WE NEED A NATIONAL POLL: WHO DID YOU VOTE FOR?

[TruthIsAll]

Seems obvious to me.
They should be very accurate.
After all, no undecideds.

It's either A or B.
As accurate as a verified paper ballot.
No machine to switch the votes.
No question on who counts the votes.
No faulty punch cards.
No SOS to disenfranchise.

We expect the pollsters will do an honest job.
Is that too much to expect?

But pollsters are experienced.
They can adjust.
They know.

With a full breakdown by:
Gender
Party ID
Age
Educ
etc.

And also some state polls:
FL, OH, NY, PA, TX, IA, NM, NV

Independents polsters like Zogby, ARG - and the usual corporates- Newsweek, ABC, FOX, NBC, CBS are constantly doing polls.

Why don't they do some post-election polls?
Do they dare?

"Who did you vote for"?
Simple.

Choose your sample-size, gentleman.

N-size MOE
1000 3.16%
1500 2.58%
2000 2.24%
2500 2.00%
3000 1.83%

[mirrera]

1. Mitofski should do it FREE

If you had a business that sells a product, and that product turns out to be "faulty"...fix it. I have my doubts that the whole polling business is run the way they tell us it is run, but nevertheless, if they are gonna tell us on the one hand that certain Democrats walked into a booth and were so afraid to vote, they didn't...and then say only Democrats were brave enough to talk to pollsters...I don't think so. Or use polls to bludgeon us on issues, but not hold them accountable for the election. Ridiculous. If Mitoefski EVER was capable of doing an accurate poll, let him do it now! Go door to door... I think it is a great idea!

[glitch]

11. I don't think we want him to do it - we want clean data, not corrupted. nt

[Sick_of_Rethuggery]

2. I was thinking along the same lines...

although I think we should go further and conduct it on a massive scale ourselves, documenting every vote that was intended for Kerry.

Names and contact info, may be even an affidavit.

What if?

[Hugin]

3. Good thinking TIA. n/t

[qwghlmian]

4. You do a poll like that, and I guarantee

you will find out there was actually 90% turnout in 2004 elections.

[Wabbajack]

6. That seems impossible

[qwghlmian]

7. Why does it seem impossible -

I can guarantee you that 90% of the people you will poll will tell you they voted, whether they actully did or not.

[jkd]

49. This could work.

To accurately do this one would need a list of every voter in the country and then independently and randomly choose perhaps 15,000 to interview.

[Pacifist Patriot]

59. And that is available information.

I did not get the impression the OP was talking about interviewing by means of the phone book.

[Wabbajack]

107. Where did all those votes go?

[NationalEnquirer]

8. Right! And people will LIE.

These post election polls don't work either.
Some people change their minds quickly, alot of folks want to say they voted for the winner, so Bush would probably win at this moment.
Now, if we had a way to recount their paper ballots, then you'd have something.

[TruthIsAll]

9. Very few will LIE. Why should they?

Pollsters know how to adjust for that based on other questions.

[NationalEnquirer]

10. They lie because they can.

There is a minority of folks who will want to "go with the winning team", phonies.
Those people can skew this a few percent, well guess what, thats all it would take to throw the election, again.
As for adjusting, isn't the adjusting part what we dont want?
The more I think of it, the more I am against exit polling all together.
We just need an open count.
And then, if needed, we count TWICE, Three times, whatever.

[TruthIsAll]

14. An open count? How about a verified paper ballot? Repugs voted NO on that.

I agree: 100% hand-counted paper ballots.
Not one computer.
That's what is needed.

Lying is a very minor concern.
Liars on both sides would cancel each other.
They could also lie in a pre-election poll.
So that's no excuse.

Let the pollsters do the best they can.

[Karenca]

16. oh, c'mon would YOU lie?

Really, now, I'de rather be naked in public than say I voted for the smirking chimp.
WOULDN'T YOU???

KERRY!
KERRY!!!!
KERRY!!!!!

:dem: :dem: :dem: :dem: :dem:

[NationalEnquirer]

69. I wouldnt lie.

But there are plenty of people who would, who are NOT as committed to their ideals as we are.
Thats the 10% or so that tend to decide the election these days, depending on who was cuter in the debates or something (I know a woman who voted for Clinton/Gore because they were cuter, no lie!).

[davidgmills]

34. Another BS argument

Bush's numbers are so low now, it is likely to be the other way. Lots of Bush voter remorse.

[Pacifist Patriot]

65. I agree. If lying made a decisive impact on the results I would

go with lying the other way. Besides, some of us sporting tinfoil chapeaux are pretty darned sure the margin of victory wasn't nearly as close in either direction. If Kerry took a well-rounded statistically compelling sampling by a decent margin that would be vindication for me if no one else. I'd like to see the poll done and done very very carefully and thoroughly.

[NationalEnquirer]

70. I am saying that either way...

Folks can claim they voted for whomever they wanted to.
Post election polling is meaningless in a close election.
It is useless.
I am for RECOUNTS if neccessary.

[Pacifist Patriot]

58. Not if you poll those who signed in on the poll books.

The names of people who voted is a matter of public record. Important thing is to make sure you start off with a good database. Therefore, don't poll the population randomly, poll verified voters.

[Fly by night]

5. I thought Zogby did do a post-election poll.

I seem to remember some reference to it on his web-site shortly after the election, though I've not seen any reference to it since then. I'll revisit his web-site this morning and see if I can find anything. If I don't find anything, I'll email his company to ask the same question.

[greenmutha]

13. Zogby has the questions in his current poll!

There's a lot of stupid stuff about Valentine's Day in there, but he does ask if you voted, and who you voted for:

http://interactive.zogby.com/pollregistration/registration/

[NationalEnquirer]

72. Why doesn't Zogby say something about the election?

All I heard was him blabbering with his tail between his legs.
I think its clear that he supported Kerry, if ANYONE has some inside info, its Zogby, right?
Why is it that the Blogs and DU are the only ones out there on this?

