Check. Mate.

[TruthIsAll]

Which is true?

1) The Reluctant Bush Responder theory
or
2) The Final National Exit Poll, which states that respondents who voted for Bush outnumbered those who voted for Gore by 43-37%.

If (1) is true, then the 43-37% Bush margin which matches the recorded vote must be wrong.
And the Preliminary National Exit poll of 13,047 was right after all.
Which means that Kerry won by 51-48%.

If (2) is true, the Reluctant Bush Responder theory is false.
So election fraud is the only possible explanation for the exit poll discrepancies.
Which means that the Preliminary National Exit Poll was correct.
Which means that Kerry won.

The two choices are obviously in conflict.
Both statements cannot be true.
If one is true, the other must be false.
But BOTH must be true if you believe Bush won.
Logically, we must conclude that both explanations are FALSE.
Which means that Kerry won.

To use the Chess analogy:
Regardless of whether you believe (1) or (2), you are in Check.
And I'm not a chess-player.

[alittlelark]

1. I play occasionally....

you are correct.

The Logic courses I took in school tell me the same.

[Wilms]

2. What's notable about these logical deductions...

is how they also concur with the statistics. And my nose.

All is one (except for the squater in the WH)

[rzemanfl]

3. I am just waking up and trying to think how a Republican would

try to spin this. Let's see, people who voted for Gore in 2000 were...no, that doesn't work. People who voted for Bush in 2000 were...no, that doesn't work. Wait a minute, Gore won the popular vote in 2000, so how could Bush voters outnumber Gore voters unless the poll included more Bush voters than Gore voters, but then wouldn't the poll be skewed towards Republicans??? Ah Ha! Jesus did it! Why do you hate freedom? Whew! That's a relief!

[Mistwell]

8. The same as this Democrat spins this

I'm a Democrat, and I will tell you the same thing probably that Republicans would tell you. The truth is that the early exit polls were always, EVERY SINGLE TIME SINCE THEY WERE INVENTED IN 1988, wrong and biased in a manner that ALWAYS shows more Democrats than actual Democrat voters.

And, they are wrong for a variety of reasons studied by the major universities, much of which has little to do with reluctant responders but has a lot to do with WHERE they exit polls take place, and WHO is conducting the exit poll.

And the authors of the polls don't leave them as EARLY unweighted polls because they damn well know this, and they also know THEY ARE NOT POLLING NEARLY ENOUGH PEOPLE TO MAKE THE POLL ACCURATE ENOUGH TO DETERMINE WHO WON.

So the authors of the polls do two things to make sure nobody makes the blatant and obvious mistake that TIA seems to make every day here. First, they weight the poll and give you a final poll that is more accurate than the earlier polls they have. Second, they tell you that even that final poll is not accurate enough to use as a predictor of fraud or of who won the election, BUT IS ONLY ACCURATE TO DETERMINE THE DEMOGRAPHIC DATA OF VOTERS (Race, Gender, Age, Geographic location).

Both of these checks placed in the polls by the authors are totally ignored by TIA. He just plain tosses them out. Nevermind that they are critical to any possible analysis of exit polls, never mind that every source that puts those polls out, and who consumes those polls (like the media) knows that these factors play an important roll. No no, TIA thinks he can ignore those two factors, and pretend that early exit polls are both accurate and can be used with their MOE to determine who actually won the vote.

[TruthIsAll]

9. You totally avoid the issue. Which is it: 43/37 or Reluct. Bush Resp?

I suppose you also reject uscontvotes.org rebuttal of the Reluctant Bush Responder?

So Mistwell, I ask you: which do you believe?

1) The 43%/37$ Bush/Gore voter split from the FINAL exit POll?
Or
2) The Reluctant Bush Responder Theory?

Do you believe both are true?

A direct answer would be welcome.

[Mistwell]

10. You, asking someone ELSE for a direct answer?

Hahahahaha! Best joke I've heard all month! Classic!

Tell me TIA, given this is about the 23rd time I've asked you this direct question which you will most assuredly avoid once again, do you ever plan on answering this question:

Which statement is true about why all prior early exit polls since their invention in 1988 showing the Democrats getting more votes than the actual vote reported (sometimes by as much as 8% more than the actual vote):

1) A massive secret Republican conspiracy spanning the entire nation involving thousands of people intent on defrauding the electorate, all at great risk and to no avail in every year except this one;
2) The early exit polls are inaccurate, which would disprove all your hard work over the last six months.

Which is it TIA? Why are you so afraid of answering the question?

As for the answer to your question, you know quite well I already answered it. I will repeat it for you, in case you missed it the first 4 times. It is not just the "reluctant Bush responder" theory, but it is that combined with the locations chosen for the exit polls, the people chosen to take the exit poll answers, and the size of the exit poll survey that combine to get more Democratic responders than Republican responders. All exit poll authors admit this, and major Universities did studies on the subject and came to that same conclusion. The exit pollers build it into their theory so they can refine the final results and future exit polls as best they can. They all will also tell you that there is simply no way that the exit polls taken in the US are large enough to test for fraud or even accurately predict election results, and they set their margin of error for DEMOGRAPHICS DATA ONLY (Race, Gender, Geographic Region, etc...and NOT how people vote). They need a LOT more money to hire a sufficient number of exit pollsters to conduct an exit poll to test for the thing you are trying to test for. You friggen know this. You know you are abusing the data for your own egotisitcal ends of calling attention to yourself here at DU. You damn well know you've got major flaws in your basic assumptions about exit polling data, which you bury your head in the sand about any time someone calls you out for it...often by calling the people who hold you accountable Republicans or Freepers or worse names.

