a torrent of e-mail flooded in from angry Democrats insisting that Bush's relative lack of popularity only reinforced their belief that the 2004 election was stolen.

Regular readers are well aware that I'm not a conspiracy theorist. My natural journalistic skepticism applies not just to politicians and people in power, but to wild-eyed theories as well. The Talking Points column that followed the polling piece debunked the idea that House Majority Leader Tom DeLay was the victim of a vast left-wing conspiracy.

Similarly, it strains credulity to think that there was some sort of massive, coordinated effort to steal an election. Such a conspiracy would have had to cross state lines, involve hundreds or thousands of people and trickle down from the heights of power to the lowest precinct worker.

Yet there's lots of chatter in the blogosphere, but little coverage in the mainstream media, of a study that suggests the early exit polls that showed Kerry beating Bush may have been accurate after all. The study, conducted on behalf of U.S. Count Votes, a non-partisan but left-leaning non-profit organization.

The statisticians who performed the analysis for U.S. Count Votes, led by the University of Pennsylvania's Steven F. Freeman and Temple University's Josh Mittledorf, have not been eager to use the word conspiracy. After all, they're scientists. Their job is dispassionate, quantitative analysis. But in some ways they seem to be playing a game, too, because the study clearly leaves the impression that the authors believe there was wholesale fraud in the 2004 presidential election.

The methodology and math of the study are far too complicated to get into in detail here. But here is a link to the entire study for your reading pleasure.

Among other things, the study reports that some of the largest discrepancies between exit polls and final vote tallies occurred inexplicably in battleground states.

"This discrepancy between exit polls and the official election results has triggered a controversy which has yet to be resolved," according the report.

Last year's exit poll was conducted by Edison Media Research and Mitofsky International for ABC, NBC, CBS, CNN, Fox News and the Associated Press, and many reports, including one from Edison and Mitofsky, have found flaws in the poll.

The U.S. Count Votes analysis noted that Edison-Mitofsky's "national poll results projected a Kerry victory by 3.0 percent, whereas the official count had Bush winning by 2.5 percent. Several methods have been used to estimate the probability that the national exit poll results would be as different as they were from the national popular vote by random chance. These estimates range from 1 in 16.5 million to 1 in 1,240. No matter how one calculates it, the discrepancy cannot be attributed to chance."

Even Edison and Mitofsky have acknowledged that the problem between exit polls and the vote count could not be explained simply by statistical error. One of their primary explanations: Kerry voters were more likely than Bush voters to participate in the exit poll.

But the U.S. Vote Counts analysis refutes that point and suggests that Bush voters actually participated at a higher rate than Kerry voters.

"We conclude that the hypothesis that the voters' intent was not accurately recorded or counted cannot be ruled out and needs further investigation," the U.S. Count Votes report suggests.

I called the exit poll coalition to get its take on the U.S. Count Votes analysis, and was called back by none other than Warren Mitofsky, the president of Mitofsky International and a pioneer of the art and science of exit polling.

Until now, Mitofsky has not spoken publicly about the USCV study. He thinks it's bunk.

But it's not a comfortable case for him to make. "I think they're wasting everybody's time, frankly," Mitofsky said. "I know they're very serious about believing that there was fraud, but I don't happen to share their view. I find myself in the awkward position of having to argue that the exit polls were wrong.

"This is not the first election with errors and the simplest explanation is probably the right one. I think fraud on a massive scale that their conclusion essentially requires is totally implausible. To make it plausible it would have a lot of people working together, and you know from being in the news business how hard it is to keep something secret. I just think their whole explanation is implausible."

I called Mitteldorf for an explanation, and he said he only knows the "what," not the "why." The "what" is that the unprecedented discrepancy between exit poll and vote counts cannot be explained merely by statistical error. It is possible though, he said, that there was widespread fraud particularly in key battleground states without a conspiracy.

"It doesn't necessarily take a conspiracy," he said. "It could just be that there was an atmosphere [from Republican leaders] of 'Hey, we really need to win this election, wink, wink. Whatever you do, we'll stand behind you. There will be no investigation because Republicans control the courts and everything, especially in places like Ohio.'

"It could just be that there are thousands of people working independently but with the knowledge that they are being protected and will never be prosecuted for this crime. . . . But I don't know. That's not my area of expertise."

Polling, Mitosfky argues, is not Mitteldorf's area of expertise. He and others have taken the USCV statisticians to task for shoddy work.

"The trouble is they make their case very passionately and not very scholarly," Mitofsky says. "I don't get the impression that any of these people have conducted surveys on a large scale."

Others have questioned the methodology and conclusions of the USCV report as well and not just Republicans. Democratic pollster Mark Blumenthal, who writes the mysterypollster.com blog, accuses USCV of blithely ignoring crucial parts of the Edison-Mitofsky report that attempted to explain the reasons for statistical error particularly the failure of polltakers to follow crucial directions.

"The Edison-Mitofsky report includes overwhelming evidence that the error rates were worse when interviewers were younger, relatively less experienced, less well educated or faced bigger challenges in ing voters at random," Blumenthal writes . "The USCV report makes no mention of the effects of interviewer experience or characteristics and blithely dismisses the other factors as 'irrelevant' because any one considered alone fails to explain all [statistical error]. Collectively, these various interviewer characteristics are an indicator of an underlying factor that we cannot measure: How truly 'random' was the ion of voters at each precinct? It is a bit odd to say the least that the USCV did not consider the possibility that the cumulative impact of these factors might explain more error than any one individually."Ultimately, the USCV report is interesting. But is it anything more than that? Given the statistical complexity of the information, I don't feel qualified to answer that question after a few days of investigation. Scientists and statisticians will continue to debate these issues for months, if not years to come. And certainly some mainstream media journalists will keep digging although investigating something does not guarantee an outcome.

People on the right will cry bias because I even acknowledged this report. People on the left will cry bias because I didn't endorse it. Rather than tell people what to believe, I prefer to let people decide for themselves. And some people on both sides will say I'm the one who's weasely because I don't take a position.

The bigger question to me is what Democrats have to gain from focusing on the past. Anger over the 2000 election didn't help the party in 2002 or 2004. And even though Kerry has claimed that voter intimidation did occur in some places, not even Kerry or his top aides and advisers have glommed onto the USCV report.

Blumenthal, who has worked for Democrats for two decades, said in an interview that it is unfortunate that the debate over an implausible conspiracy might overshadow the real debate over things like voter suppression, intimidation and disenfranchisement. Blumenthal says his disbelief in the mass conspiracy theory doesn't diminish his belief that Republicans have attempted to suppress votes, at times.

"In order to advocate for a change you shouldn't have to buy into [the theory] that five million votes were stolen," Blumenthal said. "I'm not saying that there is no possibility that there wasn't a vote stolen here and there. But do the exit polls present a case for fraud. I don't think so."