Steve, if you're trying to make a case for electoral fraud, what does exit polling have to do with it? You're saying that the discrepancy between Mitofsky's data and the actual vote tallies is prime facie evidence of some sort of fraud, but that's horseshit. People aren't obligated to tell poll-takers the truth or even to talk to them at all. The only numbers that count are the ones generated at the ballot boxes themselves.

By your logic, the findings of private polling firms are more legitimate than the results arrived at by state and local governments --- not all of which, of course, are run by those evil Rethuglikkkans. Is that even supposed to make sense?

Polls can never really be accurate. If they were, there'd be no need to actually have a vote: we'd just elect our representatives based on what John Zogby says.

I see this obsession of yours with the polling data and vote totals as a natural consequence of the Democratic Party's insistence on using statistical sampling in the Federal Census. You know: to compensate for missing all of those homeless people and [undocumented workers]. More people = more money. Or, in this case, more votes for the right kind of candidates.

If they were going to commit electoral fraud, why wouldn't they get a bright-n-early start? Why would they do exit polls to find out they were losing, and THEN commit electoral fraud?

Oh, wait. They're idiots.

don't get tricked into focusing on somebody's exit poll shenanigans. The real frauds, i.e. disenfranchisement, purposely under equipped and undermanned voting stations in heavy Dem districts, curious touch-screen antics that only seem to harm the Dem candidate, are right out in plain sight for anyone who cares. But the GOP Media won't touch it and us citizens sit idly by and whimper (unlike the Uki's)Notice what happened in Washington state when a Repug lost a close vote......it made all the national media outlets for days in a row.

Toby

Oh my oh my me thinks you doth protest too much!!!!\\A bit of an overreaction to stats which are always extremely important on the great hunt for { fooling the people twice" or is it once? What is your problem?

Rita, I'll be happy to respond to your comments as soon as they wake up and get some coffee in them.

Maybe I am just slow….. but I still do not understand how everyone can protest and complain about the Ukrainian elections because the EXIT POLLS did not match the actual results….. YET here in the “Land of the Free and Home of the Brave” where democracy has survived for such a long time…..Exit Polls mean nothing when they do not conform to the announced official results.

Does it mean the USA is such a “up right and truthful nation” that there is absolutely no way in the world that election fraud could ever occur here?
Or does it simply mean everyone in these United States is willing to turn over and play dead simply cause the “officials” running the election claim that candidate W won?

When the officials running the election in Ukraine said a candidate won …. It was the protests that caused the re-vote to occur …. Not simply someone meekly asking that a recount be done …. Knowing full well that a majority of these “Touch-Screens” have no way to verify (with Paper trail) that a vote recorded for a candidate was in fact really a vote for the candidate.

Does it mean the USA is such a “up right and truthful nation” that there is absolutely no way in the world that election fraud could ever occur here?

Well, the thing is, in the Ukraine there was a pretty clear indication up front that the [fake] winner was going to lose. When he didn't lose, it looked pretty suspicious.

Almost every pollster showed Bush with a tight lead going into the election, so it's tough to justify massive street protests. (However disappointed I may be with America for being chock full of paint-eating 'tards.)

I do not understand how the readers of this board put so much credence in the 2004 election exit polls. The contempt you all dump on Bush and his supporters leaves me with the conclusion that it is highly credible that Bush supporters were much more likely to decline to publicly express who they voted for ("none of your business, frankly"), than the self-righteous Kerry supporters (with their superiority complexes).

[I]t is highly credible that Bush supporters were much more likely to decline to publicly express who they voted for ("none of your business, frankly")

The phenomenon you're describing is called "shame."

I believe Skeptic just further proves my point.

There is such a belief in the USA that the all powerful
Officials in charge (read all levels of Government)
Are such up right and honest citizens and public servants --
There is no way in the world that the elections were anything
But fair and honest.
(just like that news organization that is fair and balanced)

The fact is the exit polls were extremely out of sync with the actual results.
Yet, rather than being upset and demanding an audit of the election results, the majority of the people wish the whole thing will just be forgotten…..
For people to get over it…

Maybe Bush won Fair and Square…….
But what is the harm with doing an actual Audit?