[TruthIsAll]

11. Mistwell, you are obviously at a loss to answer the question...

"As for the answer to your question, you know quite well I already answered it. I will repeat it for you, in case you missed it the first 4 times. It is not just the "reluctant Bush responder" theory, but it is that combined with the locations chosen for the exit polls, the people chosen to take the exit poll answers, and the size of the exit poll survey that combine to get more Democratic responders than Republican responders. All exit poll authors admit this, and major Universities did studies on the subject and came to that same conclusion. The exit pollsters build it into their theory so they can refine the final results and future exit polls as best they can. They all will also tell you that there is simply no way that the exit polls taken in the US are large enough to test for fraud or even accurately predict election results, and they set their margin of error for DEMOGRAPHICS DATA ONLY (Race, Gender, Geographic Region, etc...and NOT how people vote). They need a LOT more money to hire a sufficient number of exit pollsters to conduct an exit poll to test for the thing you are trying to test for. You friggen know this. You know you are abusing the data for your own egotisitcal ends of calling attention to yourself here at DU. You damn well know you've got major flaws in your basic assumptions about exit polling data, which you bury your head in the sand about any time someone calls you out for it...often by calling the people who hold you accountable Republicans or Freepers or worse names".

Are you finished, yet? You say "They need a lot more money to do a proper exit poll". Are you saying this with a straight face?

And you claim that I am the one abusing the data, for which Mitofsky claims a 1% MOE?

You contradict yourself. Yes, the Dem percntages are always overstated in the polls. For a variety of reasons.

Yes, natural spoilage is a major factor in the discrepancy. But there is also the question of fraud. And even you agree that fraud has occurred. But you won't accept the fact that the early, uncontaminated exit polls catch it. Or that the final exit poll was manipulated with weights adjusted to match the a corrupted vote count.

You are in total denial. Your shrillness indicates that deep down you know you have lost the argument. There is no longer anyplace for you to squirm. The more you protest, the worse it gets.

The facts are clear to the vast majority of DUers, yet you and a few die-hards continue to ignore them.

What is it? Pride? What would it take for you to see the light and accept the truth: this election was stolen every which way from Sunday.

You don't have to listen to me. Just listen to the professionals at UScountvotes.org - those renowned mathematics and statistical professors who have come right out in support of Freeman et al.

So, Mistwell, once again, answer the question:
Do you believe the Final Exit Poll with it's bogus, contradictory weights?
Or do you believe the Preliminary Poll?
Which one is it?

You are in check.
Time to lost your King.

You can't use the Bush responder excuse.
You can't use claim that the Final exit poll is legit.
You have run out of excuses.
What are you left with?

It's YOUR egotism, not mine, which is at question.
I've been consistent from Day One.

The Final pre-election polls were right.
The initial exit polls were right.
The vote counts were wrong.
Kerry won.

I am in agreement the professors at UScountvotes.org.
You are in sync with the MSM.

You seem to prefer CNN and FOX and MSNBC and BushCo.
Your ranting has gotten personal.

I'm still the same TIA who you once said was respected at DU.
I haven't changed one iota.

The vast majority of DUers believe my work is credible, even more so today than in the months prior to the election.

I will no longer waste energy to try and convince you. If you don't get it by now, you never will.

[Mistwell]

12. Answers

Which statement is true about why all prior early exit polls since their invention in 1988 showing the Democrats getting more votes than the actual vote reported (sometimes by as much as 8% more than the actual vote):

1) A massive secret Republican conspiracy spanning the entire nation involving thousands of people intent on defrauding the electorate, all at great risk and to no avail in every year except this one;
2) The early exit polls are inaccurate, which would disprove all your hard work over the last six months.

Which is it TIA? Why are you so afraid of answering the question?

Why did you avoid that question again, TIA? What do you have to hide?

"You say "They need a lot more money to do a proper exit poll". Are you saying this with a straight face?"

Not only a straight face, but with the backing of pretty much every expert on the subject on the planet. Other countries pay for an exit poll to specifically catch fraud, and they pay a LOT more so that the pollsters can have a LOT more people on duty to ask people the questions at the polling stations. To really test for fraud, you need people in the rural areas as well as the urban ones.

And, BEFORE the exit poll mess all started, I remember some posts BY YOU that agreed with this, that the exit polls needed more people on them, so at this point you want to explain this complete reversal of yours?

"And you claim that I am the one abusing the data, for which Mitofsky claims a 1% MOE?"

Mitofsky specifically claims a 1% MOE for the demographics conclusions within the poll only. He specifically says that the "how people voted" data is NOT covered by the 1% MOE. You friggen know that too!

"You contradict yourself. Yes, the Dem percntages are always overstated in the polls. For a variety of reasons. Yes, natural spoilage is a major factor in the discrepancy. But there is also the question of fraud. And even you agree that fraud has occurred. But you won't accept the fact that the early, uncontaminated exit polls catch it. Or that the final exit poll was manipulated with weights adjusted to match the a corrupted vote count."

It is not a contradiction to agree that you think there was fraud, but to also think the exit polls are so flawed as to be incapable of detecting that fraud. I don't know how much fraud there was, and whether it was enough to tip the scale to one side or the other. I do know that the exit polls, as conducted in this nation, are far more flawed than the actual vote count itself.

"You are in total denial. Your shrillness indicates that deep down you know you have lost the argument. There is no longer anyplace for you to squirm. The more you protest, the worse it gets."

That is incredibly funny. You're the one who dodges the question at every opportunity (and did it again, for the 24th time, this post). You're the one so blatantly in denial, that as the number of people who disagree with you increases, the more entrenched you become on this topic.

"The facts are clear to the vast majority of DUers, yet you and a few die-hards continue to ignore them. What is it? Pride? What would it take for you to see the light and accept the truth: this election was stolen every which way from Sunday."

Most people in this particular forum of this particular board agree there was fraud (as do I). It looks, however, like the overwhelming majority of people TODAY no longer agree with you about the exit polls. Your positive responses are WAY down, as are your total number of responses. Other fraud threads are not encountering that level of a problem, showing it's not a systematic thing with the fraud issue in general, but just your own position on the issue.

"You don't have to listen to me. Just listen to the professionals at UScountvotes.org - those renowned mathematics and statistical professors who have come right out in support of Freeman et al."

First, they are not exit poll experts, they are partisan statisticians. Second, they have ignored numerous and detailed responses to their work showing massive gaps and outright inaccuracies in their work. As a statistician yourself, you know damn well that you can find a dozen people to agree with any position you take...that doesn't make it the right position.

