Simon: To Those Awaiting Prelim Analysis of Exit Polling....

[understandinglife]

.....Data Released 12/31/04:

It's a lot of data to wade through but what has been released is more or less in line with what we had. What is curious and potentially very misleading is that the data from the 12:20 - 12:25 a.m. time of updates (which I was able to capture on election night, but which otherwise apparently would not exist anymore) is missing from this otherwise complete set. This is crucial, since it shows that (to take the full national sample), Kerry maintained his 2.6% lead (rounded to 3%) when 13,047 of the eventual "13,660," rather than the 11,027 from the last pre-"adjusted" batch released here (7:33 p.m.), had been counted.

From the data released it would look somewhat plausible that a late Bush swing (between 11,027 counted and 13,660 counted) could have accounted for the shift in the EPs from Kerry51/Bush48 to Bush51/Kerry48; but the missing sweep (the 12:20 a.m. timeframe) shows that this was not possible.

So my question is, where is that group of data; or did whoever released this stuff forget that I printed out the missing link? It appears that a partial set of data was released (missing a crucial piece) possibly in the hope of creating the impression that all was according to Hoyle. It is critical to examine this newly released data in conjunction with the screenshots which I possess and have distributed to certain recipients who (for obvious reasons) will not be named.

Among these screenshots, the national sample at 12:23 a.m. is public and can be referenced as Appendix A of the Simon/Baiman paper at:

<http://www.freepress.org/departments/display/19/2004/1054>

It reproduced a bit fuzzy so I'll recap: 12:23 a.m.; 13,047 respondents; Male(46%) 52%B/47%K, Female(54%) 45%B/54%K; Total 48.2%B/50.8%K.

It is also very much worth scrutinizing the breakdown by party ID for that 12:23 a.m. sample:

Dem 38% (B9%/K90%), Rep 35% (B92%/K7%), Ind 26% (B44%/K52%). Compare that to the "adjusted" sample from 1:24 p.m. Wednesday: Dem 37% (B11%/K89%), Rep 37% (B93%/K6%), Ind 26% (B48%/K49%). Remember: the # of respondents barely changed, so the changes are due almost entirely to "renormalization" which, if it is without justification, can better be called flat out fudging.

What we see is that the sample shifts from 38%Dem/35% Rep to 37%Dem/37%Rep (because of the huge effect of party ID on candidate preference, this shift in weighting is very much more powerful in altering the overall results than any reweighting by gender) and Independents lurch over to Bush by 7% (from B44%/K52% to B48%/K49%). Without the missing screenshot from 12:23, an argument might be made that the 2500 or so late exit poll respondents (after 7:33 p.m.) account for these shifts—anyone analyzing just the data released today would be excused for drawing such a conclusion. The missing 12:23 a.m. data shows that such a conclusion would be erroneous (it was data manipulation to match the "actual" vote counts, and not an increase in the size of the respondent group, which produced the pro-Bush shifts).

Edison/Mitofsky must present a legitimate reason for skewing their own polls to overrepresent Reublican and underrepresent Democratic voters (remember the controversy over some right-wing pre-election polls which did essentially the same thing?), while throwing a substantial share of the Independent vote from Kerry over to Bush. The only arguable reason for doing so is what has been dubbed the "reluctant Bush responder" hypothesis—the assumption being that (against all logic and observational evidence and against the evidence of the polls themselves) Republicans won the turnout battle in virtually every state across the country but that this implausible Republican advantage was masked by their innate comparative reluctance to participate in the exit polls.

No EVIDENCE, however, has been advanced to support the reluctant Bush responder hypothesis other than the tautology that the vote count had to be right and therefore the exit polls must have been wrong and this is the only way to explain it.

We still await release of the missing late exit poll data and, of course, of the raw data which would permit independent analysis of the raw numerical facts of the case.

—Jonathan Simon

(source, private email forwarded to me)

Peace.

A very, very, very "orange" and happy New Year to all of you.


"Its 7 Jan 2005: do you know who your president is?"

[althecat]

1. In relation to the where is the missing data....

This is what Mitofsky supplied... I asked for the 4pm 8pm and midnight runs.. they provided the corrected runs instead of the midnight run.

[understandinglife]

2. Wonder why? ;-) (n/t)

[Pirate Smile]

18. Didn't they say they had a computer crash right before they were about

to release the last run? I remember hearing something about this.

[Lauri]

34. Just Read about the "COMPUTER CRASH" - article from 11/4/04

http://www.washingtonpost.com/wp-dyn/articles/A23580-2004Nov3.html

"To compound the problem further, a server at Edison/Mitofsky malfunctioned shortly before 11 p.m. The glitch prevented access to any exit poll results until technicians got a backup system operational at 1:33 a.m. yesterday.

The crash occurred barely minutes before the consortium was to update its exit polling with the results of later interviewing that found Bush with a one-point lead. Instead, journalists were left relying on preliminary exit poll results released at 8:15 p.m., which still showed Kerry ahead by three percentage points."

Interesting timing.....

[Bill Bored]

36. No one on DU likes this theory

but I have suggested that Rove had the earlier polls hacked (on Mitofsky's server) to make it appear as if Kerry were ahead to get out the GOP vote.

Now, if this happened (and I know most of you would prefer to truly-believe otherwise), a server crash could be the time when the real exit poll data was inevitably reinserted into the mix, revealing that Shrub had actually taken a narrow lead.

[Lauri]

47. Why would it be assumed that the data after the crash was real?

Why wouldn't it be the time that the "new" data was inserted? The percentages seemed to swing between 12:30 & 1:30 = Going from a Kerry lead to a Bush lead. Why would it be assumed that the "new" data showing Bush ahead with be the "real exit poll data"? I don't follow.

[McCamy Taylor]

52. GOP base always votes, cant squeeze more blood from a turnip

is why usual GOP strategy is to keep Dems from voting, or, in this election, to switch votes from DEms to GOP or to create ficticious GOP votes.

The mood in Freeperville when they saw the Kerry is winning polls wasnt "Let's get out there and Vote!" it was "Oh, my God we are gonna lose!"

[davidgmills]

72. This argument is more compelling to me.

The question has always been do you tell people ahead of time what is happening and there are two thoughts about this: (1) people waon't vote because they think their candidate has lost, or (2)people will vote because they think they can pull out a victory at tthe last minute.

No 2 might be a pursausive argument if the exit polls were showing a very slim lead but that was not the case. A substantial lead, which Kerry had, would I think give people a defeatist attitude and if anything tend to make them not to put forth the effort.

[stirringstill]

60. History of Computer Crashes

There were no exit poll numbers available for the 2002 elections. Guess why. A computer crash. Apparently Voter News Service (VNS) changed software in 2002 to the tune of 10 million dollars paid to Battelle of Columbus, Ohio. Interesting article covering the demise of VNS

http://www.ropercenter.uconn.edu/pdf/141004.pdf

[nodictators]

71. The "Computer Crash" story is very hard to believe

Are they claiming they are running a national presidential exit poll and they have only one server. And that their one and only server craps out at a key time but, happily, provides a great cover story for the pro-Bush revision of the vote percentages.

