Kerry won Nat. Exit Poll (11027 respondents) posted on CNN @ 7:58PM

[TruthIsAll]

The weighted national percentages are almost identical (Kerry
51-48%) to the 12:22am Preliminary Exit Poll update (13047
respondents) posted on the WP/NEP web site. There was
virtually no change in weights from the 2020 additional
respondents.

Kerry led at 7:58PM.
He maintained his 3% lead at 12:22 AM.

Then how come the BIG FLIP to Bush (51-48%) in the Final Poll
with just 613 additional respondents (13660)?

Simple. To match the vote count by adjusting the weights.

So Kerry won the state and national pre-election polls (both
EV and popular vote).

He also won the state exit polls and the two
"preliminary" national exit polls, posted at 7:58pm
and 12:22am, respectively.

But not the Final National Exit Poll.

http://www.exitpollz.org/CNN_national2.htm

	Weighted Vote by Category			
				
Gender Pct	BUSH	KERRY	NADER
Male	46.0%	23.46%	21.62%	0.46%
Female 54.0% 24.30%	29.16%	0.54%
	100.0%	47.76%	50.78%	1.00%

Race/Gender				
WhiteMale	36.0%	20.88%	14.76%	0.36%
White Female	41.0%	21.32%	19.27%	0.41%
NonWMale	10.0%	2.90%	7.00%	0.10%
NonWFemale	13.0%	2.86%	10.01%	0.13%
	      100.0%	47.96%	51.04%	1.00%

Race				
White	77.0%	42.35%	33.88%	0.77%
Black	11.0%	1.10%	9.90%	0.00%
Latino	9.0%	3.74%	5.09%	0.18%
Asian	2.0%	0.76%	1.22%	2.00%
Other	1.0%	0.39%	0.59%	0.02%
	100.0%	48.34%	50.68%	2.97%

Age				
18-29	17.0%	7.31%	9.52%	0.17%
30-44	27.0%	13.50%	13.23%	0.27%
45-59	30.0%	14.10%	15.30%	0.60%
60+	26.0%	13.26%	12.48%	0.26%
	100.0%	48.17%	50.53%	1.30%

Income				
0-15	9.0%	2.97%	5.94%	0.09%
15-30	15.0%	5.85%	8.85%	0.30%
30-50	22.0%	10.34%	11.44%	0.22%
50-75	23.0%	12.19%	10.35%	0.46%
75-100	13.0%	6.50%	6.37%	0.13%
100-150	11.0%	5.83%	4.95%	0.22%
150-200	4.0%	2.12%	1.88%	0.00%
200+	4.0%	2.32%	1.64%	0.04%
	101.0%	48.12%	51.42%	1.46%

Education				
No HS	4.0%	1.88%	2.08%	0.04%
HS	22.0%	10.56%	11.22%	0.22%
SomeCol	31.0%	15.81%	14.57%	0.62%
ColGrad	26.0%	13.13%	12.61%	0.26%
PostGr 17.0%	6.80%	9.86%	0.34%
	100.0%	48.18%	50.34%	1.48%

Party_ID				
Dem	38.0%	3.42%	34.20%	0.38%
Rep	36.0%	33.12%	2.52%	0.36%
Ind	26.0%	11.70%	13.52%	0.52%
	100.0%	48.24%	50.24%	1.26%

Ideology				
Lib	22.0%	2.64%	19.14%	0.22%
Mod	45.0%	18.90%	25.65%	0.45%
Conserv	33.0%	27.06%	5.28%	0.33%
	100.0%	48.60%	50.07%	1.00%

Voted Before?				
No	11.0%	4.73%	6.16%	0.00%
Yes	89.0%	42.72%	44.50%	0.89%
	100.0%	47.45%	50.66%	0.89%