[ClintCooper2003]

75. Online polling is totally inaccurate. Zogby showed Kerry behind...

in Tennessee by only 4 points the day before the election. Zogby's telephone polls were far more accurate.

[Fly by night]

88. The Tennessee results are not free from question.

As I remember, none of the polls conducted immediately prior to the election in Tennessee predicted Bush's "reported" margin of victory here. Even those that showed Bush with a substantial lead undercut his final "reported" vote total here.

Of course, over 70% of Tennessee's votes were cast on touch-screen machines without a paper trail and there was a concerted effort to depress the Democratic vote here (including Republicans interferring with blue-collar voter registration efforts that I was personally involved with managing, as well as harassment of voters in Democratic precincts going to and at the polls -- something which the EIRS has documented). In addition, many of the counties which still used paper ballots were using tabulators (manufactured by ES$S) for the first time and relying on ES$S techs to manage the tabulators before and during the election. And while it would be easy to recount those ballots, Tennessee law currently prohibits re-counts (even by election commission staff) without a court order.

Bottom line: while I expect that Bush did win Tennessee, I have a great deal of doubt about the authenticity of his "reported" margin of victory here. We have much work to do with our legislature, and one of the first things we will work to do is be allowed to re-count our paper ballots while we still can. They will be held for another 20 months so we have time. We will re-count the votes in my middle Tennessee county as soon as we can, because we think there was somthing mighty fishy about our state House legislative race also.

[ClintCooper2003]

120. The Tennesse results matched the exit polls, though.

[Fly by night]

125. That's why we need to know the TN precincts that were included.

In fact, Tennessee was one of the few states where Kerry's vote exceeded the exit polls slightly. That doesn't surprise me, given the considerable anti-Bush sentiment being expressed among the conservative country people who I helped register to vote, as well as the visible disgust with Bush being expressed in the wealthy neighborhoods of Williamson County, a very Republican county south of Nashville. What did (and does) surprise me is that neither the "reported" vote nor the exit polls in Tennessee came close to any of the pre-election polls.

Given Mitofsky's apparent complicity in the election theft (despite him apparently "getting religion" recently and submitting a report which contained hard evidence he did not need to include which debunked for us his primary theory for the poll/"reported" vote disconnect), I would be interested in the representativeness of the precincts chosen by Mitofsky to poll in Tennessee. Selecting heavily Republican precincts to poll would pad the exit polls as easily as manipulating the 70%+ of our votes that was cast for the first time on touch-screen machines without paper trails in Tennessee.

[davidgmills]

126. In Memphis we had touchscreen as well

Never much thought about it until about six months ago. Seems so odd now that I was so naive.

Bush support here in the eastern suburbs was not nearly as obvious or vocal as it was four years ago. I had begun to believe the pre election polls were right and that it might be close here.

[Carolab]

98. Do you or does anyone have a direct link to the poll itself?

I've been registered for a long time and always got the polls. I just re-registered but they haven't sent the registration link back.

[Laura PourMeADrink]

108. Zogby does have a post election poll -- but you have to be a

paid subscriber. (like you have to be to get his state polls) I don't think this is the same thing as being an interactive poll participant. At least that is the way I understood it.

[yodermon]

12. Only problem:

there is a noticeable trend for people to lie, and side with the "winner" in an exit poll that is done after the results are known.

My question: if this has been measured & quantified in the past, can a new exit poll be weighted/controlled for this phenomenon? Regardless, it will be more fodder for the naysayers, because if Kerry comes out ahead outside the MOE, they can claim this fudge factor.

Do it anyway.

[anaxarchos]

17. It's called "the bandwagon effect" and it is well documented...

TIA is right though. It is possible to ask a series of questions that allow you to adjust the response for "bandwagon effect". The problem is agreement. The issue is not scientific, it is politics. And, as we all know, politics always trumps...

Having said all that, I'm with you.

Do it anyway.

[skids]

21. It was done, shortly after the elections.

I'm sure a good google session would find the relevant pages.

The results were close to the official national vote tally.

Which would mean one of three things:

1) There was no bandwagon effect, even though it is an established phenomenon.

or

2) There was a bandwagon effect, and there was also a cancelling pro-democratic bias in selection by the pollsters.

or

3) The official vote tally was wrong, the sampling in the survey was either spot-on or biased towards republicans, and there was a bandwagon effect.

Unfortunately, the results are wide open to speculation.

[anaxarchos]

31. I believe it was just asked as a question in approval polls...

... after the election. It was commented on too. I've used this citation a couple of times. This is Jamaican journalist, Ken Maxwell who has stalked many Caribbean elections for fraud.

http://www.jamaicaobserver.com/columns/html/20041106t20...

"One crucial statistic made me quite sure that the election was stolen. It is a well-recorded phenomenon that after an election result is known, more people will claim to have voted for the winner than actually did.

After this election, is a remarkable fact that only 51 per cent of the US electorate said they were happy Mr Bush had been elected. The post-election bandwagon effect is well documented.

"Response error tied to over-estimation of voting is one of the oldest and most persistent types of response error to be documented. . reports that such response errors tend to range between 12 and 16 per cent. with the error tending to be larger the closer a survey was done to the election". ( Robert H Prisuta, A post-election Bandwagon Effect 1992 and Stanley Presser: Can Context Changes Reduce Vote Over-reporting?; Public Opinion Quarterly, Wier 1990)

In this case, and as far as I can discover, only in this case does the percentage claiming to have voted for the winner fall below the percentage actually voting for him.

The US press in its cocoon of fantasy, pretends to believe that this result is possible and accurate.
No one can - without his consent - be deprived of his rights. It says so even in Third World constitutions."


I think what TIA is proposing here is even more direct....

[skids]

37. You know what might be helpful...

If any such poll actually has or will happen, and the methods were kept the same as pre-election polls that queried people who had voted early or by absentee, then comparing the results among early voters to prior polls would tell us loads. So if it gets done, people should be asked what day they voted on.