"So, Mistwell, once again, answer the question:
Do you believe the Final Exit Poll with it's bogus, contradictory weights?
Or do you believe the Preliminary Poll?
Which one is it?"

The answer is I DO NOT BELIEVE EITHER ONE. I've made that so clear a child could figure it out. Exit polls, as conducted in the US to date, are not up to the task of predicting an election outcome. Hence, their margin or error is set for demographics data, and not "how people vote" data. You want a good exit poll that can actually track how people vote in a manner that is accurate to predict an election outcome, then you and everyone else in the nation is going to have to agree to pay (a LOT) for that exit poll.

"It's YOUR egotism, not mine, which is at question.
I've been consistent from Day One."

Yes, you have. Consistency, as they say, is the hobgoblin of little minds. You have consistently said that the exit polls can be used to detect fraud in the actual vote, knowing from day one that they cannot and that the MOE is specifically stated as not being good for anything other than the demographic data of race, gender, and geographic region. Talk about ego!

"The vast majority of DUers believe my work is credible, even more so today than in the months prior to the election."

I disagree, as stated above.

Which statement is true about why all prior early exit polls since their invention in 1988 showing the Democrats getting more votes than the actual vote reported (sometimes by as much as 8% more than the actual vote):

1) A massive secret Republican conspiracy spanning the entire nation involving thousands of people intent on defrauding the electorate, all at great risk and to no avail in every year except this one;
2) The early exit polls are inaccurate, which would disprove all your hard work over the last six months.

Which is it TIA? Why are you so afraid of answering the question?

[TruthIsAll]

13. You are wrong when you claim 8% exit poll deviations. It's 3.0%.

ARE WE TO BELIEVE THAT THE BUSH 51-48% WIN IS JUST A
"SLIGHT SHIFT" FROM THE KERRY 51-48% WIN IN THE
PRELIMINARY NATIONAL EXIT POLL?

Preliminary Exit Poll        Final Vote	          Deviation	
	Dem	Rep		Dem	Rep		Dem	Rep
1988	50.3	49.7		46	54		-4.3	4.3
1992	46	33.2		43	38		-3	4.8
1996	52.2	37.5		49	41		-3.2	3.5
2000	48.5	46.2		48.4	47.9		-0.1	1.7
2004	50.74	47.93		48.28	50.73		-2.46	2.8
								
Average	49.55	42.91		46.94	46.33		-2.61	3.42

The Preliminary exit poll is weighted to match the recorded,
official vote tallies in the Final published poll. 

Since the final "weighted" 1996 and 2000 exit polls
had Party_ID weights of 39 Dem/35 Rep /27 Independent, then if
the “preliminary” 2004 Party_ID weighting was 38/35/27...WHAT
WAS THE EQUIVALENT PRELIMINARY PARTY-ID WEIGHTING MIX IN 1996
AND 2000? COULD IT HAVE BEEN SOMETHING LIKE 41/33/26?

The "preliminary" 2004 National Exit Poll is a
randomly selected 13,047 sample with a 1.0% Margin of Error.

The weightings appear very reasonable:
Party-ID is 38 Dem/35 Rep/27 Independent.
Kerry is the 51-48% winner based on all category weightings
used to calculate the vote. But this 13,047
"preliminary" poll vanished into the ether, never
to be heard from - except on the Net.

Since the poll consisted of "raw" data (even though
the sample was randomly selected - see the notes) it needed to
be "weighted" to match the recorded vote - along
with adding 613 new voters to the original sample, bringing
the final total to 13,660 respondents. And thus we now have
the official National Exit Poll. 

All of these 613 "new" voters must have voted for
Bush in order for the exit poll numbers to match the vote. Of
course, the demographic weightings also had to be changed
accordingly. 

The preliminary 13,047 poll had a Party-ID mix
(randomly-sampled) of 38 Dem/35 Repub/27 Independent. This
demographic also happens to COINCIDE with the Party-ID mix
for the FINAL, WEIGHTED exit polls in 1996 and 2000
(39/35/26). 

SO THE PARTY-ID WEIGHTINGS HAD TO BE CHANGED TO 37/37/26 IN
ORDER TO MATCH THE BUSH 2004 RECORDED VOTE.

Voila!

The final exit poll of 13660 shows that Bush won by 51-48%. 
NOTE THE EDISON/MITOFSKY CAVEAT: "FINAL PERCENTAGES MAY
SHIFT SLIGHTLY"

Remember the Lovin' Spoonful in the sixties?
"Do You Believe in Magic"?
................................................................

[anaxarchos]

17. How exactly is this possible?

"You have consistently said that the exit polls can be used to detect fraud in the actual vote, knowing from day one that they cannot and that the MOE is specifically stated as not being good for anything other than the demographic data of race, gender, and geographic region."

Explain how the above is possible, please... (Just a preview, but it sounds like nonsense to me).

As to massive conspiracies, I posted last night on your conspiracy thread. This should get interesting.

[Name removed]

21. Deleted message

Message removed by moderator. Click here to review the message board rules.

[Mistwell]

34. Answer

"DID YOU FREQUENT OTHER REPUKE BOARDS AND KNOCK DUERS, AS ONE DUER HAS CLAIMED? SO MUCH FOR YOU BEING A DEMOCRAT."

The answer is no, I did not, and yes, I am a Democrat. Your rumormongering against me simply because I disagree with you on the small issue of exit polls is disgusting. If you are calling me a liar, let's hear it. Stop beating around the bush.

As for the answer to the rest of your question - you read Mitofsky's detailed statement about what goes into the exit polls. You know he specifically excludes the "how people voted" data from that MOE. Portions that say "Gender X voted for V by Z%" are specifically not covered by the margin of error. And I know you know that, because you said previously you read the details of the exit poll. So, I'll give you the benefit of the doubt and assume you just forgot...

[johnaries]

24. Misty, I assume that you're frustrated

because it's obvious that the election was stolen and not enough people can see the obvious. But don't take out your frustration on TIA.

