Mythbusters: 2004 Exit Polls and the myth that Kerry was up

[maseman]

I saw this posted by PNWMOM on a different thread. This is no attack on you PNWMOM but just some clarification.

Here's what you said in your post..."Remember early in the day on Election Day 2004? Exit polls showed Kerry winning by a landslide -- 7 points separated him and Bush. But those 7 points completely disappeared by the end of the day, and the final exit polls were "adjusted" to accord with the "actual" voting results."

The reason why we saw some early exit polling Kerry way ahead is because of two reasons. First exit polling data is NOT supposed to be leaked or even discussed before it is completed at day's end. The data in 2004 was leaked by a reporter who used a cell phone to text some early exit polling reults. Most of the early exit poll data is obviously in the eastern time zone and most heavily in the northeast where it is extremely heavy Dem compared to the rest of the country. So of course the exit polling leaked early showed Kerry ahead since the polling data at the point was in mostly Dem states.

Second, the early exit polling data was still a very small sample size which adds in more standard error and statistical "bounce." Small sample coupled with data from heavy Dem areas coupled with a reporter releasing data early combined for some serious optimism including even myself. I had no idea what was going on that day but saw the early numbers, got excited, and then got depressed around 10PM that night. However I know the people who run the exit polls and know that that is exactly what happened.

This year security is going to be much more thorough to NOT allow data to be released early by leaks. Not until sample sizes are large enough and data is solid will we see any exit polling data.

Last, the numbers are never "adjusted" afterward for the exit polls. It takes a few days of final data collection and tabulation to report the final exit poll results. Data can change slightly but the exit poll is what it is regardless of the final general election vote outcome.

[DemocratSinceBirth]

1. Raw Exit Polls Had Dukakis Beating Bush Pere In 88

And Reagan 54-Mondale 46

Nuff said...

[BeFree]

44. But

The OP says:

Last, the numbers are never "adjusted" afterward for the exit polls. It takes a few days of final data collection and tabulation to report the final exit poll results. Data can change slightly but the exit poll is what it is regardless of the final general election vote outcome.


The 2004 exit polls were severely adjusted and once adjusted came up with numbers that were unsupportable and shameful. Basically, the adjusted polls said that 110% of bush 2000 voters voted again in 2004!

So the writer quite obviously has only a clue. Not to be believed.

[OnTheOtherHand]

50. familiar nonsense

What the exit poll shows, in a nutshell, is that a lot of people couldn't remember who they voted for. At least, that's the most likely interpretation; I personally doubt anyone bothers to lie about that.

As you know, the same thing happens in other polls: more people report having voted for the incumbent than reasonably could have.

[BeFree]

53. The raw numbers didn't show that

The raw numbers were balanced. It was the final adjusted exit poll that was out of balance. And you oughtta know that by now.

But you do make me laugh when you write ".....a lot of people couldn't remember who they voted for."

You are quite the comedian. Eh?

[OnTheOtherHand]

54. you're ignoring every other poll

If you don't look at evidence, of course you won't believe it.

[Buzz Clik]

2. I was always amazed by those who used exit polling numbers as evidence of fraud in OH.

Exit polling is brain candy for the impatient. Take it as a really rough snapshot and nothing more.

[maseman]

3. That statement is for all polling

[Buzz Clik]

5. Well, ok. I agree sorta, and sorta disagree.

Your standard polling can be scientific with a solid marging of error. Exit polling is much more difficult, less scientific, and is associated with far greater errors.

So, I agree that all polling is imperfect, but exit polling is much worse. Attempting to read exit polls in the same light as typical polls is where the problem begins. After years of being exposed to standard polls we know that leads of 3 or 4% are not real, but when leads get beyond 5%, it's pretty solid. For exit polling, it's not as clear but we know that the exit polling is far more random.

[maseman]

8. Well mixed bag here

Exit polling is in some ways a better poll for two reasons. First the sample size is much larger than a standard n=1,000 for the national polling. Sample sizes will be better for less statistical bounce. Second, in person polling is better than telephone because you have two people face to face. You can tell if someone doesn't understand a question, etc. Face to face is usually better than telephone.

As for randomness...it is certainly random. But you do have the worry of will people be honest. Also is one party much more likely to answer questions than the other.