Which, by the way, is something I'm surprised that TIA hasn't latched onto: there were three surveys done before the election which included, in effect, an exit poll of absentee voters.

Estimates of early voter turnout seem to point to around 24% (Source AP Nancy Benac, AP Robert Tanner, Harris Poll, CBS/NYT poll) Of course the ratio varied drastically from state to state, and if claims of the other side are to be believed, in certain states Dems had a better showing in absentee than at the polls... but one would be naive not to wonder whether some of that may have been due to bad election day counting.

Keep in mind that, though the trend has started to reverse and though a reverse behavior can be seen in heavily Democratic localities, conservative seniors still do tend to dominate the early/absentee vote overall:

Absentee/early voters:

Oct 28-30 CBS/NYT: They are a bit older: one-quarter are 65 or over, and eight in ten are above age 45.

Oct 28-30 CBS/NYT: Twenty percent say they have already voted by mail or early voting, and 3 percent more plan to vote by absentee or early voting before Tuesday. Of those who have already voted, 51 percent say they voted for Mr. Bush (ed. 43% for Kerry.)

Oct 20th, Zogby: among those who have already voted, Bush leads 50% to 48%.

Harris, poll Oct 29-31: we estimate that approximately 24 percent of all this year's votes have already been cast. These voters have given an early 50 to 44 percent lead to President Bush over Senator Kerry.

...that's just what I had on file. There may be others.

[anamandujano]

41. Awesome information but can't access the link.

I got the "page cannot be found" message. It's great that you managed to find this.

[anaxarchos]

42. Sorry, Try this one....

http://www.jamaicaobserver.com/columns/html/20041106t200000-0500_68984_obs_a_lobotomy_for_democracy.asp

John Maxwell is a legendary curmudgeon but quite knowledgeable.

[anamandujano]

43. Thanks, it works.

Hope everyone reads it.

[davidgmills]

35. Nobody's on Bush's bandwagon

[Pacifist Patriot]

61. Under ordinary circumstances perhaps, but given

the 44% approval rating in inauguration day I'd worry more about a trend for people to lie and side with the "loser" this time. ;)

[Horse with no Name]

15. In my small backwoods East Texas town

I have encountered MANY people that voted for Kerry.
Surprises me. But where better to pad numbers than in a state that nobody would think to count anyway?

[zimba]

18. Good idea.

As part of the GAO investigation, there should be a poll of this nature done, but with a polling size of at least 2000. For reference if nothing else.

[emcguffie]

19. Yes. Yes. That's a start. N/T

[Mistwell]

20. Problem

Is getting people to take the poll.

If you are upset about the results, you want to vote in this poll to register your discontent. If you are satisfied with the results, you feel less need to vote in this poll.

So, it would be very hard to get a fair sample of people to participate in this poll.

I'm not saying it CANNOT be done, I just think it's harder to do than just saying it will turn out fair.

[PaganPreacher]

22. False premise: A or B

You created a false dichotomy, TIA.

The choices are not "A or B".

The totality of choices also includes:
1. Did not vote at all.
2. Did not vote for a Presidential candidate, but voted downticket.
3. Voted for a third party candidate who was listed on the ballot.
4. Wrote in a candidate.

There are also other factors:
1. People who refuse to answer, because you are asking for personal information (I would not tell your pollster a damn thing, because the details of my life are nobody's business but mine.)
2. People who lie (those who want to skew your results, because they are opposed to your efforts. All it takes is one word from The Wizard of Rove, and 60 million Americans would lie to the pollster, right?)

The only way to determine the percentage of voters accurately is to do a 100% statistical sample of voters in every state, with 0% opt-out, 0% participation by non-2004 voters, and 0% false or mistaken answers (some people have already forgotten who they voted for- if you don't believe me, go down to your local nursing home and ask the residents what they had for lunch yesterday. Those people voted.) Anything less than 100% participation creates the same guessing game that we have had since 11/02/04.

By the way, if your pollster came to my door, I'd chase him off with a shotgun. "Secret balloting" is just that....secret.


Support the right to keep and arm bears.

The Pagan Preacher
I don't turn the other cheek.

[TruthIsAll]

23. I didn't mean A or B in the literal sense. Of course other questions would

I didn't mean A or B in the binary sense. Of course other questions would be asked.

Put your shotgun back.

I was thinking in terms of a TELEPHONE poll.
Remember those?

And factor in those cell-phones users.
You know, the 19-30 year old crowd of newly registered, young, democrats. Lots of young, liberal women.

And no need to poll LIKELY voters, either.
We only need the ones who VOTED.
OF COURSE, THEY HAD TO BE REGISTERED, ALSO.

HEHEHEH...

[Mistwell]

24. Non-response rate issue

And how do you propose to deal with the MASSIVE non-response rate of phone polls? Particularly if you are adding in cell phone users who are otherwise immune from such calls, and have to pay for such calls themselves on many calling plans?

Come on TIA, you've seen the studies of non-response rates to phone polls...you know it doesn't work as well as you are claiming it works.

[PaganPreacher]

27. OK, a phone poll (not to be confused with a "phone pole").....

1. Who is going to pay for this poll?

2. How do you factor in people like me, who never give out real personal information on the phone, and who make it a point to screw with pollsters and telemarketers?

3. What about unlisted phone numbers?

4. How do you confirm that the person to whom the pollster is talking was an actual voter?

5. How do you confirm that a respondent is the same person that he/she claims to be on the phone?

6. How do you discount those people who refuse to give information over the phone, for fear that your pollster could be an identity thief? (note that this is different from #1, above)

7. How do you account for those voters who don't have a telephone, or who share a phone (our nursing home residents, again)?

8. How do you separate "multiple voters in the same household" from "fraudulent registrations?"

9. How do you overcome the Wizard of Rove, telling all of the flying monkeys to lie to any pollster that calls? Your pollster might find that the Constitution Party candidate got 50% of the national vote!


Loki, the Trickster, tormented for his treachery.


The Pagan Preacher
I don't turn the other cheek.