Surely, a competent outfit would have something like a RAID 5 setup, and redundant servers, computers, etc.

[missouri dem 2]

3. We were screwed. And a very happy New Year to you.

The best New Years present would be the smoking gun that would expose these criminals. Lets hope that someone will step up with the evidence.

[papau]

4. Excellent data- if Kerry or Hillary do not object 1/6, they are dead as to

any support in the primaries in 08.

[Clark2008]

6. I still think Hillary won't run

She's just the right wing's decoy.

I'm thinking she's too smart to run. She knows she's polarizing.

[Bill of Rights]

5. Thank you for sharing this email

This is the information we have been waiting for.

[berniew1]

7. Have you sent this to Scoop in new zealand??

[understandinglife]

14. I haven't but I imagine Simon has shared it with all the ....

....relevant parties.

Thank you.

[bleever]

25. Look up: he's right above you in post #1.

Our very own althecat.

:toast:

[IndyOp]

27. Scoop-in-New-Zealand posted in the very first response above (n/t)

[k8conant]

8. Washington Post exit polls with 13,047 respondents:

http://media.washingtonpost.com/wp-srv/politics/elections/2004/graphics/exitPolls.html

[Quakerfriend]

9. Thank you once again

understandinglife! You the bomb!

One question, so the 12:23am screen shots that you've got differ from the ones that Mitofsky sent to althecat?? Is this correct?

[Quakerfriend]

10. Thank you once again

understandinglife! You the bomb!

One question, so the 12:23am screen shots that you've got differ from the ones that Mitofsky sent to althecat?? Is this correct?

[liam97]

11. Can there be a class action suit against Mitofsky

for misinformation to the public?

[k8conant]

12. Mitofsky answers only to his clients...

not to peons like us. :kick:

[liam97]

15. Yes, but Mitosfsky is not a feudal lord

and there must be ways of stopping them from thinking they are

[mulethree]

19. freedom of the press?

Mitofsky hides behind the press, the press paid for the poll
and mostly were the ones who published the data.

I have argued that the public donated the raw data, and never signed
anything assigning ownership to the NEP. So They can own the analysis and 'value added' all they want, but the raw data can't be
'owned'.

Any way, The press has constitutional protections. How do you
hold them to account except in the 'court of public opinion' which
they largely control themselves?

[RevCheesehead]

57. I'll readily admit that I'm no constitutional expert, but...

I thought that the press is permitted to protect anonymous sources, and permitted freedom to publish dissenting viewpoints.

But I DON'T think the press is allowed to be complicit in a cover-up of a crime - especially since they are the ones relaying public information. If they knowingly published falsified data, "they" are in big trouble, and no constitutional protection will save them from punishment.

I think we're getting closer, folks.

[The Judged]

62. R U alleging that the Re-uglicans are BLACKMAILING the press/MSM?

[RevCheesehead]

66. No, not at all.

I'm saying I don't think the press can "hide" behind any claims of constitutional protection - that is, if they are involved in the manipulation of the exit polls, or knowingly reported poll or election results which were falsified.

[The Judged]

67. AP is said to have had direct access to the vote count. Rumor has it ...

that the MSM is silent, because they are complicit AND may be charged with crimes.

I asked if you were saying that the MSM is being blackmailed by the Re-uglicans, because their silence is very unified and smacks of weakness and cowering.

IMO Either the Re-uglicans have the MSM by the short hairs, the MSM's Corporate-parent-company owners have locked down the MSM's objective coverage of Election 2004's problems to protect and advance their economic interests, or the MSM, in retrospect, realizes that they are complicit in the heist of the century and millennium: Election 2004.

Clue to MSM: It is never too late to come clean and save your soul!

[safari]

55. Re: Potential Class action lawsuit

I am not a lawyer but i mentioned this a few weeks ago. Mitofsky knew full well that his info was for public consumption regardless who paid for it or who he was working for. If a class action lawsuit was brought against him he must show what safe guards were put in place to avoid disseminating false or misleading information that can and did affect the financial markets. He will therefore have to defend himself and then you will get your answer.

[davidgmills]

80. My take on hte subject from another thread

No doubt there are contracts between him and the networks


Most likely, since the networks want the "scoop," and are paying him to collect the information, he is obligated only to divulge it to them as part of his contract for some period of time.

If there wasn't such a clause in the contract, he would be free to give the information to anybody and the networks would have paid for the collection of information, but lose the scoop.

I'm sure he's bound not to disburse it for a certain period of time.

This data is property and Mytofsky has property rights to it. Alhecat said he had the right to buy this information, so Mytofsky must have the right to sell it after the scoop period is over. If the data is for sale, purchase would be the easiest option.

If he won't sell it, a suit by a state AG might be able to acquire it. States have the right under their eminent domain laws to take the property of private citizens when doing so is in the interst of the general public. The state would have some interest in disclosure of the data because the data is being collected pursuant to a state function and in many (mayby nearly all)cases being collected on state property. But this would have to be an an action by the state and if so, maybe the only data Mytofsky has to give up is that particular state's. The state would have to pay for it because this data is Mytofsky's property and under the constitiution, the state can't take a private citizen's property without just compensation for the property.

Unfortunately, when Conyers requested it, he was really asking for it as a private citizen, not as a governmental request. He didn't offer to pay for it as far as a I know. If the data is worth a lot of money, why should Mytofsky give it up for nothing?

If a private citizen wanted the information, and Mytofsy refused to sell it, I think a private citizen would have a very difficult getting the information.

This took some thinking to work through this. Other lawyers may want to chime in. I welcome their thoughts and ideas.

[understandinglife]

13. You're welcome. The referenced screen shot...

...is Simon writing about what he has. I have quoted his email notice in this post.

Thank you.

[Time for change]

16. Thank you UL -- I hope some of this adds some perspective

The 2.6% lead that Kerry had at 7:33, after 11,027 responses from the exit poll mean that he had a 287 vote lead in this sample at that time. As Jonathon simon points out in his report
http://www.scoop.co.nz/stories/pdfs/PopularVotePaper181... , the odds against this much discrepancy from the official vot tally (2.8% in favor of Bush) approach a million to one (Actually, his report includes 13,047 responsdents, but that would change the odds much).

In order for Bush to make up the lost ground from 2.6% down to even, he would have had to obtain a vote spread over Kerry for the remaining 2,633 respondents of 10.9% -- 13.6% better than what he got for the first 11,027 respondents. The odds against this happening by chance are astronomical. In order for him to make up enough lost ground to obtain the 2.8% lead that was shown for the official vote tally, he would have had to make up more than twice that lost ground. Are we to believe that the "reluctant Bush responder effect" only went into action after 7:33 p.m.?