Exit polling is not really used to find out the D versus R win and loss but to find out what was on voters minds such as war, economy, moral issues, etc.

[DemocratSinceBirth]

9. That , LIHOP, And MIHOP Are The Most Divisive Issues On DU

Professor Sam Wang used meta-analysis( he aggregated all the polls) to predict the results of 04 and nailed the pop and EC Vote...

I do think Gore got jobbed in 00...But that was more about incompetence and gaming the system than out right fraud...

[dkofos]

4. Exit polling used to be accurate in this country and still is for the rest if the world.

Unless the media is owned by the opposition.

[maseman]

7. But the polling company is not owned by the media

They are not affiliated with the media except as a client paid to do the research. Media however can use data from any poll to spin spin spin.

[Frumious B]

6. Thanks for this. I'm not convinced by the massive "vote flipping" theorists myself.

I'm glad to see a rational explanation. Hopefully, in a couple of days they all get to move on from "Obama can't win because all the machines are rigged against him" to "They only allowed Obama to win because they expect him to fail." That'll be a nice change of place, eh?

[backscatter712]

10. Bullfuckingshit.

Exit polling is used in the rest of the world as gold-standard evidence to prove election fraud.

Unless the exit polls are owned by the right-wing corporate media, and cooked in secret boiler rooms to ensure the published exit poll results matched the cooked election results.

That's what happened in 2004.

Which is why we have independent election verification exit polling done by activists this time.

Sorry, right-wing thugs. You can't cook those numbers. And these will be polls explicitly designed to catch election fraud.

[OwnedByFerrets]

11. Thank you!! Finally a intellegent post.

Those who think that Kerry actually lost are fucking idoits.

[demwing]

25. Fucking Idoits?

Is that at all similar to "Get a brain morans!" ?

[OwnedByFerrets]

60. FU...I'm sure you've NEVER made a fucking typo.

[demwing]

61. I make them all the damn time, lol

I still find it hilarious when you insult someone's intelligence, and misspell your insult.

BWAHAHAHAH!!!!!

[beachmom]

27. Oh, yeah? Provide the evidence. Not statistical data, but precisely

how it was stolen. Not one theorist has done that. Not even RFK, Jr. The only thing I am convinced of is voter suppression works. But you can't contest an election based on votes not cast.

I have examined poll data prior to Election Day, and it clearly showed Bush overall being slightly ahead in places like Ohio and Florida. And that is how it ended up on Election Day.

[maseman]

12. WRONG again!

You are nuts and delusional bud. No cooking in back rooms. No right-wing media ownership. I was there. You think I am a RW shill or something? The EXACT same company that did the exit poll in 2004 (Edison Research) is the same company that did the 2006 and now 2008 elections.

The polls were leaked early and were not representative of the country. You can argue something you don't know anything about but that doesn't mean you are correct.

[DemocratSinceBirth]

13. See Post Nine

I hate these debates...

But I sure as shit don't think Dukakis was robbed or Reagan-Mondale was remotely that close...

[ProSense]

14. What did the exit polls in Ohio indicate

when 350,000 people were turned away from the polls?

[maseman]

15. That isn't an election exit poll problem

That was a SOS Ken Blackwell is a facsist problem that is a huge voter fraud scandal. That has nothing to do with exit polling.

Neither does Machines in Ohio that flipped votes. I don't trust the Diebold or ES&S machines any more than the next person. But the point that I made in the original post doesn't have anything to do with this.

[ProSense]

16. The point you made in the OP has everything to do with this

That was a SOS Ken Blackwell is a facsist problem that is a huge voter fraud scandal. That has nothing to do with exit polling.

Neither does Machines in Ohio that flipped votes. I don't trust the Diebold or ES&S machines any more than the next person. But the point that I made in the original post doesn't have anything to do with this.

You can't claim the exit polls were wrong and the pre-election polls were right, and then claim huge election fraud in Ohio and machine fraud.

If those 350,000 people were allowed to vote, the exit polls would have been right.

[foo_bar]

20. exits aren't supposed to poll people who didn't vote

If those 350,000 people were allowed to vote, the exit polls would have been right.

Are you saying the people who were denied told pollsters they voted (i.e., how they would have voted)?

[ProSense]

23. No, that wouldn't make sense.

The point is that if 350,000 votes weren't suppressed, Kerry would have won Ohio.