[anamandujano]

40. A poll of those selected from actual sign in sheets on election day

and early voting.

Nobody would want to lie. Those who voted for Kerry would want their vote registered. Those who voted for Bush would want to prove that he really won.

This could be a secret mailing. Mail 3 to 4 times as many as would be needed. Anyone screwing with the US mail (although I've read the repukes did do this with fake registrations), could be nailed for it.

Great thinking Truth is all, it's just a matter of finding the right approach.

I just read that in Iraq, for some reason, election results are not expected for 10 days. Now, why do we have to have our "winner" by midnight or 2am????

edit to add--the democrats should do this with those selected from our sign in sheets, i.e. those dems that actually voted. We can then see how accurately our votes were counted, how much "spoilage" there really was and how many of Kerry's votes "migrated". Of course, when I say democrats, I don't mean those fat cats at the top of the heap that decide everything and are happy with what we now have as a result.

[PaganPreacher]

44. anamandujano, a faulty premise:

"Nobody would want to lie"

Wanna bet?

A post-election poll that showed Peroutka receiving 20% and Badnarek receiving 20%, would destroy any credibility from exit polling. A certain Wizard of Rove can make it happen with the wave of his wand and a well-placed word (just as he co-opted all the Democrats in Ohio's election boards, right?)

Since Bush has already been inaugurated, he won the election. There are no do-overs in Presidential politics. Bush voters have nothing to prove, and absolutely no reason to answer a poll asking them who they voted for- their candidate is in the Winner's Circle until 2008.

I mentioned that I like to screw with telemarketers and pollsters. I never, ever give out personal information to cold-callers. If someone doesn't respect my privacy, I'll do my best to ruin his day.

If your pollster sends me a questionnaire asking for personal information, including "who did I vote for?", I'll tell him that I am a gay crocodile wrestler with ADHD, who voted for Anita Bryant.



Do you know why? Because, secret balloting is secret, and it's nobody's damn business but my own who I voted for.

And, I'm not the only person in America to feel this way.


The Pagan Preacher
Disclaimer: I am not a gay crocodile wrestler with ADHD. I did not vote for Anita Bryant.

[qwghlmian]

45. ROTFL.

points for flair.

[anamandujano]

46. Rove is not a wizard, he's a dickwad with a huge propaganda arm

aka "the liberal press" backing up his every proclamation.

If you insist that the repukes would want to lie in order to destroy any polling result and that Rove himself would put them up to it, that in itself speaks volumes.

A poll of dem voters could determine how many were really counted, although perhaps your hero could convince the dems to lie also.

[fooj]

25. Why not? Sounds like a plan...

[floridadem30]

26. excellent idea

[super simian]

28. OMG!

What a great idea!

[scottxyz]

29. We need to replace the corrupt voting system

We need to create an accurate, fair voting system in its place.

The old system is neither transparent nor believable.

We don't need just another exit poll - we need to DESIGN AND IMPLEMENT A TRUE VOTING SYSTEM.

[pauldp]

30. Hell Yeah! How do we get it done? n/t

[Bouvet_Island]

32. That´s a great idea!

If you get a random enough sample, you are in business.

It would be important that the precincts to sample wouldn´t be sorted based on the election results from this election. Old fashioned randoms numbers, preferrably not drawn by a computer...

It could be combined with a poll on "if the american people want secret computer program voting", I´d be fucking interested in seeing a large survey showing the support for that.

It wouldn´t be impossible to accomplish, Zogby (?) and a large paper that wants to please the people that wrote all those fucking letters.

I think the Texas vote could be an incredibly nice choice of location, for obvious reasons.

I think Bush´s hometown or area in particular should be hit with this type of probe, we could be very lucky. I am sure there are people that would be willing to pay for this consider the possible catch.

Think of it, how would you explain that people are sticking with the looser? Any "opposite" bandwagon effect would be ... interesting... And just like, even if it doesn´t make all people believe it was stolen, you can´t buy that bad PR for Bush. People that "voted for you" saying they didn´t, that´d make him a singular! Beautiful!

[davidgmills]

33. How long have I been advocating this now?

About 2 months.

[Bouvet_Island]

36. Great!

Keep going!

[jkd]

50. Yes, I believe we had a brief exchange about this possibility

[Pacifist Patriot]

62. You, me and many others.

I fail to see why people would rally around the exit polls but resist post-election polling. I don't know about you, but two and a half months later I still remember for whom I voted. I think the dishonesty factor would be the same whether you're talking to someone on Nov 2 or Feb 2. Those who are going to lie are going to lie.

[juajen]

123. I know. I'm hours late with this, but

I suggested this also on at least two or three threads, one of which I started. Post election polling of dems only. This would give us the info we need. Were our votes counted? Of course that would leave out the repuglicans that voted for Kerry, and I think there were a lot. So, maybe just counting the dems is a bad idea. I don't know.

On another thread, I proposed a parallel election in 06. You know, have dem volunteers outside every precinct, asking people to vote again on paper ballots. I believe you could just say you're testing the validity of the voting machines and most people, I think, would be glad to vote again. There are a lot of pugs that now do not trust these machines.

[Carolab]

38. TIA, have you asked Zogby?

He recommended a "blue ribbon panel" at the first Conyers hearing.

I have thought this, myself, many times--why doesn't John Zogby just do a poll of his own?

[greenmutha]

60. Zogby *is* doing it! Check out post#13:

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=304986&mesg_id=305232&page=

[Lisabtrucking]

39. Voted for Kerry/Edwards. At least thats what I thought I did. n/t

[zann725]

47. I think we shold be able to just "re-vote" (duplicate our vote) on-line

via our computers (which have I.P. addresses), and then just like when you order something on-line, we should have a way to check into central data base of "cast votes" and see if our vote was "counted" the way we entered it. If it did NOT register as we voted, then we need to notify (via email---traceable) the discrepancy. We would also print out our vote on-screen after we make it.

I don't see why this couldn't work. If a person does NOT have a computer, then we could "verify" which computer (I.P. address) they plan to use.