The odds against this happening by chance are long enough. When the existing screen shots are added, that clinches the case (My son has a screen shot for Ohio taken after 1:00 a.m. on Wednesday, showing Kerry with a 6 point lead for women and a 2 point lead for men).

[Joe Chi Minh]

17. "Are we to believe

that the "reluctant Bush responder effect" only went into action after 7:33 p.m."?

Chortle... chortle..!

[understandinglife]

20. Thank you for posting this and....

....I'm still LOLing about this sentence:

""Are we to believe that the "reluctant Bush responder effect" only went into action after 7:33 p.m.?""

Peace.

[dzika]

22. Thank you so much for posting this!

[corbett]

21. Tautology Is All They've Had Since 2001

The circular arguments which we're seeing all over the place sure are interesting. Let's hope that Mr. Conyers and company are ready with the ice pick rebuttals.

For my part, I firmly believe that they are!

[hootinholler]

26. Um I think the word you seek is sophistry not tautology N/T

-Hoot

[understandinglife]

23. From the DU archives, some relevant posts from 2-3 Nov 2004:

HOW DO YOU LOSE VOTES DURING A TALLY?

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x19715#23015

Why Did CNN Change Their Exit Poll Data for Ohio After 1:00 AM?

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x19715#22525

Kerry winning Exit Polls - FRAUD LOOKS PROBABLE

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=132&topic_id=1290765

Proof CNN is tampering with the election to cover the fraud

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=132&topic_id=1290765#1295180

CNN just changed their Ohio exit poll page

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=132x1293911

Peace.

"Its 7 Jan 2005: do you know who your president is?"

[MelissaB]

24. Wonderful links

Thank you!

[FreepFryer]

28. backstory?

Hi folks.

I'm having trouble tracking down the thread about the backstory for this data - who leaked it, why, when, etc. I have seen the Raw Story post but it doesn't say much.

Can someone add the link here?

I'd like to be very confident that this data has not been supplied conveniently.

As always, a word of thanks for all of those giving so much of their effort to investigate the massive irregularities present in the 2004 Election.

[s-cubed]

29. kick

[Melissa G]

30. Hi Jonathon and UL thought you guys and maybe TIA

could help Freeman in locating the now missing off the Internet documents which the Florida papers did documenting that Gore won in 2000 if all the votes were counted. His revised paper on another DU thread is saying Gore lost Florida and I would love for us to get that idea out of his writing with some links to back it up.
Thanks in advance for your help!

[understandinglife]

31. Regarding FL in 2000: Here are the references you request

Go to Wikipedia 2000 Election:

http://en.wikipedia.org/wiki/U.S._presidential_election,_2000

Then scroll to the University of Chicago Florida Ballot project.

Direct link:

http://www.norc.uchicago.edu/fl/index.asp

But I find this resource to be the most helpful I've found.

http://www.gwu.edu/~action/chrnuncr.html

Among other important information (quite relevant today), take a look at the US Civil Righst Commission information as well as the relevant quote to your question:

"On Jan. 10, 2001, the National Opinion Research Center at the University of Chicago was brought on to conduct the inventory. In the following months, teams of coders examined and coded undervotes and overvotes from all counties totaling 175,010 ballots. The news organizations reported their findings on Nov. 12, 2001. Their analyses showed that if the recounts underway, but stopped by the U.S. Supreme Court, had been completed, Bush would still have won by a narrow margin, but if disputed ballots statewide had been recounted Gore would have eked out a slim majority."

In other words, if you remove all the contrivances and schemes to void access to participate in the franchise and just count each citizens vote you get a different result than the catastrophe called Bush.

Here's another reference to the UC study:

http://www.consortiumnews.com/2001/112101a.html

Peace.

[Melissa G]

33. Thanks UL! I am going to email this to Freeman

:yourock:

[Bill Bored]

40. UL, is this the one published on 9/12/01 by the NYT?

Also, let's say the failed state of FL did count all the votes; would the result that Gore won FL using ANY counting standard have been aparent at the time of the election? And if so, why did the previous recounts come up with split decisions?
Thanks.

[understandinglife]

64. I don't have answers readily available to your questions...

...I'll try to spend more time with wikipedia and google and see what I can find -- should have some time late Sunday afternoon.

Thank you for your questions.

Peace.

[Melissa G]

104. Hi Understanding life ,TFC , Steven Freeman sent a thank you

to DU. Here is the link. Thanks for your and everyone on this thread's help.
http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=230454&mesg_id=230454

[TruthIsAll]

96. THEY NEVER COUNTED THE 110K OVERVOTES. GORE WON THEM 75K-35K

THEY NEVER COUNTED THE OVERVOTES.
GORE WON THEM BY 40,000.

GORE WAS DOUBLE PUNCHED ON APPROX 75,000 BALLOTS.
BUSH WAS DOUBLE PUNCHED ON APPROX 30,000 BALLOTS.

[Time for change]

38. I don't have a website address, but I do have books on the subject

By the narrow mandate that the Florida Supreme Court was given, it is true that Gore probably would have lost by a narrow amount. This is based on two considerations:

1) That mandate did not specify that over-votes were to be counted. The Gore team was not aware that the over-votes contained a lot more for them than the under-votes (Several hundred or several thousand). These were ballots that had more than one vote for President on them, but Gore's name written in at the bottom. This was due to confusion over the butterfly ballot.

2) The only county that the Florida Supreme Court did not mandate to be re-counted by hand was Broward. This was because Broward was the only county that had already been re-counted and certified. However, what was not realized at the time was that it was mis-counted, giving Bush many more votes than he deserved (Probably because of GOP pressure during the hand re-count). This came to light during the Miami Herald recount.

If either the over-votes had been counted, or Broward had been re-counted again, fairly, Gore would have won. Otherwise, no.

Of course this doesn't take into account:

3) The thousands of votes lost because of the butterfly ballot (which can be documented statistically.)

4) The tens of thousands of disenfranchased voters who were disenfranchised because they were a close match to felons.

5) The cheating over the absentee ballots.

6) Any other cheating that may have and probably did go on, related to electronic voting, which I am not able to document.


I can document most of this from the books I have, if anyone is interested.

[Melissa G]

53. Thanks Time for Change! n/t

[Time for change]

68. Let me know if you're interested in any of those references

It will take some time to dig them out because I had to study the tables, etc. a lot in order to figure some of this stuff out, and it was a long time ago -- but I still have the books. So if you're interested in documenting any of that I'd be happy to help.

[McCamy Taylor]

32. I am so confused, can some one spell out to me what this means?

[IndyPriest]

35. Among other things, it shows that the final PUBLISHED exit polls

which have been used to "demonstrate" an apparent strong late-evening push for the schrub, were fudged. The ACTUAL late-night exit polls -- which were caught and saved by Simon -- show Kerry maintaining his lead.