[foo_bar]

26. no argument there, just "the exit polls would have been right" part

For all we know the exits might have been 60-40 Kerry in Ohio if illegally suppressed votes were taken into the equation, it's not as though the answer they got from people leaving polls was the "correct" one in terms of suppression or intimidation based fraud.

[maseman]

31. No one is arguing that point

But exit polls do not include people who didn't vote nor are they supposed to. That is a different (and important) issue but not what I was talking about.

[ProSense]

34. You must understand that

if exit polling indicated a trend toward Kerry, and people were systematically turned away from the polls who were likely to vote for Kerry, that would explain the odd results.

To claim that the pre-election polls were right (and they were all within the margin of error) and exclude the reality of vote suppression and other fraud, doesn't make sense.

Despite overwhelming evidence that have shown the machines to be a huge problem, some still refuse to acknowledge the impact of machine irregularities in 2004.

Even if someone cannot come to grips with machine fraud, ignoring the impact of vote suppression on the outcome of the election, and then using that outcome to justify the pre-election polls that were withing the margin of error, makes no sense.


Edited typos.

[OnTheOtherHand]

52. no, we don't "understand" that at all

How are exit polls supposed to capture a trend among people who were turned away from the polls?

We can try to sort out the magnitude of vote suppression in Ohio and elsewhere, but the exit polls aren't likely to help.

[grantcart]

38. Really? Having lived overseas for decades and following local elections

The only time 'exit polls' have been used in the way that you refer to is where there are hostile regimes and international observers have been called in as official voting observers.

Can you site a single occurence in the developed world where exit polls were used to prove election fraud?

[yodermon]

55. Ukraine 2005 as mentioned downthread. n/t

[OnTheOtherHand]

56. how were they used to "prove" vote fraud?

(I assume you mean 2004.)

Did the Ukrainian courts cite them? Did international observers cite them? What exactly are you claiming?

I do think that the first Ukraine run-off was stolen, and that the (multiple) exit polls provided some additional evidence for that conclusion. But I don't think they proved it.

[Drunken Irishman]

39. Dude, exit polls, like every other type of poll, has a MOE.

The state polls showing Kerry up were slim leads. He never had a large lead in Ohio or Florida. In Ohio, it was something like 51-47 and four points is well within the MOE of an exit poll.

Add in the fact these numbers were released around THREE EST and you can see why they were probably wrong. The polls were still going to remain open for hours, so the results should have been questioned.

[Overseas]

17. IF I had not been watching election 2004 live for hours into the night I might believe you.

But I saw with my own eyes and to my own great relief that the Abu Ghraib administration was coming to an end in election 2004. In the later hours on election night, exit polling and vote totals were in agreement, with Kerry 51 and Bush 48 and it seemed like we would have our victory. I saw that on TV. During the night, things changed. THere was an unprecedented SIX POINT SHIFT in vote totals-- Kerry DOWN THREE, and Bush UP THREE points. Then yes, it really happened, the exit polling was adjusted to reflect the new vote counts. Thank goodness others have compiled the televised information and other reports in comprehensive documentaries like "Stealing America: Vote by Vote" www.stealingamericathemovie.org , and election integrity reports that can be easily accessed at sites like www.velvetrevolution.us, www.bradblog.com, www.verifiedvoting.org .

We criticized other counties' elections when their exit polling differed from their actual vote counts. Exit polling has been more accurate than other kinds of polling because, we had been told for years, those questions ask "What did you just do?" as opposed to "what is your future intent?" That is why exit polling had been used to judge the fairness of international elections.

I realize that our numbers from many different polls are comparatively higher in 2008 than they were in 2004, so I am hoping our clear lead will be allowed to stand once we vote. But in 2004 on election day, while things were too close for my comfort, given the abuses of America's core principles by the Bush Cheney gang, it still looked like Kerry would win a narrow victory.

I watched election 2004 on TV and blogged that night and read reports of voting anomalies coming in from swing states all over the country. I was reporting on election manipulation using the GOP's "dribs and drabs method" -- a few hundred votes here and there, flipped electronically or misdirected via robocalls and robofliers or discouraged by hours long lines due to "careless" allocation of voting machines. Each incident reported would be brushed away with "that's too small to affect the outcome." No one added up all the incidents until later reports were compiled and then the magnitude of all those dribs and drabs added together.