[qwghlmian]

48. IP addresses are not unique per individual

for example a couple of thousand AOL users may share one IP address.

[sacxtra]

130. don't forget about proxies n/t

tcpip online voting is even worse than voting across modems using ansi or zmodem.

[sacxtra]

131. don't forget about proxies

tcpip online voting is even worse than voting across modems using ansi or zmodem.

[tritsofme]

51. Post election poll right here for ya TIA

(In the election for president this year, did things come up that kept you from voting, or did you happen to vote?) Did you happen to vote for John Kerry, George W. Bush, or someone else?

Questionnaire: Gallup Poll Social Series: Health and Health Care
Questionnaire Field Date: 11/07/2004-11/10/2004
Questionnaire Sample Size: 1016

Question Mean: N/A
Question Total N: 863


Scale % N
John Kerry 44.64 385
George W. Bush 50.25 433
Someone else 2.18 19
DON'T KNOW 0.41 4
REFUSED 2.52 22

http://brain.gallup.com/documents/questionnaire.aspx?STUDY=P0411042

[yurbud]

121. No data at your link, just asstrix

[tritsofme]

52. Another poll:

(In the election for president this past November 2nd, did things come up that kept you from voting, or did you happen to vote?) Did you happen to vote for John Kerry, George W. Bush, or someone else?

Questionnaire: November Wave 1
Questionnaire Field Date: 11/19/2004-11/21/2004
Questionnaire Sample Size: 1015

Question Mean: N/A
Question Total N: 855


Scale % N
John Kerry 40.57 347
George W. Bush 51.22 438
Someone else 3.28 28
DON'T KNOW 0.27 2
REFUSED 4.67 40

http://brain.gallup.com/documents/questionnaire.aspx?STUDY=P0411044

[TruthIsAll]

56. 800 sample size? Gallup? Sorry, trit, we can do better than that.

Trit, I know you love Gallup. You defended every one of their off-the-wall, heavily Repub weighted LV polls prior to the election.


BTW, why no more replies in that other thread?
You know, the one in which had the Mystery Pollster link where he claims that Kerry led in 11 of 15 final pre-election polls.

Anyway, what is your point on these two polls?
That Kerry really lost?
Or that post-election polls are worthless?

[Pacifist Patriot]

64. Consider me one of the statistically challenged math-phobes yet

I was underwhelmed by those two poll results. A combined 1,800 people out of how many millions of votes? Where did they sample, Georgia?

Gallup is the organization that gave in to the Republicans about the definition of "likely voter" isn't it?

[anaxarchos]

77. They 'invented' it.... n/t

[tritsofme]

81. You asked for a poll, I showed it to you

MoE of 3.3% isn't that bad.

I bet if you looked through the internals of any other polling outfit you would see they asked similar questions, but there nothing particularly newsworthy about it.

I stopped replying to other thread when you started resorting to personal attacks to the point where a post of yours was deleted.

Take the poll for what it is.

[TruthIsAll]

84. Your list of 50 LV's was from a Repub. site.Next time, if you are going to

try to rebut any of my posts, please do your homework first.

You were attacking my integrity by doing so. That was not the first time. I made it a point of saying up front that I used RV polls, if available - and LV if they were not - and gave very valid reasons for doing so.

I used those 18 polls since July and maintained the latest data in my Election Model, of which you are well aware.

The MP agreed that Kerry was leading in the final polls. As soon as that was posted, all you guys disappeared.

If you have a valid argument, that's one thing. But to continuously attempt to demean my choice of 18 polls as "cherry-picking" ignores the logic and rationale in using them and only serves to diminish yours.

[tritsofme]

93. I simply disagree with you that

RV polls were more applicable to this election.

Its my opinion, I base it on the facts as I see them, that's not an attack on you.

I think everyone knows that RCP is a RW site, but that didn't make it a less valuable source prior to the election. It was a great place to see every new poll that got released, both on the national and state level.

I have never used MP to justify any of my arguments.

And I really don't even see the point of this thread in the first place. I'm sure you are well aware that most pollsters asked questions like Gallup did in their post-election surveys, and that there is nothing particularly newsworthy about them.

[TruthIsAll]

104. NOW you admit RCP is a right-wing site. So why did you use it?

Why didn't you spend a little effort to present a fair picture of what the final polls looked like right before the election?

You could have shown BOTH LV and RV polls. But you could not resist the temptation to cherry-pick, could you?

Trit, I'm sorry, I can no longer believe anything you say. Even when you say that LV polls were more applicable in this election, I have to believe you are in denial.

Your posts are almost always the same, defending the whoremedia while denigrating independent pollsters, like Zogby. It's makes one suspicious that you must have an agenda in doing so.

I don't understand how you can constantly defend Gallup's insane LV model, even after it was so far off the mark in 2000 and 2004. I'm not talking about the final pre-election polls, but rather the early ones from Sept., which showed Bush with a 12 point lead. These polls were overweighted for Repubs, as Mitofsky has done in the National Exit Poll - to skew the results for Bush.

Trit, it's an obvious scam. They fooled a lot of people and continue to get away with it. They rely on the ignorance of those who don't know a poll from a pole.

I wonder how long they can continue to get away with it. And you appear to be in lock-step with them.

Good luck. It's only America we are talking about.

[tritsofme]

106. Its not as though

I am suddenly embracing LV polls after the election because many support the official outcome, for months prior to the election I said that LV polls would be more applicable to this election.

There's nothing to admit in that RCP is run by RWers, its well known by anybody that clicks on the link, however that does not stop them from posting a variety of polls onto a list and averaging the most recent ones.

It wouldn't make sense to list both RV polls and LV polls from the same source, I'm sure you could find it if you want, but its of no relavence to me.

I'm definitely not going to reopen Gallup's LV model discussion, and how you cannot overweight a random sample.

I don't see how you can consider Zogby any more independant than Gallup. Zogby was sponsored by Reuters and NBC/WSJ.