I am not a "statistics expert": however, I AM a certified Black Belt Six Sigma expert. Therefore, I have been certified by very exacting Business Standards to use many business tools including data collection and Statistical Analysis. Of course, I realize this means nothing to you. But one thing I've learned is that any data capture attempt is flawed, and before you accept any single result you must compare that result against other input and analysis before you can build an accurate picture. The official vote count is legally considered the final authority and therefore is legally considered the most accurate "poll", but that doesn't mean that it is the most accurate in reality.

I've been following TIA's analyzes, and I've also been following Mitovfsky's and other expert analyzes of the exit polls vs the "election" result.


It is safe to assume that TIA would be biased in his attempts at the statistical analysis of the exit polls vs the "official results". However, TIA has thoroughly documented every assumption he has made, and when new information was available he adjusted his analysis accordingly.

Plain and simply, all evidence indicates that there was a problem with the data collection in the Official Election Process. If, as a Six Sigma Black Belt, I had tried to present the Election Results to my Executive Operating Committee as accurate as opposed to all of the other data to the contrary, I would have been laughed out of the room and told to re-consider my data collection methods and possibly need to go through re-training (if they decided not to fire me on the spot).

What TIA has done is prove that the current data collection process for election polling is flawed. His work is also substantiated by outside research. The "excuses" trying to explain the discrepancies that TIA has pointed out do not pass the "smell" test.

There is something wrong with our election process, and TIA has helped prove it!

[Nederland]

40. Disagree

However, TIA has thoroughly documented every assumption he has made, and when new information was available he adjusted his analysis accordingly.

This is simply not true. TIA's analysis has always assumed that the exit poll sample is representative of the total voter population. That assumption is not only unproven, but completely unacknowledged by TIA as being significant. It reflects a poor understanding of statistics to think that you can simply take the size of the sample pool and plug it into Excel to get an accurate MOE figure.

[FogerRox]

14. this is funny --sounds like a Holly wood movie

1) A massive secret Republican conspiracy spanning the entire nation involving thousands of people intent on defrauding the electorate,



SO it would take thousands to hack the vote in 2004?

I'm so glad I come to DU to get SO well <wink> informed.

Roj runs off to write in Dairy-.-.-.-.-.-.---.


Th , , , ,ou, , , ,s,a, , ,nds of p, ,eop, , , le

.-.-.-. oh wait, , what did they use them chinese things with all the beads.-.-.-.- abacus, Right? that s it yeah I knew that was it.

Bush cheated with Abacuses,,

wo , , ,rk , ing , , ,on, A, ba, cu, ses

[anamandujano]

4. They thrive on faulty logic.

The greater the WTF value, the better it seems to go for them.

Thank you for keeping up your posts TIA, they are some of the few scraps of truth that keep us hanging on to reality.

[Melissa G]

5. Here's to kicking the squatter out of the white house! n/t

[kiwi_expat]

6. I must be missing something.

"(1) The Reluctant Bush Responder theory....
If (1) is true, then the 43-37% Bush margin which matches the recorded vote must be wrong.
And the Preliminary National Exit poll of 13,047 was right after all.
Which means that Kerry won by 51-48%."

Why would the preliminary exit poll be right if the Reluctant Bush Responder Theory (RBRT) was true? I thought that the RBRT was offered as an explanation for the preliminary poll being wrong and the (massaged) final exit poll being right.

Please do not attack me. I'm just asking. Personally I don't believe the RBRT. And I am an agnostic on the accuracy of the national preliminary exit poll.

[TruthIsAll]

7. Hope this clears it up.

Don't worry. I never "attack" anybody.
I'm really not a bad guy. And legitimate questions are always welcome.

"Why would the preliminary exit poll be right if the Reluctant Bush Responder Theory (RBRT) was true? I thought that the RBRT was offered as an explanation for the preliminary poll being wrong and the (massaged) final exit poll being right".

1) The point is this: RBRT cannot be true, because it conflicts with the 43-37% Bush/Gore voter split from the Final Exit Poll.

YOU CANNOT HAVE IT BOTH WAYS.
NEITHER CONCLUSION MAKES SENSE WITHOUT THE OTHER CONFIRMING.
THEREFORE, IF ONE IS FALSE, THEY MUST BOTH BE FALSE.

If you believe the 43/37 split you cannot believe the RBRT. And vice-versa. Therefore you cannot believe the Final Poll which says Bush won. Therefore, the Preliminary Poll of 13,047, which says Kerry won, must be correct.

Furthermore, UScountvotes.org has rebutted the RBRT by analyzing Mitofsky's 77-page report. On the contrary, they found a pro-Bush bias.

[kiwi_expat]

15. I think there is another possibility.

Thanks for your response. I understand the point you are trying to make, now.

However, I think there is another POSSIBILITY: The "final" massaged exit poll Bush/Kerry totals could be correct, but the "who did you vote for in 2000" totals were just not massaged correctly. (In fact, I have the recollection that a subsequent re-massaged final exit poll, showed the extra Bush votes coming from new voters rather than 2000 Gore voters. Which is much more plausible.)

Personally, I do not believe that the final Bush/Kerry totals are correct. However, I do think it is possible that Bush won the national popular vote. I don't know.

BUT I AM CONVINCED THAT KERRY WON OHIO - AND IS OUR TRUE PRESIDENT.

[FogerRox]

16. We killed Bush on Voter Reg. there is a young girl in my town

For MTV rock the vote just ran a volunteer program that registered 450 THOUSAND new voters---just one 21 yr old girl from NJ.

Check this out---Who , , , didnt want to vote for Bush in 2000, but did want to vote for Bush in 2004?

Who?

Doesnt make sense--does it?

If you didnt vote for him the 1st time---why on earth would you vote for him the 2nd term----

[Peace Patriot]

28. Yeah, I was going to say that, too, FogerRox. Dems beat the pants off...

...Repubs in new voter registration, 57% to 41%!

Steven Freeman lays it out in his 2nd paper: Gore repeat voters + Dem blowout in new voter registration in 2004 + big jump of Nader voters to Kerry = a 4 to 8 million vote Kerry win.