I saw myself that exit polling in Ohio had Kerry 51 and Bush 48 and that agreed with the vote totals late in the evening. That was exit polling data broadcast continuously on election night. Thank goodness someone else captured screen shots of those totals. I remember being very glad that the Abu Ghraib Administration was going to end. Finally, the destructive Bush administration would end. The exit polling and vote count had Kerry with a three point lead. But during the midnight hours, voting totals diverged from the exit poll data BY SIX POINTS-- Kerry down three and Bush up three.

After the vote totals reported by the Republican Secretary of State diverged so much, the exit poll data was adjusted to match the reported vote totals. In other countries, such a discrepancy would not lead to adjusting the exit polling, it would be evidence of potential election fraud.

So the networks launched into tortured discussions of how exit polling is faulty, even though we had used the very same internationally renowned pollster to judge other countries' elections in the past. Those tortured discussions also ignored the anomalies being reported, like purging voters from the rolls and distributing far fewer machines per voter to Democratic areas than Republican, even though Democratic registrations were way up, etc. Reports compiling those oddities didn't come out until months later.

IF YOU HAVE FORGOTTEN or did not see it with your own eyes, you can review the TV footage from that night in the documentary available for free download now: Stealing America: Vote by Vote
www.stealingamericathemovie.org

[Mojorabbit]

35. I watched it too

that night and I agree with you totally.

[suston96]

18. Exit polling caused a silent revolution in the Ukraine in 2004.........

http://en.wikipedia.org/wiki/Orange_Revolution

Protests began on the eve of the second round of voting, as the official count differed markedly from exit poll results which gave Yushchenko up to an 11% lead, while official results gave the election win to Yanukovych by 3%. While Yanukovych supporters have claimed that Yushchenko's connections to the Ukrainian media explain this disparity, the Yushchenko team publicized evidence of many incidents of electoral fraud in favor of the government-backed Yanukovych, witnessed by many local and foreign observers. These accusations were reinforced by similar allegations, though at a lesser scale, during the first presidential run of October 31.

[foo_bar]

22. at least that's what our State Department wants you to think

A key part of the media game has been the claim that Yushchenko won according to "exit polls". What is not said is that the people doing these "exit polls" as voters left voting places were US-trained and paid by an entity known as Freedom House, a neo-conservative operation in Washington. Freedom House trained some 1,000 poll observers, who loudly declared an 11-point lead for Yushchenko. Those claims triggered the mass marches claiming fraud.

http://www.atimes.com/atimes/Central_Asia/GA20Ag01.html

Western-funded exit polls showed Yuschenko was gonna win, Russia-funded that Yanukovych was gonna win. We only heard about the former tho.

http://www.indymedia.org.uk/en/regions/liverpool/2004/12/302679.html

More provocatively, the US and other western embassies paid for exit polls, prompting Russia to do likewise, though apparently to a lesser extent. The US's own election this month showed how wrong exit polls can be. But they provide a powerful mobilising effect, making it easier to persuade people to mount civil disobedience or seize public buildings on the grounds the election must have been stolen if the official results diverge.

http://www.guardian.co.uk/world/2004/nov/26/ukraine.comment

Pro-Yanukovich television stations used Russian exit poll results to vouch support for him as outcries against falsification increased and the CEC stalled in delivering official results. In addition, the exit poll did not really exist in Ukrainian political culture before 2004. Voters, who already believed the vlada would falsify the election months in advance, had a new tool at their disposal to reconfirm their suspicions. Steele posits that the exit polls, funded by the US and Russia, influenced the voters to take to the streets.

http://leopolis.blogspot.com/2004_12_01_leopolis_archive.html

---

In related shenanigans:

The recall's proponents sponsored an exit poll, supervised by Penn, Schoen & Berland, an American firm whose clients have included Bill Clinton and Michael Bloomberg. Sometime before the polls closed on Aug. 15, Penn, Schoen reported that 59 percent of Venezuelan voters had said yes to throwing the president out of office.

A few hours later, the official count, by an election commission under Mr. Chávez's control, declared him the winner, with 58 percent of the total. Both the Organization of American States and the Carter Center, the Atlanta-based human rights organization founded by Jimmy Carter, said that their observers had seen no irregularities at the polls. In response to the exit poll, they called for a random audit at selected polling stations and again found nothing suspicious.