[McCamy Taylor]

37. That's what I thought but wanted to be sure. So, in other words, it is

like a doctor is being sued, and he alters the patient's records before sending them to the lawyer but doesnt realize that the patient (at the urging of her lawyer) got a copy of them a few weeks before, and so now the real records are there along with the forged records, making it clear 1. what the doctor really did, 2. that the doctor attempted fraud and 3, that the doctor had better settle because he has just committed a major professional no no.


Is there a similar ethic when it comes to the sociologic statistical sciences? If this were an academic paper and the author was caught fudging his facts he would be in major hot water.

Who is this Mitfosky guy? Does he pretend to have an academic reputation? Maybe the academic world should take him on, chatise him for being unprofessional.

[IndyPriest]

41. Very good analogy!

Mitfosky was pretty near the inventor of exit polling. Been doing it for a LONG time, and now has himself in some pretty deep shit for what has come down.

[BeFree]

39. I will try

For most of the day, Kerry had a sizable lead...2.6%

Even after 11,000+ exit polling votes had come in Kerry still held his lead. That was about 7:30

Then, as the last 2,000+ votes came in, Bush overcame that 2.6% lead, and was credited with even more than that.. That was @ 12:30

It is virtually impossible that the last 2,000 votes counted would have been so different in their percentages than the first 11,000 votes.

What we have learned is that the exit pollers began fudging their own numbers in the last 2000+ votes so as to not look so crooked.

All of this is based upon a few assumptions.

First is that exit polling is designed to give an accurate picture of the final complete vote. History has proven, time and again, that exit polling was accurate for many an ellection right up to the year 2002 midterms when suddenly they veered away from the finals.

The second is that the M$M would not have paid $10 million dollars for what they thought might be screwed up exit poll numbers. IOW, they thought what they were buying would be damn near dead on.

What we have been saying is that given the majority of exit poll numbers showing Kerry winning, and the virtual impossibility of just 20% of those numbers reversing that trend, the final numbers were cooked to match the tabulations coming from the states.

What we are saying is that the missing raw exit poll data, once we are able to get our hands on it, will show how the books were cooked.

Hope this helps.

[Lauri]

42. This info conveniently disappeared in Mitofsky's computer malfunction

According to Mitofsky, they don't exist....

"To compound the problem further, a server at Edison/Mitofsky malfunctioned shortly before 11 p.m. The glitch prevented access to any exit poll results until technicians got a backup system operational at 1:33 a.m. yesterday."

http://www.washingtonpost.com/wp-dyn/articles/A23580-2004Nov3.html

[McCamy Taylor]

48. Sounds better than "My dog ate it"

:nopity:


I really hate it when academic types and professionals sell out. I mean, it should not be any worse than anyone else selling their soul, but for some reason it seems worse to me, I dont know why. Maybe it is because society has done so much for them, educated them, given them meaningful rewarding professions rather than the McJobs everyone else has.

[Lauri]

51. Not all Professionals Sell Out - There is Hope

http://www.cpsr.org/issues/vote/evoteproject

"Computer Professionals for Social Responsibility = CPSR is a global organization promoting the responsible use of computer technology. Founded in 1981, CPSR educates policymakers and the public on a wide range of issues. CPSR has incubated numerous projects such as Privaterra, the Public Sphere Project, EPIC (the Electronic Privacy Information Center), the 21st Century Project, the Civil Society Project, and the CFP (Computers, Freedom & Privacy) Conference. Originally founded by U.S. computer scientists, CPSR now has members in over 30 countries on six continents."

http://www.cpsr.org/issues/vote/evoteproject

"CPSR has worked in partnership with Verified Voting Foundation in a coalition of dozens of prominent national organizations, representing a combined membership of over three million Americans, on an Election Incident Reporting Project, to prevent election problems from disenfranchising voters or changing election outcomes inappropriately. The Project brings the experience of voting rights groups together with the technical."

groups.http://www.verifiedvoting.org/

"We advocate the use of voter-verified paper ballots (VVPBs) for all elections in the United States, so voters can inspect individual permanent records of their ballots before they are cast and so meaningful recounts may be conducted. We also insist that electronic voting equipment and software be open to public scrutiny and that random, surprise recounts be conducted on a regular basis to audit election equipment."

[go west young man]

43. You guys kick ass!!!!

I just thought I'd say that its cool to be able to get on these boards and see people working as hard as you all do. I don't post very often but when I see you all doing real investigative research and working together to do what the MSM is too pathetic to do it gets my blood up. You rock DUers. And I for one think the orange hats in time square (wheather coincidence or not) are a great sign!!! Discover! Discover! Discover!

[1democracy]

58. Election night scenario...could timing be a smoking gun?

So Bush&co were going along on election day hoping for a Bush win with the voter suppression of specific Kerry supporters (e.g. blacks, poor) and with the help of voting machines where things such as the default setting for Bush along with other abnormalities were already operating.

(Recall: Exit polls ask voters how they voted as they exited the voting booth. Counts, which continue on much later, reflect the reports of the actual ballot counts).

The exit polls were going to be sticky. If the exit polls were virtually even for Bush and Kerry, there would be no problem, since a small (10%?)but sufficient proportion of the Kerry votes were already destined to be counted for Bush. But if the Kerry votes (determined by exit polls) were too large, then when the actual vote counts came in and Bush won, there was a possibility that the vote count would be questioned instead of the polling method.

Kerry was in the lead all night. It could be a problem. It all hinged on whether the media was prepared to believe there was a huge 'miscount' (we would say fraud) or would they rather believe their own polls were wrong. Faced with a call from Rove, supposedly telling them their exit polls were wrong and his were correct and that Bush was taking Ohio etc, they caved in and corrected their final exit polls to agree with the "actual" counts being reported.

As other DU posters here have suggested, the "last minute" Bush voter rush that was hypothesized actually never happened; it was an explanation for the the excess Bush votes that showed up after the last exit poll had been conducted. This is why the final exit poll data (unaltered/ uncorrected)is so crucial for this analysis. This exit poll info should prove whether or not these voters exist!!!If exit polls at 8 p.m. still showed the Kerry lead, one must conclude either the actual vote tabulation was altered for a Bush win or, as other DU posts have warned me, the exit polling model was flawed. I am not sure how to distinguish between these two. Perhaps it can be shown that these same exit poll data were accurate for some other (non manipulated) races or that the same method employed in another year's election was accurate.

The computer blackout/ power outage sounds suspicious, like an excuse to reformulate/rework the data. It might also be an excuse for the pollsters to ignore the final exit poll data as "lost" or corrupted, (as long as someone hasn't saved it on their computers) and sufficient time for them to import new (whose?)numbers to recalculate the results.