Ppl, please don't forget that the exit polls were within the usual range (1-2%) of accuracy in ALL STATES but the swing states using optical scanning devices. The odds of that happening are astronomical.

There are several problems with both exit polls and the post.

First, there is a complete disconnect between Mitofsky's post factum data fitting and what may or may not have happened on election day. Mitofsky's "work" was more marketing than historical record--he wants to appear more accurate than his results actually were so that the parties will consider hiring him to do polling in the future. Check that--judging from the skewed and repeatedly changed numbers, he's only marketing to one party, and it ain't the Democrats.

Second, traditionally, exit polls skew Democrat by 2-3%. That does not mean that every single exit poll should be adjusted by 6% in favor of Republicans, but, on average, we can expect this kind of scenario. This, of course, only works for straight polling, not for the weighted variety. Once you start weighing poll results differently, the balance of the skew goes out the window.

What is interesting in the sets of polls from the last elections is that the oversampling seems to be coming from... women. That is, the proportion of women in the exit polls seems to be greater than their proportion in the active electorate (those who voted). This may, in part, explain Mikofsky's finagling of the numbers in the South--if women's share of the vote (and they tended to vote more Democratic than their male counterparts in all regions) was overrepresented in the exit poll numbers, it would make sense to weight the male share heavier in computing the final weighted averages. This could easily account for the 54-58 point difference between the two sets of numbers AND preserve the vote breakdown by gender (but not the gender balance, obviously).

Now, we all now that Toby is a troll. I am actually surprized at the relative mildness of his comment today. However, he's dead wrong. As paradoxical as this might seem, appropriately collected exit poll data may well be more accurate than actual vote count in representing the wished of the electorate. Remember, exit polls represent the way the people think they actually voted (assuming they are honest in their reporting), ignoring misvotes, wrong votes, double votes, computer errors, etc. It is quite clear that the 2000 Florida exit polls were far more consistently accurate in determining the voters' intent than were the actual ballots (this has nothing to do with the issue of whether we should have counted them differently or not). We cannot hope to ever count every vote as it was intended to be cast--the combination of human and technology errors is insurmountable. But, as long as such errors are consistent, we should accept this inaccuracy and allow for the process to go forward. What I see from the last three election cycles is that the number of errors is going up rather than down and the errors tend to skew to one side. The exit polls, again, assuming they are collected properly, being more consistent, point out a problem with the current voting system. They do not, of course, offer a solution--we are not going to elect people by polling.

Call your senators to ask them to join Congressman John Conyers in challenging the Ohio electors:

1-877-762-8762

For the life of me, I cannot see the harm in investigating the Ohio election - Whether there has been fraud, or merely election errors that can be fixed, an investigation will help us determine.
I question the motives of those who would deny us such an investigation.

I don't know if I'd call Toby a troll, he does, sometimes, actually make sense. However, that doesn't make him right.

The point of the exit polling data questions are as an indicator that something is amiss and needs looking in to. They're not proof, but circumstantial evidence that may or may not make a case, and that is all they are. We need to get under the hood of all these machines and software and see what they are actually doing, and not taking the word of their providers.

At the least, why did people have to wait in line for hours in order to vote? That needs to be corrected, even though we know the real reason why, and the correction starts with the impeachment of Blackwell for dereliction of the duties of his office while serving another master.

I didn't study the pollster stuff, but when did weighting the data to get it to conform to the desired result become an acceptable polling tactic?

We all know about statistics, but really!

The phenomenon you're describing is called "shame."

I think this is actually the explanation for the bad poll data. The effect is seen on controversial ballot measures, such as those which are anti-gay or anti-immigrant, where polls don't show the support they end up receiving once people are in the privacy of the ballot box. Basically, Bush voters felt persecuted for favoring an incompetent lying candidate. Go figure!