Mr. Schoen acknowledged in an interview that the poll's field workers were recruited by a group that helped organize the recall, but he said the volunteers had been trained to conduct the poll professionally, and that his firm would have no reason to put its reputation at risk by participating in a fraudulent poll. The recall's supporters continue to believe the election was stolen.

http://www.nytimes.com/2004/10/17/weekinreview/17plis.html?_r=1&oref=slogin

According to co-director Maria Corina Machado, Súmate is an objective non-partisan civil association. When asked why Súmate has worked exclusively with the Venezuelan opposition since its inception in 2002, Machado said that their overtures to the government were regularly rebuffed. Machado neglected to mention that one of the reasons the government may have been hesitant to work with her group is because she was a participant in the 2002 coup that briefly overthrew Chávez—she signed the infamous decree of dictator-for-a-day Pedro Carmona. She is currently being investigated for treason, for having received funds from a foreign government (the U.S.) earmarked for ousting the Chavez government.

http://www.venezuelanalysis.com/analysis/644

[suston96]

28. Fact is a new election in the Ukraine after that silent revolution reversed the original election.

The exit polls were right in the Ukraine. They were right here in 2004 also in Ohio. Because of our easily corruptible election process there is no way to prove this.

They are still investigating criminal complaints about the elections in 2004 in Ohio - the capital of election corruptibility.

The problem is the way people think: as long as my candidate and my party win, who gives a shit about corrupted elections.

[foo_bar]

29. are you suggesting Chavez stole the election, with Jimmy Carter's complicity?

Mr. Schoen acknowledged in an interview that the poll's field workers were recruited by a group that helped organize the recall, but he said the volunteers had been trained to conduct the poll professionally, and that his firm would have no reason to put its reputation at risk by participating in a fraudulent poll. The recall's supporters continue to believe the election was stolen.

http://www.nytimes.com/2004/10/17/weekinreview/17plis.html?_r=1&oref=slogin

A few hours later, the official count, by an election commission under Mr. Chávez's control, declared him the winner, with 58 percent of the total. Both the Organization of American States and the Carter Center, the Atlanta-based human rights organization founded by Jimmy Carter, said that their observers had seen no irregularities at the polls. In response to the exit poll, they called for a random audit at selected polling stations and again found nothing suspicious.

(ibid.)

I mean you're reiterating a far right wing theory, for what it's worth.

[suston96]

32. Don't know what you are talking about. Chavez, Jimmy Carter?

The revote in the Ukraine after the silent revolution proved the exit polls were correct. Exit polls have been extremely useful in the past, in every country where democratic elections are held.


Orange-clad demonstrators gather in the Independence Square in Kiev on 22 November, 2004.
On some days, the number of protesters in the center of Kiev reached hundreds of thousands (one million by some estimates).

[foo_bar]

33. "in every country where democratic elections are held"; except Venezuela, and everywhere else?

Since 2000, Carter Center delegations have overseen local and national elections in Peru, the Dominican Republic, Venezuela, Nicaragua, Guyana, East Timor, Zambia, Sierra Leone, China, Kenya, Mozambique, Guatemala, Indonesia, Congo, and Ethiopia. In most of those cases, the Center opposed the use of independent exit polls to verify whether its assessments of the elections' integrity were accurate.

http://www.discoverthenetworks.org/groupProfile.asp?grpid=7314

If you believe Uncle Sugar's Freedom Haus had our best interests at heart in the Ukraine, fine, but when you write "in every country where democratic elections are held" it seems as though you aren't reading any of these citations.

as long as my candidate and my party win, who gives a shit about corrupted elections

Were you referring to Chavez or Jimmy Carter (seeing how I referenced their "stolen election(s)" just prior to your comment)? Or are you suggesting Democrats stole the 2006 election?

[suston96]

36. I am talking about the Ukraine in 2004. You are responding with Chavez and Carter....

I am talking about elections in this country that have been corrupted and have no validating process except for pre-election and exit polling.

What a fucking disgrace! The prime historical model of democracy which started direct rule by the people, cannot come up with a tamper proof electoral process.