Question; would a decision have been made and transmitted to ??? to institute the plan to more drastically alter the vote late on election day, or would the entire plan have been put in place much earlier? When would the phantom votes have appeared? Were the "phantom votes/over votes (more than 100% of registered voter turnout)activated at the end of the day on Nov 2, corresponding with the last minute Republican rush to the polls?

It seems like what would be a crucial part of the investigation of into the whole Nov 2 scenario, would be to check the timing and pattern of the appearance of the mysterious votes. Did they all happen at once or were they being counted at same rate all day? When did the over votes occur? All at the end? In several bursts? Did this coordinate in all precincts at the same time, or start in one and branch out into others as if connected? What would the pattern look like if it all took place at the level of the tabulators?I believe the timing of the appearance of these votes might be a chance for a real smoking gun...Since voting machines record the time of votes, some type of analysis of these questions should be possible to investigate.

[anamandujano]

84. 2 things

There are threads here (with charts) showing that exit polls were accurate (matched vote tallies) in counties where there was a paper trail. Hopefully someone can help you find the information.

The media has no problem reporting what Rove tells them, no matter how much it goes against any common sense. They are dupes, fools, @sswipes--Rove treats them that way and actually, they deserve to be treated that way. They are paid lackeys. Full stop!.

[1democracy]

85. Reply

Thanks. I'd like to see the graphs.

[anamandujano]

87. edit--Okay, right here

and the thread

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=104x2612492#2612573

from post #9

[Nederland]

44. You don't get it

No EVIDENCE, however, has been advanced to support the reluctant Bush responder hypothesis other than the tautology that the vote count had to be right and therefore the exit polls must have been wrong and this is the only way to explain it.

You've got this completely backward--the burden of proof is on us, not them. From a legal standpoint, if you want to prove fraud you cannot simply assume that the exit polls were right, you have got to prove it. The reason this is true is because in court you will be asked to explain why you believe that a sub-sample of actual voters should be considered more accurate than the full sample (actual results).

Just look at what happened in Ukraine. Yes, exit polls told us right off that there was something fishy going on, and that lead to investigations. Within days, those investigations showed numerous examples of voter suppression and intimidation, in some cases with video evidence. In this case, Conyers has been investigating for weeks now and he has no hard physical evidence to demonstrate fraud. None. The closest we've ever gotten is an account of a Triad employee doing maintenance on a tabulator, but an investigation of that went nowhere.

We need proof, not more mathematical analysis that merely assumes that the exit polls were right and the vote wrong.

[McCamy Taylor]

46. That''s why the Greens and Libertarians are in Federal Court trying

to get an accutal count. And a look at those voter logs too.

[sickinohio]

69. Question - did I read somewhere that Blackwell

locked down all poll books in all counties in Ohio until Jan 6th? Am I right or wrong? Where can I find this documented? Thanks

[BeFree]

49. Get this

The proof needed is in the hands of Mitofsky, et al.

Instead of preaching to us what we ought to do, I would suggest you turn to helping us get the proof. It's there, and they are hiding it from us.

Answer me this, Nederland, why are they hiding the raw numbers from us? And, do you really want to see the proof?

[McCamy Taylor]

50. Everything we need is "proprietary"--the softwear of Diebold machines

the exit polls. It is time that we stopped letting our elections be run like a business.

[BeFree]

54. Secrets all

We don't need no steenkin' secrets! Let the sunshine in.

It's like I told my county election board: "That's the last time this county will use those damn machines"

[davidgmills]

70. What Congress Needs and What the Court Needs are not the same

Any congressman or senator can object to acceptance of the vote of electors based on his own requirements.

But even in a court of law, I really do believe that the best evidence of systemic fraud is mathematical evidence. If that evidence is widely accepted in the scientific community after substantial peer review, then even a court would look at the evidence as being sufficient.

The only real question for a court would be is whether there is sufficient evidence to believe that the purported winner of the election did in fact not win. That could be done mathematically if the evidence were strong enough. You don't necessarily have to prove how the erroneous result occurred. You just have to prove that an erroneous result did in fact occur. Usually you only have to do that by "a preponderance of the evidence," or at most by "clear and convincing evidence." Accepted margins of error far surpass any clear and convincing standard.

[Nederland]

88. Nonsense

Let me show you what would happen to our mathematical "evidence" in a court of law.

Democratic Witness: ...so as you can see, the odds of President Bush winning this election are a gazillion to one.

Republican Defense: Doesn't your analysis assume that the exit poll samples are representative of the voting population as a whole?

Democratic Witness: Yes.

Republican Defense: Can you prove the samples were representative?

Democratic Witness: No. We don't know the demographic make up of the actual voting population.

Republican Defense: Why not?

Democratic Witness: Its not recorded. The only thing that is recorded for the entire voting population is their vote, and in a small set of states, their party affiliation.

Republican Defense: So you not only do you not know what the demographic makeup of the voting population is, you can never know.

Democratic Witness: Correct.

Republican Defense: So its impossible to prove that the exit samples are representative.

Democratic Witness: Ummmm. Yeah.

Judge: Case dismissed.

[davidgmills]

89. I'm sorry but that wouldn't be the testimony.

Sorry double post. See below.

[davidgmills]

91. That would not be the testimony>

The testimony of Freeman or Baiman is all you need.

You assume that it needs to be representative. Who says? You? Who are you? You are nobody. You have no credentials.

Somebody has to say it has to be representative.
Somebody else can say it doesn't.

Who do you beleive? Well whoever makes the best argument should win but judges and lawyers make terrible "judges" of scientific information. They suck. I'm speaking about my colleagues and I know. They don"t have the training. Neither do juries.

If a judge or jury believes Freeman or Baiman over whoever is on the other side you win. That doesn't mean Freeman or Baiman were right; it jsut means to the trier of the case they were more convincing.

[Nederland]

92. You are so wrong

I'm not sure where to start. Nobody says that a sample doesn't need to be representative. Nobody. Not even TruthIsAll would say that. Imagine if you did an exit poll that contained only Republicans. Naturally such a poll would show that Bush won the election by an overwhelming margin. How would you feel about Republicans then claiming that Democrats cheated because the vote count was closer? You'd laugh in their face, right?

Well that's what happens when you think it doesn't matter if a sample is representative. Hello...

[Nederland]

93. More

You assume that it needs to be representative. Who says? You? Who are you? You are nobody. You have no credentials.

Since you do not trust me, perhaps you could look at the following links to educate yourself on how polls work.

http://www.learner.org/channel/courses/learningmath/data/session1/part_d/

In data analysis, we use graphs, tables, and numerical summaries to study the variation present in our data. Often, we want to extend our interpretation to a larger group beyond the particular group studied. Such generalizations are only valid, however, if the data we examine are representative of that larger group. If not, our interpretation may misrepresent the larger group


http://www.mori.com/mrr/2000/c001103.shtml

Is a larger sample always better than a smaller sample? All other things being equal, yes, but all other things are rarely equal. A small but representative sample is far better than a large and unrepresentative sample.

