The Unexplained Exit Poll Discrepancy--250 million to 1 odds

The Unexplained Exit Poll Discrepancy

A BUZZFLASH NEWS ALERT

BuzzFlash was forwarded a copy of a new research paper (271k PDF) on the exit polls from the 2004 election.

In "The Unexplained Exit Poll Discrepancy," Dr. Steven F. Freeman says:

"As much as we can say in social science 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."

The odds of those exit poll statistical anomalies occurring by chance are 250,000,000 to one. 250 MILLION to ONE.

He concludes the paper with this:

"Systematic fraud or mistabulation is a premature conclusion, but the election's unexplained exit poll discrepancies make it an unavoidable hypothesis, one that is the responsibility of the media, academia, polling agencies, and the public to investigate."

http://www.buzzflash.com/alerts/04/11/ale04090.html

I heard the BushCo whores yapping about the exit polls being "off" on Theft Night that there was trouble brewing. It was just too weird, and too sudden, and too specific as to WHERE it seemed to be happening.

These fuckers must think the whole damn country is as dumb as the bag-of-hammers crowd that supports them.

Sorry, Freeptards...all you double-digit IQers better start counting the days 'til January 20th; that's all Chimpy has left.

I like trhat term for Nove 2

I think his data set is flawed. The exit polls were released before the polls closed, so, he should only be counting votes cast through the reporting of the exit polls.

Second, once Ohio provisionals and other votes are counted, Kerry's total should approach a percentage within the 75%-possible range (pg 6 of the pdf).

I think everyone needs to gather more data before making fools out of themselves; that does no one any good.

Ask him to describe what the data is. Not just "exit polls" or "result, " but actually go in depth. When were the polls taken -- the first three hours? the whole day? the entire election including early votes? What was the methodolgy -- did the pollsters use ten people in a row? did they take every 15th or so voter? etc. Same thing with the results -- do they include provisional ballots, is there a way to control for (rightfully) spoiled ballots, etc. The flaws can easily be explained by those factors, so we should know what we are dealing with.

Thanks. Considering the exit poll data Freeman used was consistent with that of the data used in every other state with results that matched the exit polls, I think his study is sound. However, I'm always open to other interpretations. Please do post your analysis. I'll bookmark this thread and check back frequently. In the meantime, I'm certain Zogby is on this question like green on grass. It is his future as a pollster that's at stake, after all.

Welcome to DU.

We don't know what the data represents, b/c we do not know what methods the pollsters used. That's what I'm asking for. If we take Freeman's paper at face, _he_ is asking for the same thing -- he wants an explanation as to why there is such a discrepency.

One of the starkest examples of my complaint comes from Freeman's paper, footnote 2: the exit poll data changed. Freeman admits that his data set is likely flawed due to his inability to obtain the raw data. (see page 3). There's been no explanation as to why, other than to "correct" it for demographic needs (and even that explanation is more rumor than fact). Perhaps pollsters were observing something that required a correction other than a demographic recentering? That fits nicely with Occam's Razor.

There is plenty of reason to believe that the exit poll data is generally flawed -- Kerry +10 in New Hampshire, +8 in PA, and +9 in MN are almost impossible given the surrounding states. Freeman himself suggests exit poll problems as possible explanation in his third-to-last paragraph, stating that one of the most important investigations is "examination of the exit poll's methodology and findings." He also notes that "if the polling consortium would release their data, that would allow us to do more definitive analyses."

That's the point -- we need more information about the polls. There are two explanations here, and if it turns out that the polls were the problem, the term "conspiracy theorist" will rightly be used as a pejorative.

The exit poll methodology proved accurate in nearly all 50 states, just as it has for numerous past presidential elections. It's the 2004 exceptions that intrigue. Many people don't want us looking at those, though. I wonder why that is? Hmmm. It's a puzzlement. Luckily, the Nader-sponsored recount in NH will shed a great deal of light on the subject.

"I think his data set is flawed. The exit polls were released before the polls closed, so he should only be counting votes cast through the reporting of the exit polls."

Hahahahaha. We can't even get an accurate count for the whole day. I suppose you think it is possible to get a 4pm total?

I actually looked at this study once and think they may have used the wrong data-- the later ones that MIXED IN THE ACTUAL RESULTS. It took me forever to pull up so I am not going to try again. So it could be the correct result would be 500 million to one, or..........whatever.

A very simple way to check to see if the survey is flawed rather than the vote is to compare the presidential exit polls to the Senate exit polls. I am not sure anyone has yet done that in a rigorous way.

Re post 3: he explains where his data came from. If you want to be helpful, why don't you collect all the exit polling data and do an analysis?

Do you have it?

With all the reluctant evidence we have about Bush and Cheney I positively 110% do not believe them or their cheerleaders. The odds are so astronomical that the only way to block them is to throw me in a political mental asylum.

Trying again to document mathematically the entire extent of the crime(is NOTHING ever enough?) is an enterprise that the media has raised the odds of success to 1000000:one against the truth winning.

Bush- WWI profiteer for BOTH sides
Bush- WWII financier to Hitler and the camps, slap on wrist, member of Congress, eugenics fan
Bush I- Iran Contra to name one, but admittedly a lower profile as the first Bush worthy to inherit the WH.
Bush II- loose living, hater of the poor, miserable and crooked businessman, number one fan of Cultic RW loons, cheated in two elections to ruin the US and all its "values".

Oh, yeah. Legit with a large bunch of snide courtiers who would be seen as journalists and an opposition party living in another zone of reality altogether and loving it.