Until a valid redundancy is added to our electoral process whereby every vote cast and all tabulations are manually accountable then pre-election and exit polling must be used as they have been successfully in the past.

[foo_bar]

41. so Freedom House exit polls in the Ukraine = good, the same exit polls in other countries = ?

as long as my candidate and my party win, who gives a shit about corrupted elections

Can you clarify, since you can't discuss the logical conclusion of your beliefs?

elections in this country that have been corrupted and have no validating process except for pre-election and exit polling

The real validation mechanism is auditing, and having a record to audit in the first place. Exit polls are nice but they're like audits that depend on people's say-so and only examine a small fraction of the sample a (robust) random audit would examine. Of course audits can be corrupted too, but so can exit polls with much less effort (if you believe the librul media in the UK and Venezuela), so clinging to this vision of the Orange Revolution as a fun grassroots democracy party is a simplistic view, one which happens to coincide with the administration's simplistic view, and the view of reactionaries in Sumate and the GOP alike. I guess the question is, will this belief system supercede being a Democrat on November 5th? After enough 2006-style elections the freepers will pretty much own this "issue" if they don't already:

The margin of error between Carter’s certified fair-and-square ballots and the independent exit poll results constituted a swing of almost forty points — a statistical impossibility. Chavez counted on Carter leaning his way — Carter’s history of promoting anti-American dictators is no secret.

http://thespisjournal.wordpress.com/2006/09/23/the-jimmy-carterhugo-chavez-alliance/ -a wingnut

A statistical impossibility unless the exit poll is corrupt (or simply the 1 in 20 that exceed the MoE at 95% conf), which appears to have been the case in Venezuela, and possibly the Ukraine if you believe the Guardian (or even indymedia) more than the bush administration's statements at the time.

[suston96]

45. I am not talking about Chavez and Carter.....Stuff that because I am not interested....

Americans don't give a shit about fixing elections in this country as long as their candidate and their party wins.

Matter of fact, I believe that there are many, a disturbing number, who want corruptible elections - that is, who seek access to manipulating election results, starting with voter suppression and through the actual voting and tabulation processes.

That is why pre-election and exit polling are vital factors until a tamper proof process is developed and put in place.

That's my final word on this. I am repeating myself. By all means continue your defense of corrupted elections.

[foo_bar]

48. did you catch the paragraph about auditing results? no thoughts?

That is why pre-election and exit polling are vital factors until a tamper proof process is developed and put in place.

If you want to defend corrupted elections, insist on exit polls as a validation mechanism. Right-wingers in Venezuela (reputedly CIA funded) and possibly the Ukraine attempted to subvert democracy with that very technique, so it's too bad you can't be interested in information you don't already believe.

Stuff that because I am not interested....

Okay, but if you believe exit polls have magical properties making them impervious to corruption in a way elections (and especially audited elections with a paper trail) aren't, you're defending corrupted elections, maybe even in the Ukraine given the spread between differently funded exit polls and the players involved.

[slackmaster]

19. I refuse to talk to exit pollsters

My vote is nobody's business but mine.

[karynnj]

21. The exit poll were at the state level not the national level - Why not use the reasons given by the

people who did the exit polling?

The numbers stayed in Kerry's favor though the early evening. These were also not "raw" numbers, but numbers weighted by experimental design factors that were based on past elections, new registrations, and other things. When the actual weights (based on the number of people voting in various places becomes known), they are used. The unexpected large turn out of evangelicals could have played a role here.) Kerry was still ahead at that point in Ohio. They also are routinely "reconciled" with the official numbers after the fact.

Had what you said been true, the lead statistician would never have issued a report defending his design and explaining that the implementation was flawed because of "shy" Bush voters.

The main intent of the exit polls had been to understand the dynamics of the election. Which groups voted which way and for which reasons. As they exited, it became common to use them to predict the states, However, people waited to determine that the actual results from the sampled precincts were matching the result and that the sampled precints were similar to early results in some of the non-sampled precincts in the same strata.

I had expected in 2004 to hear an admission that the experimental design of the exit polls was flawed. This could happen even with the best statistitians in the world. One way is that there could be a factor that made otherwise similar people vote in very different ways that was not included in the design - then if a disproportionate number of precincts of one type were chosen.

The report blamed the implementation instead. There conjecture was that as mostly young people were picked as the people doing the exit poll, there were far more Bush voters who refused to be exit polled.