[davidgmills]

95. You obviously have no clue how lawsuits work

What you just cited is hearsay. I don't have to read it. I know it is hearsay.

Now if Joe Blow expert says the group must be representative and Joe Schmo expert says it doesn't have to be, somebody has got to decide who is right and they don't get to go look up your cites.

Or, if Joe Blow expert says the sample was representative and Joe Schmo expert says it was not, somebody has to decide who is right about that.

All I am saying is that judges and juries are very poor "judges" when it comes to assessing the quality of expert testimony. They might decide which expert is right based on who parts his hair the straightest. Many times the suave expert wins the case and the non-suave expert loses the case even when the non-suave expert is right. Happens every day.

[Nederland]

98. Challenge

Find me a person with a Phd in statistics that says that a sample doesn't have to be representative and I'll give you $1000.

Time to put up or shut up.

[davidgmills]

100. You missed the whole point of the argument.

In reading my first post, I see that I did not make myself clear. Let me start over.

You assume that a Democratic expert would have to admit that he could not prove with scientific certainty the sample was not representative and that such an admission would be dispositive of the case. I say that even if he had to admit that, such an admission is nowhere near dispositive of the case.

I read Freeman's article today and he flatly said he thought the sample was representative. Your assumption, the way you have stated it is that he would have to prove to almost all certainty that the sample was representative. He wouldn't need the kind of proof you require. Believe it or not all that is required of a expert is his opinion. Freeman just gave his opinion today.

Expert testimony is a battle of opinions not a battle of scientific certainty. That was the point I was trying to make.

[Nederland]

101. Irrelevant

Post #88 explain why you can never prove that the exit poll sample was representative. It doesn't matter whether or not Freeman believes that the samples were representative--the bottom line is that he can't prove that they were. In a court of law you need proof, not belief, in order to convict.

[davidgmills]

103. You do not understand the burden of proof

First of all, election law suits are civil cases not criminal cases. The burden of proof in a civil case is not beyond a reasonable doubt (roughly a 99% probability). In a civil case the burden of proof is clear and convincing evidence (roughly a 75% probablity) or even a preponderance of the evidence (just over 50%).

At most all you would need to do in a civil case is prove that there was a 75% probability that the sample was representative.

What I am also trying to say is that under the Federal Rules of Civil Procedure and all the states which have essentially adopted it, when the subject matter is beyond the common understanding of a layman, experts are allowed to explain to the judge or jury what they think of the situation. They do this by giving their opinion.

In a personal injury case, a doctor may give his opinion that an automobile wreck caused the injured person to have a broken arm. The doctor who testifies almost never witnesses the accident and he doesn't have personal knowledge of the accident at all. He still gets to give his opinion based on the medical history the patient has given him.

Experts get to give their opinions on many subjects, even on subjects contained in articles and books (normally excluded by the hearsy rule). But ultimately their testimony always comes down to a matter of opinion.

Which is why I say Freeman can look at exit poll data and conclude the samples were representative and conclude in his opinion that they were random or clustered or whatever and testify what his opinion is. A judge or jury is not required to believe his opinion. If they do, his side wins. If they don't his side may lose.

[Nederland]

105. Irrelevant

You just don't get it. Its not that Freeman has a small amount of evidence that the samples were representative--its that he has absolutely none. You mention that a civil case requires 50% evidence. Well, Freeman has nothing. That's 0% and therefore nothing will come of it.

[davidgmills]

106. First Of All You Obviously Know Nothing About Court

I've told you that and you don't listen. So I will explain it one more time.

I agree with you that nothing will come of these cases as they now exist because they are missing one of the two elements of an election case, which are: (1) irregularities or fraud, and (2) proof that the irregularities or fraud could have changed the outcome.

Guess which proof is missing? Unlike 2004 where the election was so close in Florida that proof of (2) was self evident, there is no such proof this time unless you do it with statistics.

How do you get the statistical evidence? With a poll. If the exit poll was not good enough to suit the critics, you do another one and do one that is.

Would Freeman like a better poll? Sure, so would I. Of course today the IPSOS poll shows that Americans now prefer Kerry 48% to 44%. More proof and support that the exit polls were right.

Listen one more time. This is a battle of expert oninion not scientific certainty. Freeman qualifies as an expert. He gets to give his opinion. Some other expert like you can give your opinion he is wrong. A judge or jury decides which one of your opinions they like best. Got it now?

[Nederland]

108. You still don't get it

Expert opinion is fine, but it needs to be based on something. In this case there isn't even data to base an opinion on. Get it?

[IndyPriest]

77. Hold a sec. I don't think we know all that Conyers has found

do we? To assert that he has found no hard evidence - none! - is quite remarkable, given that he's going to hand out his report this week to senators (and to the media, no doubt!). I doubt he'd mount a full court press based on "circumstantial" or "subjective" evidence. But, we'll see soon enough!

[jwmealy]

102. It's a little involved, but Proof is Right There

Hi,

You're right. What is needed is a model of analysis that disproves the hypothesis of the reluctant responder, or proves that even if that hypothesis accounts for some of the "red shift" phenomenon, it doesn't account for the major portion of it. I have performed such an analysis, and it has lately been confirmed by a professional statistician. The probability that the 2004 presidential election was NOT electronically engineered is thousands to one against. The legal proof is now not far away--all it takes is more digging.

www.selftest.net/Analysis.htm

(data sets at www.selftest.net/redshift.htm)

Peace,

jwmealy

[McCamy Taylor]

45. I can guess what Mitofsky has been promised...

Everything is going to be a monopoly in the neo-con Bush world and there will be a place for the Exit Poll. We saw how they used them in the Ukraine to make sure that the election fraud of the candidate we were not supporting was revealed. They can be fudged, if necessary, to reveal fraud that never happened. With the sophisticated system of electronic vote fraud which Rove, Diebold and the others have cooked up they absolutely have to have continuous accurate monitering of the vote nationwide to keep the numbers believable----what if other organizations are doing exit polls too? Thye may even create "vote fraud" by Democratic candidates whom they want to smear (Rove has already done this in other ways such as with the Bush debate tapes) and use exit polls to catch the "culprits".

I am guessing that Mitofsky has been told "either you are for us or you are against us", i.e if he plays ball he will be the official exit pollster of the Bush Dynasty and if he does not some guys from Texas A & M that nobody has ever heard of will get the job and his company will be investigated by the SEC or FCC or IRS.

[euler]

56. When the other side puts Mitofsky on the stand to testify...

...about the 2004 exit poll and the early release raw numbers, Mitofsky will testify that the exit poll raw numbers are unsuited for any type of valid statistical analysis. Indeed, he and those who work for him are already on record saying this very thing several times since the election.