Religion				
Prot	53.0%	29.68%	22.79%	0.53%
Cath	27.0%	13.23%	13.50%	0.27%
Jewish	3.0%	0.69%	2.31%	0.00%
Other	7.0%	1.40%	5.25%	0.35%
None	10.0%	2.90%	7.00%	0.10%
	100.0%	47.90%	50.85%	1.25%

In Military?				
Yes	18.0%	9.90%	7.74%	0.36%
No	82.0%	37.72%	43.46%	0.82%
	100.0%	47.62%	51.20%	1.18%

When Decided				
Today	6.0%	2.46%	3.24%	0.30%
3days	3.0%	1.26%	1.62%	0.12%
Week	2.0%	1.02%	0.96%	0.02%
Month	10.0%	3.80%	6.10%	0.10%
Before	79.0%	39.50%	39.50%	0.00%
	100.0%	48.04%	51.42%	0.54%

Family Financial Situation				
Better	31.0%	24.49%	6.51%	0.00%
Worse	29.0%	5.22%	23.49%	0.29%
Same	40.0%	18.40%	21.20%	0.40%
	100.0%	48.11%	51.20%	0.69%
				
Confident Votes counted accurately?				
Yes	90.0%	45.00%	44.10%	0.90%
No	10.0%	2.70%	7.10%	0.10%
	100.0%	47.70%	51.20%	1.00%

Vote in 2000				
No	17.0%	6.63%	10.03%	0.17%
Gore	38.0%	3.04%	34.58%	0.38%
Bush	41.0%	36.90%	3.69%	0.00%
Other	4.0%	0.52%	2.60%	0.64%
	100.0%	47.09%	50.90%	1.19%
				
Region				
NE	22.3%	9.01%	13.02%	0.22%
MidWest	25.3%	12.37%	12.63%	0.25%
South	31.3%	16.88%	14.06%	0.31%
West	21.3%	9.67%	11.37%	0.21%
	100.0%	47.93%	51.07%	1.00%

Size of community				
BigCity	13.0%	4.55%	8.45%	0.00%
SmCity  19.0%	8.74%	10.07%	0.19%
Suburbs	44.0%	22.00%	21.56%	0.44%
SmTown 8.0%	3.76%	4.08%	0.16%
Rural	16.0%	9.12%	6.72%	0.16%
	100.0%	48.17%	50.88%	0.95%

[fooj]

1. Just in case...please forward these stats to Conyers and whoever else

you deem appropriate. You are one tenacious warrior, TIA! Peace.

[merwin]

2. What we really need is the raw data... which he will never give out.

[TruthIsAll]

3. This IS effectively the raw data, weighted by actual responses.

Just multiply the percentages to get the totals.

[TruthIsAll]

4. kick for discussion n/t

.

[abbiehoff]

5. The one thing that always stands out for me in these polls

is that they seem to have polled so many more 2000 Bush voters than 2000 Gore voters even though Gore won the popular vote. How could Bush possibly come out the winner if in fact those who voted for Gore in 2000 voted for Kerry in nearly the same percentage as those who voted for Bush twice, and those who did not vote at all in 2000 were much more likely to vote for Kerry.

Vote in 2000
No 17.0% 6.63% 10.03% 0.17%
Gore 38.0% 3.04% 34.58% 0.38%
Bush 41.0% 36.90% 3.69% 0.00%
Other 4.0% 0.52% 2.60% 0.64%
100.0% 47.09% 50.90% 1.19%

I guess the answer is that many people who voted for Gore in 2000 just didn't bother to vote at all in 2004 even though there were record numbers of voters.

[TruthIsAll]

6. This should be obvious, but they are either ill-informed, in denial

of the truth or lacking ability to analyze the facts.

Or they value personal comfort more than democracy itself.

[anaxarchos]

10. It's evidence of fraud either way...

Argue it one way and the exit polls are improperly weighted to "match" the vote. Argue it the other, and the Republicans are remarkably good at voter suppression.

This is really why it's getting a little funky around here...

The NEP analysis took the wind out of the sails of a lot of the "nay" theorists.