And I won't speak to my employment situation to you on this forum either.

Also I have never denigrated Zogby's national numbers, those were legitimate, however I do not trust his online Interactive polls, or those YouGov polls that are conducted online as well.

You still haven't said what exactly was wrong with the post-election polls that I posted.

[ClintCooper2003]

100. Only 855 responses counted from a sample size of 1015. That leaves a

whopping 15.8% unaccounted for. Totally irrelevant nonsense, in my view.

[tritsofme]

101. Probably because they didn't vote.

Some people were lying when they said they voted in this poll as well.

If the poll was to be truly correct than only 60% of the sample should say they voted, but as is shown in this poll, some people will say they voted even if they didn't.

Tell me again how this is any more irrelevant than this entire topic?

[ClintCooper2003]

53. I've been thinking about this for months.... it shall be done...

You will see.

[davidgmills]

54. how?

I'd like to know because I can't get any traction for the idea.

Also when I see these numbers of GAllup up here, that is quite disturbing. If he had come up with those numbers (granted sometimes he did) pre election we would have been the subject of even more ridicule.

I have always thought that an exit poll even now would be accurate and that people would not lie, but how do you explain numbers like Gallup's? If Zogby came up with different numbers, everyone would be saying once again that there is no consensus.

[anaxarchos]

63. The science is straight forward...

... the Gallup polls have 90+ questions, <1000 sample size, 3%+ voting for someone other than D or R (that should say something), all kinds of silly angles and silly results which don't let them play here... but you are still right.

No consensus....

In the end, they would poll the Bush family in Texas to "prove" Bush won... by a hair...

[ClintCooper2003]

74. Which Gallup numbers? I don't see any.

[davidgmills]

91. see post 52

[PaganPreacher]

76. David: here's how:

1. You call a polling service (you pick which one).

2. You tell the polling service what you want (presumably a 100% accurate poll).

3. You sign a contract with them.

4. You sign a check for their services and product.

5. You release the results when you get them.

Hmmmm....

I detect a common denominator..... is it "u"?

U

Nope... maybe it's "ewe"?.....


"youse", for those from New Jersey

Nope- it's you! Make it happen, Counselor!

The Pagan Preacher
I don't turn the other cheek.

[davidgmills]

90. There's this little problem

My wallet is not nearly as thick as my skull.

But, if someone could find out the cost of say a Zogby or Harris or a combination of the two, that would be a start.

Maybe we could pass the collection plate, preacher.

[qwghlmian]

96. Don't ask for "someone" if you want it done

call up Zogby or Harris and ask.

[PaganPreacher]

99. A low-cost DU workaround:

The "Democratic Underground, 50-State Post-Election Poll"

1. Statistical model:I read posts by amateur statisticians on DU every day. Those folks could develop a statistical model that they agree is accurate and predictive (one that has no assumptions). That model must have one variable, and only one: the number of votes for each candidate.

2. Test bed: Some of our DU-bees could take their local telephone book, call the first number and every tenth number after in the White Pages, and report back the results of two questions:

a. Did you vote in the 2004 Presidential election?
b. (to be asked if the answer to question "a" is "yes"): For whom did you vote?

Each "pollster" provides the raw numbers from his/her canvass, and our in-house savants figure the stats, using the agreed-upon statistical model. That gives us our test batch.

3. Run the poll: The same canvassers go back to their phone books, begin at the second name in the book and call every 10th number. Same two questions. Report back the results, and our homegrown mathematicians apply the numbers to each state's totals, giving us results for all states (not just cherry-picked states, as some here want to do.)

After running the poll, we can see if the results reflect actual poll numbers, DU expectations (the dreaded "f" word), or if the whole idea has an unforseen flaw.

The Pagan Preacher
I don't turn the other cheek.

[davidgmills]

116. not bad

are we credible?

[PaganPreacher]

122. You're asking the wrong question, Counselor.

The polling technique, raw numbers, and statistical models do not depend on the credibility of the group, at least not beyond the "are their numbers legit?" point.

We are not preparing evidence for trial; this is a preliminary check-poll, with a potential for submission to the Democratic Party for use in targeting election reform.

The operative question, then, is: "is the process valid, and are the results credible?"

The primary purpose of conducting the DU Bargain Basement Post Election Poll would be: To determine if measurable variance between poll data and official election results can be identified (with potential for the "F word," or for other skews.)

How do we determine that? After we conduct the poll, and the savants input the data into their statistical model, we would present the data to some of our university-based DU-bees for review (these people must be independent of our savants who developed the statistical model and input the data.)

We're going to need some volunteers; people who are ready to move past bitching on the 2004 election forum, and to accomplish something.

Think we can get past the monomania and ego trips, and find some of that?

The Pagan Preacher
I don't turn the other cheek.

[davidgmills]

124. but we might lie about the answers people give us

We are liberals and progressives after all. Liars all.

[count_alucard]

55. Kerry would concede again, faster even

The man was the wrong choice and I feel sorry for those who think he did 'his best'.

[Pacifist Patriot]

66. Thank you for not letting another thread go by without a

Kerry slam. I was beginning to worry we'd see this conversation go by without the obligatory speed to concession whine.

[NationalEnquirer]

71. Bullshit.

Whats the guy supposed to do, whine and whine that he really, really, REALLY won?
Although I wish he would be more for Election Reform, I mean, he HAS to know what went on, or have a gut feeling about it.
But anyways, conceding was the right thing to do, its after that that I have a problem with him.
As a candidate, I think he was OK, the only Democrat who would have won in my opinion is Bill Clinton..

[count_alucard]

102. Karl Rove should hire you

[New Earth]

73. why are ALL your posts about Kerry?

when the thread isn't even ABOUT Kerry? If you think he is a fraud, that's great, but do you have to state that in every single one of your posts?

[count_alucard]

103. and do you have to defend this impostor in every single one of your posts?

and why isn't this thread about Kerry? aren't we talking about a poll on who Americans voted for?