Many of you are now rolling your eyes at my post - "after all", you may reply to my post, "Mitofsky is just a partisan hack with an agenda" or "He is being deliberately untruthful so he can cover up the shortcomings his ability to conduct exit polls" or something similar to that. Amateurs on DU can get away with this vacant way of "defending" their analysis against other DU members, but if won't work in Congress or in a Court of law.

Warren Mitofsky is the preeminent exit poll expert on the planet, and everywhere except right here on DU, he is so recognized. Don't believe me ? Take a look at paragraphs 66-72 of Arnebeck case number 04-2088. here's one quote. You can read the rest for yourself.

<snip>

Credit for INVENTING the exit poll is generally given to Warren Mitofsky a world recognized expert in exit polling...Mitofsky has directed exit polls since 1967 for almost 3000 electoral contests. he has the distinction of conducting the first presidential exit polls in the United States, Russia, Mexico and the Philippines.

</snip>

From here, Arnebeck goes on and on with praise after praise for Warren Mitofsky.

To sum up, Mitofsky is the INVENTOR exit polling and has been conducting exit polls for 38 years. If he says (and he already has) that exit poll raw numbers are useless for anything including fraud analysis, then that's what will be accepted by the Courts, Congress and MSM. We may be able to find experts willing to challenge Mitofsky, but only Mitofsky can say that he conducted the 2004 presidential exit poll. No one knows the numbers like he does.

Dozens of DU members have been fairly persistent in their criticism of any analysis here on DU that relies on the accuracy of exit polls. However, we are always shouted at when we do. Folks, the real world is not like DU. If this goes before Court or Congress, the other side will trot Mitofsky out and he will crush the exit-poll/actual-vote discrepancy fraud theory. If we don't drop the exit poll idiocy, then we will probably start to get that long sought after MSM coverage - but it will sound like mockery.

[1democracy]

59. Humor me as I think this through... timing a smoking gun?

So Bush&co were going along on election day hoping for a Bush win with the voter suppression of specific Kerry supporters (e.g. blacks, poor) and with the help of voting machines where things such as the default setting for Bush along with other abnormalities were already operating.

(Recall: Exit polls ask voters how they voted as they exited the voting booth. Counts, which continue on much later, reflect the reports of the actual ballot counts).

The exit polls were going to be sticky. If the exit polls were virtually even for Bush and Kerry, there would be no problem, since a small (10%?)but sufficient proportion of the Kerry votes were already destined to be counted for Bush. But if the Kerry votes (determined by exit polls) were too large, then when the actual vote counts came in and Bush won, there was a possibility that the vote count would be questioned instead of the polling method.

Kerry was in the lead all night. It could be a problem. It all hinged on whether the media was prepared to believe there was a huge 'miscount' (we would say fraud) or would they rather believe their own polls were wrong. Faced with a call from Rove, supposedly telling them their exit polls were wrong and his were correct and that Bush was taking Ohio etc, they caved in and corrected their final exit polls to agree with the "actual" counts being reported.

As other DU posters here have suggested, the "last minute" Bush voter rush that was hypothesized actually never happened; it was an explanation for the the excess Bush votes that showed up after the last exit poll had been conducted. This is why the final exit poll data (unaltered/ uncorrected)is so crucial for this analysis. This exit poll info should prove whether or not these voters exist!!!If exit polls at 8 p.m. still showed the Kerry lead, one must conclude either the actual vote tabulation was altered for a Bush win or, as other DU posts have warned me, the exit polling model was flawed. I am not sure how to distinguish between these two. Perhaps it can be shown that these same exit poll data were accurate for some other (non manipulated) races or that the same method employed in another year's election was accurate.

The computer blackout/ power outage sounds suspicious, like an excuse to reformulate/rework the data. It might also be an excuse for the pollsters to ignore the final exit poll data as "lost" or corrupted, (as long as someone hasn't saved it on their computers) and sufficient time for them to import new (whose?)numbers to recalculate the results.

Question; would a decision have been made and transmitted to ??? to institute the plan to more drastically alter the vote late on election day, or would the entire plan have been put in place much earlier? When would the phantom votes have appeared? Were the "phantom votes/over votes (more than 100% of registered voter turnout)activated at the end of the day on Nov 2, corresponding with the last minute Republican rush to the polls?

It seems like what would be a crucial part of the investigation of into the whole Nov 2 scenario, would be to check the timing and pattern of the appearance of the mysterious votes. Did they all happen at once or were they being counted at same rate all day? When did the over votes occur? All at the end? In several bursts? Did this coordinate in all precincts at the same time, or start in one and branch out into others as if connected? What would the pattern look like if it all took place at the level of the tabulators?I believe the timing of the appearance of these votes might be a chance for a real smoking gun...Since voting machines record the time of votes, some type of analysis of these questions should be possible to investigate.

TLL

[Bill of Rights]

74. Excellent post!

It all hinged on whether the media was prepared to believe there was a huge 'miscount' (we would say fraud) or would they rather believe their own polls were wrong. Faced with a call from Rove, supposedly telling them their exit polls were wrong and his were correct and that Bush was taking Ohio etc, they caved in and corrected their final exit polls to agree with the "actual" counts being reported.

[understandinglife]

76. You may find these DU archive links of interest

OH vote counts over time - graphs

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=180794

How the popular vote was decided - numbers!

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=123453

I have corresponded with seaclyr and have received very helpful responses.

Peace.

"Who bought the green shoes; Rove or Daddy?"

[1democracy]

82. Graphs of popular vote-- is there more info there?

Thanks for the links. The graphs were excellent. Was the analysis for the 2000 election ever done for comparison, as mentioned in the thread?

Also, what was the likely scenario for the postulated fraud? I'd like to calculate how many votes were lost due to voter suppression. Then do a calculation on the conversion of 10% of Kerry to Bush votes, as postulated by others to figure how many Bush could have gotten this way. This would be a broad background all day, I'd presume... except on the machines that weren't tampered with. Therefore, it would be interesting to just look at the accumulating votes* as they came in on the touch screen machines alone and see what these patterns looked like, unobscured by "honest" counts.(I think AP had direct feeds on all election day data so maybe it could be obtained). Then compare with machines/methods where fraud unlikely.

RE: accumulating votes- is all of this just the summation of final data after polls have closed in each state or would there be reports of partial counts in each state as the day wore on? (If so, this would be a way to closely track shifts in results.)

Would some decision have been made during Nov 2 to generate the vapor votes? How would that look on a graph? What would show it most clearly?

[RevCheesehead]

61. OK, I think I understand your point - that manipulated poll data is not a

crime, because it does not lead to proof of election tampering.

I find this stuff fascinating, because I see the manipulation of the exit polls as the crime done to cover up the real crime. And anyone who watches TV crime shows knows that the suspects are caught in their attempts to cover up their crime.