And, you're right, too, that it's not just the numbers. Given this blowout success in new voter registration, you have to figure a HIGH LEVEL OF ENTHUSIASM among Gore repeat voters, because they are the people--family, friends, co-workers, volunteers--who got the non-voters to vote.

So, to believe Bush won, you have to believe that

a) new voters were flocking to the Democratic Party in unprecedented numbers--to vote for Bush!???

and

b) the family, friends, co-workers and volunteers who were pushing them to register Democratic and vote were themselves planning to vote for...Bush????

Totally silly.

The fact of the matter is--and we all know this--there was great enthusiasm for OUSTING Bush, and all grass roots groups put aside their differences with the DNC and Kerry (on Iraq, for instance) to accomplish that. And we did.

And I think quite a few Republicans turn against Bush as well--and some of their votes got "disappeared," too.

[DELUSIONAL]

18. Of Course he is -- too bad he wasn't willing to fight for what is right.

It is so easy to add a few votes here and there -- a few adjustments in a whole lot of precincts would give Der Bushita the "popular" vote.

His handlers weren't satisfied that the bag the Electoral Vote -- the had to beat down the Gore effect and had to bag the "popular" vote.

With all the computer connected to the Internet -- it would be very easy to hack the main tabulators and add/subtract votes.

Plus the real convincing argument for me -- 99.9% of the "errors and anomalies" went to bush. For example on some voting machines -- vote straight line democratic -- vote deflated to bush -- WTF???????

[TruthIsAll]

19. Glad to see you are convinced Kerry won, but you are still...

not getting it.

Kerry won the new voters by 59-39% in the preliminary poll.

Every demographic was massaged in the final exit poll,
including new voters.

National Exit Poll Time Line

Preliminary 1:  Nov. 2,  7:38pm, 11027 respondents
Preliminary 2:  Nov. 3, 12:22am, 13047 ""
Final 3:        Nov. 3,  2:04pm, 13660 ""

There was NO CHANGE IN THE AVERAGE OF THE WEIGHTED NATIONAL
PERCENTAGES after 2020 additional respondents were
interviewed between 7:38pm and 12:22am.

  Time Sample Bush  Kerry     Nader
 7:38pm 11027	47.91%	50.76%	1.14%
12:22am 13047	47.90%	50.76%	1.00%

But look what happens between 12:22am and 2:04pm (after just
613 additional respondents):
 2:04pm 13660	50.89%	48.04%	0.30%

Kerry gained in the Party ID split between 7:38pm and
12:22am, going from 38 Dem/36 Rep/26 Independent to 38 Dem/35
Rep/27 Independent (2020 respondents). The effect of this
slight change was to increase his national weighted average
from 50.24% to 50.69%.
 
 Time  Sample Bush Kerry	Nader
 7:38pm 11027	48.24%	50.24%	1.26%
12:22am 13047	47.77%	50.69%	0.92%

But the next day, the Party ID weights mysteriously changed
to 37/37/26 when the final 613 respondents were sampled -
quite a feat.
 2:04pm 13660	49.69%	49.15%	0.27%

Totals may not add to 100% due to rounding.

Here's the progression of the exit poll tranformation from
Kerry to Bush, from left to right, literally and
figuratively.

From an initial 11,027 on CNN at 7:38pm to a pristine 13,047
on WP at 12:22am to a soiled 13660 on CNN at 2:04pm

	CNN	7:38	pm			WaPo	12:22	am	    		CNN	2:04	pm
	11027			               13047		                    13660

	Mix1	Bush	Kerry	Nader		Mix2	Bush	Kerry	Nader		Mix3	Bush	Kerry	Nader

GENDER
Male 	46.0%	51%	47%	1%		46%	52%	47%	1%		46%	55%	44%	0
Female 	54.0%	45%	54%	1%		54%	45%	54%	1%		54%	48%	51%	0
	100%	47.76%	50.78%	1.00%		100%	48.22%	50.78%	1.00%		100%	51.22%	47.78%	0.00%
														
EDUCATION														
	Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
No H.S.	4.0%	47%	52%	1%		4%	47%	52%	1%		4%	49%	50%	0
HSGrad22.0%	48%	51%	1%		22%	48%	51%	1%		22%	52%	47%	0
Colleg31.0%	51%	47%	2%		31%	51%	47%	1%		32%	54%	46%	0
ColGrd 26.0%	50.5%	48.5%	1%		26%	50%	48%	1%		26%	52%	46%	1%
Postg	17.0%	40%	58%	2%		17%	40%	58%	2%		16%	44%	55%	1%
	100%	48.18%	50.34%	1.48%		100%	48.05%	50.21%	1.17%		100%	51.24%	47.82%	0.42%
																										
RACE AND
GENDER	Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
WMale	36.0%	58%	41%	1%		36%	58%	41%	1%		36%	62%	37%	0%
WFem 41.0%	52%	47%	1%		41%	52%	47%	1%		41%	55%	44%	0%
NonWM 10.0%	29%	70%	1%		10%	28%	69%	1%		10%	30%	67%	1%
NonWF 13.0%	22%	77%	1%		13%	22%	77%	1%		12%	24%	75%	1%
	100%	47.96%	51.04%	1.00%		100%	47.86%	50.94%	1.00%		99%	50.75%	47.06%	0.22%
														
AGE	Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
18-29 	17.0%	43%	56%	1%		17%	43%	56%	1%		17%	45%	54%	0
30-44 	27.0%	50%	49%	1%		27%	50%	49%	1%		29%	53%	46%	1%
45-59 	30.0%	47%	51%	2%		30%	47%	51%	1%		30%	51%	48%	0
60+ 	26.0%	51%	48%	1%		26%	51%	48%	1%		24%	54%	46%	0
	100%	48.17%	50.53%	1.30%		100%	48.17%	50.53%	1.00%		100%	51.28%	47.96%	0.29%
														