[beachmom]

24. My suggestion is that even IF it is leaked out, view it as a random bunch of numbers

completely devoid of any meaning.

[amborin]

30. glad to see this clarified...

[Drunken Irishman]

37. The raw numbers should be ignored. However, demographic numbers are pretty spot on.

So look at those Tuesday and those are actually leaked BY the networks themselves.

[JJ]

40. As an Ohioan

It was stolen, by Kenneth Blackwell and the Republican party.

Thousands of Ohioans didn't stand in 8 and 12 hour lines to vote for 4 more years of the same shit.

Point of fact, the numbers were were "adjusted" to match the official vote count. They changed their Likely Voter model to make the exit polls match the official results.

There was a video posted on here last week, that spelled it all out. http://www.stealingamericathemovie.org/index.html (pay special attention starting at about 21 minutes in.)

Watch the video, then tell us how Kerry really lost in Ohio in 2004.

I suppose next you'll say Bush won Florida in 2000.
Gore won Florida in 2000 easily, and the theft took place in Duvall county, not in any of the counties that were contested. 26,000 African American votes "disappeared" there.

[marlakay]

43. Saw a guy on the news last night showed how in 5 min

he was able to hack into machine and change it. But I actually think we will win this time because it is so obvious everyone wants Obama this time it would be definite proof of stolen election. And this time we have democratic congress and senate to fight the results.

All they might do is fuck it up to where he wins by only a few points instead of a landslide which it should be...

[Cessna Invesco Palin]

42. You and your fancy mathematical statistics can go to hell.

Kerry obviously won Georgia and Alabama in 2004 and only a Freeper would say otherwise! So there!

[HughMoran]

46. I knew this was a mistake on your part

There are some people who'd prefer to die than agree with you.

[Mayberry Machiavelli]

47. Does anyone know if exit polls this year will reflect early voting results as well?

Has there been exit polling of early voters? Otherwise, if early turnout is heavily Obama, election day turnout in FL, CO etc. will be disproportionately McCain in exit polls because that's who's left.

[maseman]

49. Yes they are

I just asked Edison about this and they are doing pre-election surveys now on the phone and mail regarding early voting. They must do this to even be relevant.

[OnTheOtherHand]

51. I don't get the premise of your OP

I don't know what the "7 points" is about, but it isn't a myth that the national tab posted around 7:30 PM Eastern did show Kerry ahead. That's consistent with the Best Geo estimates in the E/M evaluation report (although of course the tabs wouldn't be based on Best Geos). Lots of these results were freely posted, not "leaked."

[foo_bar]

57. here's the post in question

Remember early in the day on Election Day 2004? Exit polls showed Kerry winning by a landslide -- 7 points separated him and Bush. But those 7 points completely disappeared by the end of the day, and the final exit polls were "adjusted" to accord with the "actual" voting results.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=132x7689214

In related news, Democrats are Hitler and that's before watching TV, so I'm detecting a similar theme to the rhetorical last resorts of broken glass Replicants and tinfoil content producers awaiting a similar outcome. As for the context provided by the original OP, well, pseudoFAQ number 3a stroke nine oh, even though the only question posed was "do you remember early in the day?" (maybe the Snopes approach is the way to go, beginning with a proposition and deducing which questions are raised by it)

[OnTheOtherHand]

58. yeah, the "7 points" is just mysterious

It has been interesting to stumble across mutant memes ("the United Nations uses exit polls around the world..."; "exit polls were always accurate until Florida 2002...").

I just want to be clear that the exit poll discrepancies really did exist. Otherwise, it's sort of like watching Eliza and Parry. Not that there's anything wrong with that....

[foo_bar]

59. at least Eliza asks questions

Not pertinent ones but her heart's in the right place. I wonder if this isn't adaptive radiation of the Y2K meme, the fear of social meltdown from over-reliance on conductors, even though 2000 was decided by hanging bits of wood pulp and people in black robes. From a memetic-epidemiological standpoint we're almost certain to hear anti-Obama poll conspiracies in the near future, the anti-DUers are already picking cherries and calling for sedition charges against Gallup, which means they're setting themselves up for disappointment if the past is an indication. I guess it's like SARS, you don't even have to be from the same species.