...that it is highly credible that Bush supporters were much more likely to decline to publicly express who they voted for ("none of your business, frankly"),.... such manners from the morally righteous crowd huh.

Gee wonder why those Bush supporters were likely to decline,it is highly credible those Bush supporters were also pretty embarrassed. PLus your dubious "highly credible argument" also forgets that many voters in that same poll also did so because of Moral Issues....therefore since more of those voters were Kerry supporters it is also highly credible that they were likely to be Moral Value Voters....which frankly wouldn't surprise me(unless that is you equate evangelicals as being moral values people)

As you can see I can't decide if your comment is ridiculous or ludicrous. And though I know my comment is not much better, it is still very possible to make based upon the unproven illogic of your own Skeptic.

Steve, this poll analysis stuff is way over my head, but even I can see something certainly is fishy. I look forward to your future posts on this issue.

I still do not understand how everyone can protest and complain about the Ukrainian elections because the EXIT POLLS did not match the actual results…. YET here in the “Land of the Free and Home of the Brave” where democracy has survived for such a long time…..Exit Polls mean nothing when they do not conform to the announced official results.
I felt this way 2 years before the war in Iraq and before the election. Diebold told Bush that he would deliver this election. I was amazed how there was a play dead attitude. Kerry gave up with no recount. How sad. Ukraine stood in the street, Americans went to bed and say oh well. Let see “you fire” tonight. More important! Americans are so uncaring about what is happening in their government. The lobbyist, who are all old politicians, are writing every laws that have been put for at least for the last 30 years.
And most did not even knew what was happening! Let see how his legacy will go? like Nixon.

I scrolled through the comments before posting pretty much what not stupid said so I'll limit myself to the short version: all the media told us that the Ukrainian vote was fraudulent because it didn't agree with the exit polls while simultaneously telling us that our exit polls were wrong because they didn't agree with the vote tallies. People who believe both of these propositions deserve what they get, but what about the rest of us?

Why is it people are raising a huge stink about a difference between exit polls and voting results, but those same people don't seem to care when there are more votes than voters in Washinton states governors race? I mean, sure, the exit poles may point to something fishy, but more votes than voters? Nope, nothing fishy there, why can't the Republicans just certify the vote for the Dem Governor. Whatever. After growing up in Santa Cruz, California (the wanna be Berkley) all I can say is that the new left is full of it. They only see what they want to see.

Has anybody considered that more potential GOP 'exit-pollers' were likely in a hurry to get back to work, whilst the DFL-ers were happy to milk the government-mandated time away from work to hang out at the polls a bit longer?

I, for one, had work to attend to.

VOTE DEMOCRAT
(it's easier than working)

Eric: First your stupidity is shown by the fact that the DFL is limited to Minnesota. So calling Dems. from elsewhere DFLers is pure ignorance.
Second, you can't assume other get mandated time off to vote. Next time find out how other people live before you show your ignorance.

rlprather:

Stupidity is a poor term, ignorance better, but simple mistake better yet. First draft final draft is the rule here I guess.

As far as finding out how other people live, I'm confused. Time off is mandated generally in all states. Paid time off is mandated in many states. Many employers will give time off, paid, even where not mandated. Early exit polls certainly are more likely to show those who took time off from work, relative to those who voted later in the evening. The point is responsibility and accountability, and the entitlement mentality that has grown beyond belief, to which the bumper sticker lament is directed.

Also, while I do think there is a definite trend that would support my thoughts on timing, understand that my comment was meant to draw a smile as much as anything. Most of the comments from both sides on these issues are so bent one way or the other (i.e., calling those who voted for Bush stupid) that discussion becomes impossible.

The thing is, I have and will again vote Democrat where appropriate. I just lump this exit poll fixation into the group I spoke of above in which terms like "stupid" apply.

I guess I'd take this more lightly but it seems your comment was very serious. We need to stop listening to the Michael Moores (or Rush Limbaughs)and get to the real issues.

rlprather:

Stupidity is a poor term, ignorance better, but simple mistake better yet. First draft final draft is the rule here I guess.