I see it like this: the manipulation of exit polls was done at a time when the actual vote totals were being manipulated. A convenient "server crash" gave them plenty of time to focus on the polls, while the real crime happened at the tabulators.

Imagine this scenario: a bank robbery occurs while at the same time, a gruesome murder occurs. While you have everyone distracted with the bank robbery, it becomes easier to carry out your diabolical plan to commit homicide.

What I'd like to know, relevant to the time-frame, is WHEN was the shrub family green shoes video shown on the air? I'm pretty sure it was before the variation of the exit polls, or the red shift.

[The Judged]

63. If the polls were changed to match crime induced results, then ....

Mitofsky did not need direct knowledge of the crime in order to become guilty as an accomplice by his act to legitimize the crime, even if he believed that the fudged election results were legitimate.

[anamandujano]

86. it's called COVERUP n/t

[nodictators]

73. Have you reviewed Mitofsky's work?

He did the exit poll for the California gubernatorial recall election of 2003.

First, he prominently showed the percentages for the recall question seemingly accurately. Dead on! Inside the data tables, however, he had numbers that were different by 2 percentage-points. It sure looked to me at that time that he adjusted the headline exit poll numbers to reflect the actual result.

For the rates of uncounted ballots in the replacement governor race he offered differing claims that either 4%, 6%, or 7% of the voters simply chose not to vote for any candidate. Take your pick.

BTW, the actual total rate of uncounted ballots in the replacement governor race was 8.1%. So, his 7% claim implied that only 1.1% of the voters erred in that race, which had 135 candidate choices. Yeah!

[pointsoflight]

75. Mitofsky himself has said that exit polls protect against fraud.

He sells his service to foreign governments all over the globe as an important way to verify elections and protect against fruad. So if he were to now claim that they shouldn't be used for that purpose, it would be quite easy to use his own statements against him.

[davidgmills]

78. Mitofsky May Have Started the Business

But that does not mean he is right.

His methods are only as good as the honesty of the last election. He assumes the last election was honest and because he was off in his projections he tries to correct for the "mistakes" he has made.

Freeman gives good reason to believe that exit polls are far more accurate in determining the intent of the voter than actual results.

For one thing, there is every incentive to lie cheat and steal an election because the stakes are so high. The goal is winning.

In polling, the goal is accuracy. If you are not accurate, you lose the next job.

So when a pollster is off in his projection, he assumes he is wrong, rather than assuming the election was flawed.

I think Mitofsky has been making the same mistake for years and just keeps repeating it. It can be proven that he is wrong in his assumptions.

[euler]

97. As I said, I doesn't matter what your or I think about Mitofsky.

When he speaks about exit polls, everyone listens.

[IndyPriest]

79. euler: a semi-technical question....

What do YOU mean by "raw numbers?" Your case seems to rest on the idea that such "raw" numbers are unqualified to act as markers for fraud. In your interpretation, what are these "raw numbers", what makes them "raw," and what needs to be "added" to them to make them useful?

About Arnebeck's praise of Mitofsky: the point in the suit is precisely to use M's background and expertise to show that the exit polls ARE markers for fraud. Of course, Arnebeck and Mitofsky may have different interpretations of what the USE of exit polls can be. And that is the question that many people want M to answer.

[Patsy Stone]

65. Movieweb: Nov 4th, 2004

"Critics Slam Exit Polls" - Nov 4, 2004

While memories of the 2000 debacle in which the networks embarrassed themselves by making incorrect snap judgments about the election may have contributed to the reluctance of most of the networks to call close states in Bush's favor, an equal factor, some analysts suggested, may have been early exit polls that gave Kerry a significant lead. CNN's conservative commentator Tucker Carlson, who earlier in the evening had commented that the exit polls indicated that Kerry would win, told today's (Thursday) Washington Post that the polls proved to be "totally useless. ... In fact, they may be counterproductive." On Wednesday's Today show, Brokaw commented, "The exit polls, which happily we did not report on air, were distorted." The polls had been conducted by Edison Research and Mitofsky International, who had put together an enhanced polling system, employing more sophisticated computer software and more interviewers in key states than were at work in 2000.

http://movieweb.com/news/news.php?id=5786

[ExpatriateTexan]

81. "Explanations" of Exit Poll Discrepancies

You can forget all the "explanations" that have been put forth for the exit poll discrepancies. The reluctant Bush responder hypothesis, flawed exit poll model, etc. can not explain the accuracy of exit polls in most states and the wide swing in "selected" states.

[super simian]

83. Help for the statistically challenged?

What do "weighted data" and "corrected data" mean? Just wondering what would be the point of exit polls if their findings had to be corrected by vote tabulation? The corelation disintegrates rapidly when the consistency of exit poll methodology is corrected by the free-for-all methodology of vote tabulation in this country. :shrug:

[IndyPriest]

90. "Weighted" and "corrected." Good questions.

There are people around here smarter than I am about this stuff. But from what I have read, "weighted" refers to the fact that exit polls are built to INCLUDE population distributions: age, race, gender, party ID, etc. Supposedly, good exit pollsters know what those distributions are supposed to be based on census data, past voting patterns, etc., and they purposefully aim at getting their data to match those distributions in order to get as accurate a snapshot of the vote patterns as possible. This "weighting" is one of the reasons exit polls have a much smaller margin or error than "survey polls." Some of the questions about this year's exit polling is about whether they were "weighted" correctly.

"Corrected" is a whole different horse. One part of the process of analyzing the exit polls is to compare them to "actual" votes and, as an election day goes on, begin to feed the actual votes into the exit poll numbers to "correct" for any (supposed) difficiencies in the exit polling. (I put actual in "" because, as we know from this election, trying to discover what the "actual" votes were is a nightmare.) But the theory is that if an election is clean, comparing actual votes to exit polls is supposed to "confirm" the accuracy of the voting by "correcting" for any remaining problems that may have arisen because of "incorrect" weighting.

Does that help?

[super simian]

94. Yep!

Helps and confirms my unease with "correcting" these exit polls to confim to the phony vote tabulation. I guess that's why the raw data is a smoking gun. All these statiticians who are losing sleep to figure this all out are my new heroes.

:bounce: :bounce: :bounce:

[formernaderite]

99. Perhaps I'm missing something....

I recall questioning the exit polls election day myself, because they were showing states like Virginia, North Carolina and South Carolina as too close or Kerry leading. I for one, never thought Kerry would win there. One network broke down the vote as 60% women and 40% men, at that point I knew they were skewing the date incorrectly. Mitofsky apparently only fed the networks the raw data, and they crunched it as they saw fit. If these errors were being made across the board, no wonder Ohio and Florida were incorrect. I don't doubt a certain level of fraud and irregularities took place, I am just not seeing the broad conspiracy.

[bleever]

107. Last kick for the night, and for the Knights who say "Nay!"