INCOME	Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
0-$15K 	9.0%	33%	66%	1%		9%	33%	66%	1%		8%	36%	63%	0
$15-30 	15.0%	39%	59%	2%		15%	39%	59%	1%		15%	42%	57%	0
$30-50 	22.0%	47%	52%	1%		22%	47%	52%	1%		22%	49%	50%	0
50-75 	23.0%	53%	45%	2%		23%	53%	45%	1%		23%	56%	43%	0
-100 	13.0%	50%	49%	1%		13%	50%	49%	0%		14%	55%	45%	0
-150 	11.0%	53%	45%	2%		11%	53%	45%	2%		11%	57%	42%	1%
-200 	4.0%	53%	47%	0		4%	53%	47%	0%		4%	58%	42%	0
200+ 	4.0%	58%	41%	1%		4%	58%	41%	1%		3%	63%	35%	0
	101%	48.12%	51.42%	1.46%		101%	48.12%	51.42%	0.95%		100%	51.02%	48.13%	0.11%
														
														
IDEOLOGYMix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
Lib 	22.0%	12%	87%	1%		22%	12%	86%	1%		21%	13%	85%	1%
Mod 	45.0%	42%	57%	1%		45%	41%	57%	1%		45%	45%	54%	0
Con 	33.0%	82%	16%	1%		33%	82%	16%	1%		34%	84%	15%	0
	100%	48.60%	50.07%	1.00%		100%	48.15%	49.85%	1.00%		100%	50.83%	47.95%	0.22%
														
RELIGIONMix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
Prot	53.0%	56%	43%	1%		53%	56%	43%	1%		53%	59%	40%	0
Cath 	27.0%	49%	50%	1%		27%	49%	50%	1%		27%	52%	47%	0
Jewish	3.0%	23%	77%	0		3%	23%	77%	0%		3%	25%	74%	0
Other 	7.0%	20%	75%	5%		7%	20%	75%	4%		7%	23%	74%	1%
None 	10.0%	29%	70%	1%		10%	29%	70%	1%		10%	31%	67%	1%
	100%	47.90%	50.85%	1.25%		100%	47.90%	50.85%	1.18%		100%	50.77%	47.99%	0.17%
														
														
MILITARYMix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
Yes 	18.0%	55%	43%	2%		18%	55%	43%	1%		18%	57%	41%	1%
No 	82.0%	46%	53%	1%		82%	46%	53%	1%		82%	49%	50%	1%
	100%	47.62%	51.20%	1.18%		100%	47.62%	51.20%	1.00%		100%	50.44%	48.38%	1.00%
														
														
WHEN
DECIDED	Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
Today 	6.0%	41%	54%	5%		6%	40%	53%	5%		5%	45%	52%	1%
3Days 3.0%	42%	54%	4%		3%	41%	53%	4%		4%	42%	55%	1%
Week 	2.0%	51%	48%	1%		2%	51%	48%	1%		2%	51%	48%	0
Month	10.0%	38%	61%	1%		10%	38%	60%	1%		10%	44%	54%	1%
Before79.0%	50%	50%	0%		79%	50%	50%	0%		79%	53%	46%	0%
	100%	48.04%	51.42%	0.54%		100%	47.95%	51.23%	0.54%		100%	51.22%	47.50%	0.19%
														
REGIONMix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
East	22.0%	40%	58%	1%		22%	41%	58%	1%		22%	43%	56%	1%
Midwest	25.0%	49%	50%	1%		26%	49%	50%	1%		26%	51%	48%	0
South	31.0%	54%	45%	1%		31%	54%	44%	1%		32%	58%	42%	0
West	21.0%	45%	53%	1%		21%	45%	53%	1%		20%	49%	50%	1%
	99%	47.24%	50.34%	0.99%		100%	47.95%	50.53%	1.00%		100%	51.08%	48.24%	0.42%
														
														
PARTY
ID	Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
Dem 	38.0%	9%	90%	1%		38%	9%	90%	1%		37%	11%	89%	0
Rep 	36.0%	92%	7%	1%		35%	92%	7%	0%		37%	93%	6%	0
Ind	26.0%	45%	52%	2%		27%	45%	52%	2%		26%	48%	49%	1%
	100%	48.24%	50.24%	1.26%		100%	47.77%	50.69%	0.92%		100%	49.69%	49.15%	0.27%
														
														
VOTED IN
2000	Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader		Mix	Bush	Kerry	Nader
No	17.0%	39%	59%	1%		17%	39%	59%	1%		17%	45%	54%	1%
Gore	38.0%	8%	91%	1%		38%	8%	91%	1%		37%	10%	90%	0%
Bush	41.0%	90%	9%	0%		41%	90%	9%	0%		43%	91%	9%	0%
Other	4.0%	13%	65%	16%		4%	13%	65%	16%		3%	21%	71%	3%
	100%	47.09%	50.90%	1.19%		100%	47.09%	50.90%	1.19%		100%	51.11%	48.48%	0.26%



Exit Poll Comparison Summary

Poll   Bush	Kerry    Nader

Gender			
11027	47.76%	50.78%	1.00%
13047	48.22%	50.78%	1.00%
13660	51.22%	47.78%	0.00%
			
Education			
11027	48.18%	50.34%	1.48%
13047	48.05%	50.21%	1.17%
13660	51.24%	47.82%	0.42%
			
Race/Gender			
11027	47.96%	51.04%	1.00%
13047	47.86%	50.94%	1.00%
13660	50.75%	47.06%	0.22%
			
Age			
11027	48.17%	50.53%	1.30%
13047	48.17%	50.53%	1.00%
13660	51.28%	47.96%	0.29%
			
Income			
11027	48.12%	51.42%	1.46%
13047	48.12%	51.42%	0.95%
13660	51.02%	48.13%	0.11%
			
			
Ideology			
11027	48.60%	50.07%	1.00%
13047	48.15%	49.85%	1.00%
13660	50.83%	47.95%	0.22%
			
			
Religion			
11027	47.90%	50.85%	1.25%
13047	47.90%	50.85%	1.18%
13660	50.77%	47.99%	0.17%
			