As far as finding out how other people live, I'm confused. Time off is mandated generally in all states. Paid time off is mandated in many states. Many employers will give time off, paid, even where not mandated. Early exit polls certainly are more likely to show those who took time off from work, relative to those who voted later in the evening. The point is responsibility and accountability, and the entitlement mentality that has grown beyond belief, to which the bumper sticker lament is directed.

Also, while I do think there is a definite trend that would support my thoughts on timing, understand that my comment was meant to draw a smile as much as anything. Most of the comments from both sides on these issues are so bent one way or the other (i.e., calling those who voted for Bush stupid) that discussion becomes impossible.

The thing is, I have and will again vote Democrat where appropriate. I just lump this exit poll fixation into the group I spoke of above in which terms like "stupid" apply.

I guess I'd take this more lightly but it seems your comment was very serious. We need to stop listening to the Michael Moores (or Rush Limbaughs)and get to the real issues.

What's the point of all this conspiracy nonsense? Dem's who persist in this idiotic vein do a wonderful service to conservative voters. When are the Dem's going to stop using class and racial warfare-type tactics, like suggesting long voting lines and a few touch screen glitches were limited to predominately African-American districts or poor districts, thereby suggesting a conspiracy to steal the Ohio vote? Long lines and voting glitches occured in heavily-Republican districts as well!

I for one am ready to vote for a Democrat but this kind of conspiracy crap turns me away!

rlprather:

Stupidity is a poor term, ignorance better, but simple mistake better yet. First draft final draft is the rule here I guess.

As far as finding out how other people live, I'm confused. Time off is mandated generally in all states. Paid time off is mandated in many states. Many employers will give time off, paid, even where not mandated. Early exit polls certainly are more likely to show those who took time off from work, relative to those who voted later in the evening. The point is responsibility and accountability, and the entitlement mentality that has grown beyond belief, to which the bumper sticker lament is directed.

Also, while I do think there is a definite trend that would support my thoughts on timing, understand that my comment was meant to draw a smile as much as anything. Most of the comments from both sides on these issues are so bent one way or the other (i.e., calling those who voted for Bush stupid) that discussion becomes impossible.

The thing is, I have and will again vote Democrat where appropriate. I just lump this exit poll fixation into the group I spoke of above in which terms like "stupid" apply.

I guess I'd take this more lightly but it seems your comment was very serious. We need to stop listening to the Michael Moores (or Rush Limbaughs)and get to the real issues.

rlprather:

Stupidity is a poor term, ignorance better, but simple mistake better yet. First draft final draft is the rule here I guess.

As far as finding out how other people live, I'm confused. Time off is mandated generally in all states. Paid time off is mandated in many states. Many employers will give time off, paid, even where not mandated. Early exit polls certainly are more likely to show those who took time off from work, relative to those who voted later in the evening. The point is responsibility and accountability, and the entitlement mentality that has grown beyond belief, to which the bumper sticker lament is directed.

Also, while I do think there is a definite trend that would support my thoughts on timing, understand that my comment was meant to draw a smile as much as anything. Most of the comments from both sides on these issues are so bent one way or the other (i.e., calling those who voted for Bush stupid) that discussion becomes impossible.

The thing is, I have and will again vote Democrat where appropriate. I just lump this exit poll fixation into the group I spoke of above in which terms like "stupid" apply.

I guess I'd take this more lightly but it seems your comment was very serious. We need to stop listening to the Michael Moores (or Rush Limbaughs)and get to the real issues.

Sonnyw:

Can you point to predominantly GOP precincts in Ohio that suffered from long lines of more than 3 hours?

On reading over the PDFs of the exit polls what really struck me was how hard they were trying - before the votes were even counted - to make sure the explanation for the swing to Bush was going to be religious/moral issues. The way the questions are structured, they have multiple angles of attack designed to get people to commit to being morally motivated and then classify them as voting for those reasons if they qualify on any one of the question sets. That virtually guarantees that the poll would produce a result skewed to suggest that people voted the way they did on moral/religious issues, even if they really only admitted to that motivation in an extremely limited context.