It's my opinion that John Kerry was the wrong choice, he knew that himself, and it's my right to express that view.

on edit: and weren't you one of those swearing Kerry had a COVERT PLAN? Jan 20 came and went, where's his stinking covert plan now?

[anaxarchos]

57. So.....

...for all the reasons touched on in this thread and for all the talk about "exit polls are not designed" for this and that, and for all the other talk, the NEP exit poll data is by sample, timing, MOE, methodology, etc. more accurate than any poll that could be designed now.

The exit poll data shows that Kerry won.

That conclusion is supported by pre-election polls and the lack of a post election "bounce".

Any poll done now would use the data above to test its accuracy (i.e.
demographics, etc.).

But it still won't be "accepted" as a "smoking gun".

... and that lack of acceptance will still be politically, not scientifically, motivated.

Do another poll... sure, why not? It's just another brick in the wall. Who knows, maybe it could be used to debunk some other "red shift" theories.

But, no illusions....

[Pacifist Patriot]

67. I agree. Realistic impact would be virtually nil, but...

boy howdy if it showed a clear win for Kerry I'd feel vindicated. :)

[anaxarchos]

85. I don't want to be misunderstood on this...

I am not the least bit pessimistic. I just don't believe that there is any "high court" anywhere that will fairly weigh our evidence and eliminate the need for us to fight for Democracy. I guess it's not the 'smoking gun' but the 'magic bullet' that I object to.

On the other hand, can we win? Hell yes, we can win. And it's about f**king time we put up a fight anyway.

But there are no shortcuts....

[Pacifist Patriot]

89. That's how I saw it and why I agreed.

[Bouvet_Island]

78. I´d just like to state that I disagree,

we had a full and thorough discussion on the problems with the Exit polls, in http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x302159">this thread. The important point is that the exit polls are less accurate, the larger the fraud. And that since we cannot size up the fraud, we can´t judge that accuracy, they might for example mask a much larger fraud as they stand now. Or they could by incredible luck be accurate to the last decimal, the point is that we can´t reverse engineer the accuracy from what we currently have. The Exit polls might be useful for us, but not in convincing scientists and the like, this would be superior in that respect if it would show that "unexpected" result.

As outlined by TIA, would not have it´s accuracy affected by fraud. You don´t need weighting if your sample is close to perfectly random.

It doesn´t have to be very accurate to "work", as I said even a minor kerry victory in a Bush area would be great PR, at the same time working for the people that believe and the ones that don´t believe there was/are/could be fraud. If we get access to the raw data of the poll, any "anomalies" in it could be good leads for us.

I think the pessimism in this thread is uncalled for, if you need a smoking gun to get work done, life will be disappointing.

I disagree slightly about the acceptance motivation, we have to Fight for that acceptance, but there are scientific rules of how to do it. We are presenting an anomaly, something that doesn´t fit the current "optimistic" idea of politics in the population. When we have changed how the population feels about it, the acceptance will come.

What we have to our advantage is that people are getting more sceptical about computer security every day. It also seems like the Opponents propaganga apparatus are in for a defeat with the Iraq war. this will make them more receptive to our case, if Bush´s feel good factor breaks, we score.

I think the proportion needed is at about 50% of the population believing there was/could very suspiciosly easily have been Fraud, we then "win" or at least get a real, open fight.

I am sceptical against Gallup, I believe they were more or less accused of tilting their sample in something as conservative as the IHT, that they did´t include cell phone interviews on purpose.

[Mistwell]

80. I disagree also

"The exit poll data shows that Kerry won."

No, really, it does not. And the more that myth is repeated, the more foolish the whole anti-fraud cause looks, because it is so easy to refute.

The final exit poll shows that Bush won. The pre-final exit polls, which were NOT FULLY WEIGHTED, show Kerry won. However that same exact type of pre-weighted exit poll showed EVERY DEMOCRACT IN ALL PRIOR YEARS AS THE WINNER, until they were weighted. That includes total losers like Dukakis. There isn't any question, none at all, that pre-fully-weighted exit polls ALWAYS (I repeat - ALWAYS) favor Democrats, often by more than the margin of error than even this last election.

The exit poll data does not "show" that Kerry won. It shows jack shit is what it shows, and every single actual expert in the field, including the guy who invented and conducted the exit poll (who himself is both a liberal and a Kerry supporter), along with Kerry himself, and Clinton himself, and all the rest, ALL agree that the pre-fully-weighted exit polls are not an accurate judge of who won.

I'm so damn tired of hearing this myth repeated like it is fact that it makes me sick.

I agree there was fraud. I think MAYBE Kerry won but was denied victory because of fraud (though that is looking less likely to me these days). However, I totally and uneqivically disagree with the notion that you can use these exit polls to prove any of that. They are so friggen useless it's laughable, and anyone who continues to pump that false meme that they are in any way reliable for fraud detection is decieving people (perhaps themselves as well).

[RaulVB]

82. Your statement is a very, very shameful LIE (n/t)

[anaxarchos]

83. I suppose we could go through all this again...

and we could take apart Simon, or Freeman, or TIA. But nah...

The exit polls are at variance with the election results. They are not at variance in a random way. They are at variance in a very specific way which favors Bush. No explanation for that variance has been given; no hypothesis ("shifts", clustering, non-response, WPE, statistical anomalies, methodology, explanation after "all" the data has been released, etc.) which would explain the variance has stood up. The only thing left is that "they must be wrong" because they are at variance with the vote count, or precisely what we started with.

If the vote count was not suspect, that might be a starting point but it IS suspect. There is intent, motive, means, opportunity, and history (2000, to name just one year), to suggest it was wrong.

That calls for supporting data. But the supporting data is all weighted towards the exit polls. Voting patterns analysis supports the exit polls. Pre-election polling supports the exit polls. The lack of a post election "bounce" supports the exit polls. And, most importantly, the inability FOR ANYONE to reconstruct a "Bush victory" from the admitted demographics of this election is decisive. The fact that none have seriously tried is the cherry on top.

So, until someone "refutes the data", Kerry won.