Military			
11027	47.62%	51.20%	1.18%
13047	47.62%	51.20%	1.00%
13660	50.44%	48.38%	1.00%
			
When Decided			
11027	48.04%	51.42%	0.54%
13047	47.95%	51.23%	0.54%
13660	51.22%	47.50%	0.19%
			
Region			
11027	47.24%	50.34%	0.99%
13047	47.95%	50.53%	1.00%
13660	51.08%	48.24%	0.42%
			
			
Party ID			
11027	48.24%	50.24%	1.26%
13047	47.77%	50.69%	0.92%
13660	49.69%	49.15%	0.27%
			
			
Voted in 2000			
11027	47.09%	50.90%	1.19%
13047	47.09%	50.90%	1.19%
13660	51.11%	48.48%	0.26%
			
Average			
11027	47.91%	50.76%	1.14%
13047	47.90%	50.76%	1.00%
13660	50.89%	48.04%	0.30%

[kiwi_expat]

25. I think I do understand your point.

My analysis, so far, gives me much more confidence in the preliminary precinct exit poll data than in the preliminary state and national exit poll data. That is because the individual precinct data is randomly selected but the state and national totals are not. The preliminary state and national exit poll totals are the product of "pure" precinct data weighted according to Mitofsky's model, a model which MAY OR MAY NOT be correct. (Refer to the discussion on "random sampling" in the "What are the exit poll posters here so afraid of?" thread.)

I was merely saying that it is POSSIBLE that the "final" massaged exit poll Bush/Kerry totals could be correct, but the "who did you vote for in 2000" totals were just not massaged correctly. When it comes to massaging the final totals, any result can be produced. One of them could be correct.

[msongs]

20. are president exit polls historically accurate and suddenly......

the one this year is wrong? we all know the tiny vote margin "flip" against Gore in Florida but how about the exit poll results in other states that year?

how odd that it is widely believed exit polls are accurate except for last year.


Msongs
www.msongs.com/liberaltshirts.htm

Ps - did not the whole election revote in Ukraine get its heft from the exit polls that seemingly were wrong after the first vote.

[Nederland]

23. No

Exit polls have never been accurate in predicting the final vote percentages.

[TruthIsAll]

32. Exit poll are accurate. It's the votes that aren't.

You are in such denial, you can't deal with that.
How do we know exit polls are accurate?

Simple.Let's just assume the loss of democratic votes was due to natural spoilage. This is DOCUMETED as a fact in EVERY election - that millions of votes are lost due to punch card machine anomalies..

Let's disregard fraud for now.
KISS.

So we have the Exit poll identity:
Intended vote = recorded vote + lost vote (spoilage).
I= R + S

The Exit Poll is reflection of the INTENDED VOTE, NOT the RECORDED VOTE tally.

Therefore, the Exit poll can NEVER equal the recorded vote and must be therefore be greater than the recorded vote.

And that is why the Exit Poll is closer to the truth and favors the Democrats.

Can your mathematically-challenged mind accept that?
If it can (I am making a major assumption) then let's expand the analysis to include fraud.

Intended vote = Votes Recorded + Lost (spoilage) + Stolen (fraud).
I= R + S + F

That brings the exit poll all the way there.

The exit poll deviation = Dev = I - R, where R = S + F

[kiwi_expat]

36. Your equation is a good idea. We could tweak it further.

We should add "P"(Uncounted Provisional ballots)to the equation.

In addition to Fraud, we should add "M" (Mis-punched/cast to the wrong candidate but not by fraud), if we are talking about the Intended vote for a Democratic candidate. This happens most commonly in heavily Democratic precincts.


Intended vote for a DEMOCRATIC Candidate = Votes Recorded + Lost (spoilage) + Stolen (fraud)+ Uncounted Provisionals + Mis-cast (not by fraud)
I= R + S + F + P + M


However, if we are talking about the TOTAL intended vote - not just the intended vote for a Democratic candidate - Fraud should not necessarily be added to the Recorded vote. Fraud includes, deliberately switching votes from one candidate to the other and can also involve adding phantom votes to the total count. The net result might be that Fraud should be SUBTRACTED from the TOTAL Recorded votes. Or, better yet, NET fraud should added,

And, of course, if we are talking about the TOTAL intended vote, "M" would not be a factor.

TOTAL Intended vote = Votes Recorded + Lost (spoilage) + NET fraud + Uncounted Provisionals
I= R + S + NF + P

[Nederland]

22. #3 is true

3) The precincts chosen and the pollsters used in the exit poll produce results that are skewed toward Democrats, JUST AS THEY HAVE SKEWED TOWARD DEMOCRATS EVER SINCE 1988.

Now that I've answered your question, will you answer mine? Didn't think so.

[andym]

35. Mitofosky supposedly validated his precinct model

In the NEP paper, Mitofsky seems to validate his selection of precincts and the weight assigned to them in his model by plugging in the actual election data instead of the exit-polling data for these precincts.

The problem is in the data from within the precincts (WPE). They vary widely from the actual data for any individual precinct. Since, 1988 this WPE was greatest in 2004. So the selection of the precincts is apparently not the problem (as it has been improved upon over the years), but the polling data from within any precinct is.

The reason(s) for the skewing of the data has not been conclusively proved by anyone. Though perhaps the Reluctant Bush Responder theory has been discredited by uscountvotes.org (see below)

see my post here for a reasoned discussion on what I think we really know:
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x341540#341699

and what should be done.

As for the main point of this thread: there is a big problem with the argument. The greater proportion of bush responders who voted for bush in 2000 than Kerry supporters who voted for gore in 2000, was likely artificially increased by weighting the exit polls to match the election results. In other words, it does not disprove the reluctant bush voter hypothesis, because it in effect is trying to correct for just this kind of problem. Let me restate this, it is quite likely that in the raw data there are fewer bush(2004)-bush(2000) voters (which could be due to the Reluctant Bush responder among other possibilities) than 43%, but that the weighting boosted it up.