The larger implications of this are that the overall value of the exit poll is compromised because the pollsters clearly went in looking for certain answers and didn't put their first priority on clean, accurate results.

Dave
http://www.elitistpig.com

On reading over the PDFs of the exit polls what really struck me was how hard they were trying - before the votes were even counted - to make sure the explanation for the swing to Bush was going to be religious/moral issues. The way the questions are structured, they have multiple angles of attack designed to get people to commit to being morally motivated and then classify them as voting for those reasons if they qualify on any one of the question sets. That virtually guarantees that the poll would produce a result skewed to suggest that people voted the way they did on moral/religious issues, even if they really only admitted to that motivation in an extremely limited context.

The larger implications of this are that the overall value of the exit poll is compromised because the pollsters clearly went in looking for certain answers and didn't put their first priority on clean, accurate results.

Dave
http://www.elitistpig.com

On reading over the PDFs of the exit polls what really struck me was how hard they were trying - before the votes were even counted - to make sure the explanation for the swing to Bush was going to be religious/moral issues. The way the questions are structured, they have multiple angles of attack designed to get people to commit to being morally motivated and then classify them as voting for those reasons if they qualify on any one of the question sets. That virtually guarantees that the poll would produce a result skewed to suggest that people voted the way they did on moral/religious issues, even if they really only admitted to that motivation in an extremely limited context.

The larger implications of this are that the overall value of the exit poll is compromised because the pollsters clearly went in looking for certain answers and didn't put their first priority on clean, accurate results.

Dave
http://www.elitistpig.com

ive just noticed that the NEP issued a statement on Nov2 which appears to be their final summary at the end of the day about their activities for the day - it is still on exit-poll.net under the prominent link titled "National Exit Poll Methodolgy Statement" which links to MethodsStatementNationalFinal.pdf (note the word "final" - and it still hasnt been updated).

This document specifically says "The National exit poll was conducted at a sample of 250 polling places among 11,719 Election Day voters representative of the United States. In addition, 500 absentee and/or early voters in 13 states were interviewed in a pre-election telephone poll." That is, the NEP's apparently "final" statement explicitly states that they only did 12,219 interviews.

As we know from Scoop, by 1pm the following day, the NEP was reporting that they had actually conducted 13,660 interviews - 12% more than they had officially stated - which is starting to test the limits of credulity, given that these people are supposed to count things for a living.

One hypothesis is that there was much interview-stuffing on Nov3 to produce the requisite swing to Mr Bush. To the extent that the NEP's 'final' statement is actually final (and they haven't updated it), then it would seem to render irrelevant all of the other discussions about chatty democrats and gender bias and the rest.

Is this a smoking gun?

there's a little bit more here

ive just noticed that the NEP issued a statement on Nov2 which appears to be their final summary at the end of the day about their activities for the day - it is still on exit-poll.net under the prominent link titled "National Exit Poll Methodolgy Statement" which links to MethodsStatementNationalFinal.pdf (note the word "final" - and it still hasnt been updated).

This document specifically says "The National exit poll was conducted at a sample of 250 polling places among 11,719 Election Day voters representative of the United States. In addition, 500 absentee and/or early voters in 13 states were interviewed in a pre-election telephone poll." That is, the NEP's apparently "final" statement explicitly states that they only did 12,219 interviews.

As we know from Scoop, by 1pm the following day, the NEP was reporting that they had actually conducted 13,660 interviews - 12% more than they had officially stated - which is starting to test the limits of credulity, given that these people are supposed to count things for a living.

One hypothesis is that there was much interview-stuffing on Nov3 to produce the requisite swing to Mr Bush. To the extent that the NEP's 'final' statement is actually final (and they haven't updated it), then it would seem to render irrelevant all of the other discussions about chatty democrats and gender bias and the rest.

Is this a smoking gun?

there's a little bit more here