The objections to that are sheer sophistry, not science. Casting "doubt", poking at ghosts, raising "design" issues, wondering what you may say to a pollster,... all of that is entertaining but it has no standing. None of it explains how the exit poll data skewed in the way that it did. To think otherwise is creation science - just a conclusion looking for facts.

So... despite the fact that you are "so damn tired of hearing this myth repeated like it is fact that it makes me sick", let me make you sick again (though I have no stake in it). There is no reason, so far articulated, to believe that the exit polls were wrong and that the vote count was correct. To believe that is the real "myth" (a popular story without proof) although it is a currently dominant myth I grant you.

My point also still stands that the problem is fundamentally not scientific, it is political and ideological for which your objection merely adds evidence.

[davidgmills]

94. Amen

[Bouvet_Island]

110. All this have been answered for,

In the thread I pointed to in my other post.

There is a complete and specific "refutes the data", as well as your "namecalling" flame argument in rich supply.

The "refutes" doesn´t exclude fraud, or that the exit polls indicate it, it is a simple matter of applying the consequence of your conclusion on what we do know about the methods.

I´ll present you the formula:

true polling result * (true election turnout and results + X fraudulent turnout and results) = unknown X and truth mix.

"No explanation for that variance has been given", that isn´t truthful.

Conyers have accounted for more votes than the victory margin in Ohio. New Mexico and several other states were blue. Your conclusion is correct IMO, but not how you arrive at it.

[Bouvet_Island]

86. I of course disagree with this too.

I read your sources for your posts about the bias earlier, the writer although credible stated he didn´t believe there was fraud, in a way that I found a bit funny. He also stated that he didn´t know how the current exit poll works, and somehow implied he couldn´t be sure his numbers applied. It is an interesting fact, but in this case I believe the connection isn´t there and it doesn´t fault the argument of the people that believe it did show a Kerry victory.

I have a different conclusion than you, I believe that fraud could be hidden in the current poll because of the way fraud affects the weighting. Thus, I believe the Exit polls support neither theory, it could *indicate* a Kerry victory but it would be a hard call to make.

My idea of the proportion of fraud have not decreased since christmas, on the contrary I think the events in New Mexico leaves little question there were machine fraud there and that opens a particular window.

If Bush cheated, Kerry was the winner, we don´t need to prove fraud up to the margin of victory to make that call. It doesn´t need to be direct manipulation of ballots to be either fraud or illegal, do you disagree with the conclusions in Conyers report? The situation could have been much worse in my opinion.

I think TIAs Idea in this thread is good, if you are supportive state that as I believe your post seems rather unsupportive as it stands.

* I don´t exclude democrat fraud either, but I have read enough to make a fair assessment of proportion and intent, it isn´t plausible to me that it were of any size if there were.

[anaxarchos]

87. You do not disgaree...

... I have no idea what you are talking about but the only thing you don't do is disagree.

[Bouvet_Island]

111. Eh,

is "disgaree" code for something? I assume typo, I just would like to say that I counter and accuse you of not agreeing. It is very clear, you are not agreeing.

[davidgmills]

92. So re exit poll

And your need to have weighting is history. Clustering is history. Bad design, history. Underage pollsters is history. Pick yuour problem and eliminate it.

Design an exit poll (as near perfect as you can get it)!

Then lets do it.

Then maybe you can prove to the rest of us just how bad these clustered poorly designed exit polls were.

I notice you say nothing about the topic here.

Which is how about another exit poll.

That is the topic.

Kindly stay on it.

[Pacifist Patriot]

68. If anyone wants to do the design and management of the project....

count me in as a volunteer for implementation.

[RaulVB]

79. Anybody thinking on this seriously?

I'd volunteer NO PROBLEM.

[davidgmills]

95. We are the we we have been waiting for

Its up to us or nobody so count me in.

[qwghlmian]

97. Ok - so if you want to do it

why not call up Zogby, say you're interested in doing something like this, and ask for a quote. Once you get that you will at least have an idea how much fundraising you need to do it.

[davidgmills]

117. Is Zogby the answer

I wonder now. I used to think so but I'm beginning to think we need some new blood, someone who wants to make a name for himself, but of course, he has to have credibility.

[googly]

105. Waste of time, has no constitutional validity.

Worse yet, we will be called sore losers.

[Bouvet_Island]

112. False, true, false.

see post 32.

[lynne]

115. Have to agree -

- as there are too many variables to skew results to be effective.

Way too labor intensive. How to verify if the person polled actually voted? And, if they did, there is no way to determine if their answer is accurate.

Who would lie? My mother, for one. Say's that her politics is nobody's business but hers. There's probably a whole generation of oldsters that feel similar.

Then there would be a Post-Post Election Poll to determine why the Post-Election Poll didn't answer any questions . . .

[Kashka-Kat]

118. oops posted in wrong spot try again

[lonestarnot]

109. We already know they cheated so why do we need further confirmation?

:mad: :smoke:

[Bouvet_Island]

113. To communicate this to the rest of the population.

[lonestarnot]

114. I'm for that.

:*

[Kashka-Kat]

119. Huh? Lacks credibility

If they blew off the exit polls that were done on Nov. 2, what makes you think they would they be any more inclined to believe a poll done 3 months later?

Better to help with the various recount america projects, I think.

[RockStar]

128. I suppose you have credible evidence

[Kashka-Kat]

129. evidence of what?

Sorry dont understand what you mean

I was just looking at it from purely pragmatic point of view, in terms of where my $$$$ and/or time & energy might yield the most results. Sure its an interesting question, but would it do anything to convince any of the skeptics that election fraud took place on November 2, 2004. No,I don't think so, considering they are already discounting exit polls which already took place. In effect you are asking for another "exit poll", only one which is done 3 months after the fact and hence even more prone to be discounted.

Is there some kind of pay off or expected outcome that I'm not seeing here, that would make it worth doing?

[sharman]

127. Better yet

Do a 100% poll of a few suspicious precincts.