Now even though TIA does not disprove the reluctant bush voter hypothesis, uscountvotes does a great job in showing how weak the Reluctant Bush supporter hypotheses really is using the NEP's own data.

http://uscountvotes.org/ucvAnalysis/US/USCountVotes_Re_Mitofsky-Edison.pdf

[kiwi_expat]

26. Is "Who did you vote for in 2000?" on the 2nd page of the questionnaire ?

If so, we might be able to better access the reliability of the (preliminary) responses for that question by finding out what percent of the respondents answered the questions on the second page.

It would be useful to have a look at preliminary precinct exit poll data. The number of respondents who filled out the second page is shown.

It would also be informative to see if heavily Democratic or heavily Republican precincts had a higher response rate for the second page.

minvis might be able to help, with his copy of the Univ. of Michigan's Ohio NEP "raw" precinct data. (The data is apparently difficult to decipher.)

[Peace Patriot]

29. Nine Ph.D's say otherwise, kiwi-expat--they found a skew to BUSH in...

...the responder data:

Josh Mitteldorf, Ph.D. - Temple University Statistics Department
Kathy Dopp, MS in mathematics - USCountVotes, President
Steven F. Freeman, PhD - Center for Organizational Dynamics, University of Pennsylvania
Brian Joiner, PhD - Prof. of Statistics and Director of Statistical Consulting (ret), University of Wisconsin
Frank Stenger, PhD in mathematics - School of Computing, University of Utah
Richard G. Sheehan, PhD - Department of Finance, University of Notre Dame
Elizabeth Liddle, MA - (UK) PhD candidate at the University of Nottingham
Paul F. Velleman, Ph.D. - Department of Statistical Sciences, Cornell University
Victoria Lovegren, Ph.D. - Department of Mathematics, Case Western Reserve University
Campbell B. Read, Ph.D., Professor Emeritus, Department of Statistical Science, Southern Methodist University
Also Peer Reviewed by USCountVotes’ core group of statisticians and independent reviewers.

This document can be found here: http://uscountvotes.org/ucvAnalysis/US/USCountVotes_Re_Mitofsky-Edison.pdf

[kiwi_expat]

31. Are you referring to the question "Who did you vote for in 2000?" ?

Sorry, I can't read the pdf file. Does it include a discussion of the "Who did you vote for in 2000?" question? Does it say what percent of the 2004 respondents answered that question? Does it say if Democrats or Republicans tended to answer the question more frequently? Thanks.

[Name removed]

27. Deleted message

Message removed by moderator. Click here to review the message board rules.

[Peace Patriot]

30. There is always someone like you, jsw_81--after a long, detailed...

...discussion of the facts--who chimes in with a 2-sentence hit containing words like "conspiracy," "nut," wacky," and "face reality." And "we lost."

No facts. No data. No analysis. No indication whatever that you've reviewed the mountain of evidence of an invalid election with a wrong result. No respect for the long list of Ph.D.'s in statistics and mathematics who have cried foul on the 2004 election and called for an investigation--putting their professional reputations on the line in peer-reviewed reports. Oblivious to the reports by top statisticians of large discrepancies between the electronic totals and the other methods of voting, always favoring Bush. Nothing to say about Bush partisans controlling the vote count with secret, proprietary programming code, and no paper trail. No discussion of the insecurity and hackability of these machines. No concern about the 57,000 complaints to Congress of machine failure and voter intimidation. No mention of the numerous reports of touchscreens changing Kerry votes to Bush votes--virtually all favoring Bush. Ignorant of the odds against a Bush win on every set of facts that are looked at. No comment on Bush's continued miserable approval ratings, since before the election (sinking to 48% on his inauguration day!), with nearly 60% of Americans STILL against the Iraq war, now, today.

Conspiracy. Nut. Wacky. Face reality. We lost.

Yeah, we DID lose, big time. The Bush Cartel is still running our government, despite the will of the majority.

[jsw_81]

33. Tell me this:

Why did this vast conspiracy allow Clinton to win twice?

[seriousstan]

37. GORE/LIEBERMAN 2004!! WOOHOO

Am I too late or is this the forum where hope(lessness) never dies?

If you want to make a difference, get out from behind the computer, wake your lazy as at 5 AM and work the elections for a thrilling 15 hour day.

[TruthIsAll]

38. Makes absolutely no sense.

You ain't serious, are you Stan?

Hope next time you will contribute something more insightful to the discussion.

[seriousstan]

39. How's this, instead of mathematical masturbation try working an

election as a poll worker. Get a real hands-on view. Stop intimidation and suppression at the source, when you legitimately have the power to do so. OR, play number games for your own amusement.

I will try my way, you try yours...we'll see who gets something accomplished first.

[TruthIsAll]

41. YOU will accomplish more than UScountvotes.org?

Number games?
Get serious, Stan.

And Good luck at the polls.
Say hello to all those reluctant Bush responders for me.

[TruthIsAll]

42. Kick this Smoking Gun....... n/t

,

[TruthIsAll]

43. Calling Bobby Fischer....

.

[kansasblue]

44. two questions

I have two questions. First for TIA. Unfortunately your great great analysis isn't really easy to discuss with everyone but I was wondering if you had be attempting to contact anyone to get this info out. Like Brad at Bradblog. Or even Jimmy Carter. It just seems like someone would love to run with this info.

Second question which maybe we all could research. How do they handle exit polls in other countries? Did our exit polling system differ from other countries? What I mean is that if you assume there is a possibility for fraud (and there is always a possibility) and then you have the polling company tell you they had to 'adjust' the polls to make it right. That info in other countries that would send up red flags. It's nice that we give our system the benefit of the doubt but imagine you are reading this same info coming from an election in Russia. Would the general public be so willing to accept the info? Did the Ukrainians use an adjusted exit poll to contest their election? The U.S. uses exit polls all the time as watchdogs. Are we using them the same way?

[TruthIsAll]

46. I keep trying to make it as clear as possible.

I leave it to others to analyze. And it is getting out.
Check out uscountvotes.org

Jimmy Carter?
What would he do with it?
Bring it up with his co-chair on the new election commission, James Baker III?

[bleever]

45. Still true. Kick.