[kster]

11. I've noticed that

to,but I think its a "whole lot" funky in some of the threads, some of the most useless ones seem to keep appearing at the top getting posts,while the best ones get shuffled to the bottom with little or no posts in them,I thought it was just me. Maybe something big is going to break! Lets hope. NGU

[understandinglife]

7. I continue to communicate your analysis to Conyers and others, as...

...I'm sure others are doing. It is so important to keep a steady stream of facts and analysis flowing their way.

Peace.

[skywalker_5]

8. thanks for the info n/t

[wiggs]

9. Thanks...and question for TIA

So....this weighting that was done after midnight....I think everyone agrees that the weighting happened, right? Pollsters have admitted to necessary "adjustments", as I remember. DO YOU THINK THE WEIGHTING WAS DEMOGRAPHICALLY APPROPRIATE OR NOT? Do we know the basis of weighting....age, race, political affiliation, gender? Was the correction meant to reflect the actual vote demographic? How closely did the 13000 person exit poll demographic (prior to weighting) match the actual vote demographic and did it more closely or less closely match the actual vote demographic AFTER weighting?

Seems to me that the theory of an accurate exit poll (pre-weighting) that was changed to cover up election irregularities is weaker if the weighting effort is shown to be reasonable. Have pollsters said how they did the weighting and why? If their answer is reasonable, then they could diffuse some of the confusion out there. Should be an easy question to answer.

But in any case, seems to me that scrutiny should be on the weighting...what, when, how and why.

[TruthIsAll]

12. A quick answer.

Let's cut right to the core.

This is the thread which displays the corresponding weights for the two preliminary and final exit polls:
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x340029

The FINAL 2000 exit poll closely matched the vote (within 0.5%).
Gore won the popular vote by 540,000 votes.

In 2000, the FINAL exit poll had the following Party ID weights:
Dem 39%
Rep 35%
Ind 26%
I don't know what the preliminary mix was, but Gore won the preliminary exit poll by 2%.

Note that exit poll percentages are rounded to the nearest 1%. It makes it easier to fudge the numbers later on. There is NO other explanation for not displaying the numbers to one decimal place.

In 2004, the preliminary exit poll of 11027 posted at 7:38pm on CNN had the Party ID as:
Dem 38%
Rep 36%
Ind 26%
Kerry won the poll 51-48%.

In 2004, the preliminary exit poll of 13047 (2020 respondents added to 11027) posted shortly after midnight had the Party ID as:
Dem 38%
Rep 35%
Ind 27%
There was virtually no change in the demographic weightings in going from 11027 to 13047.
Kerry's exit poll margin remained at 51-48% (average of 10-12 weighted demographics).
Kerry also won new voters by 59%-39%.

The 2004 Final exit poll of 13660 posted at 2:04 pm on Nov. 3 had this Party ID mix:
Dem 37%
Rep 37%
Ind 26%
Bush won the poll 51-48.
Yes, the poll weights were adjusted - to match a (corrupted) vote.

Now consider these questions very simple questions:
1) How did the Party ID mix change in the FINAL 2004 exit poll from the FINAL 2000 exit poll AND the TWO preliminary 2004 exit polls? The mix deviated from 38/35/27 to 37/37/26 in the final 2004 exit poll. How could this Party ID change occur when the majority of NEW voters said they voted for Kerry. Do you really think they were Republicans?

2) How could there have been a 43/37% split in 2004 of former Bush/Gore voters in 2000, as indicated in the final exit poll?
The preliminary poll had it 41 Bush/38 Gore. Did Gore voters decide to stay home?

Didn't Mitofsky float the Reluctant Bush Responder theory as an explanation for why Bush voters avoided talking to the exit pollsters? This factually unsubstantiated "theory" was completely debunked by Uscountvotes.org; in fact, they cited stats from Mitofsky's own 77-page report which indicate that just the opposite occurred: Bush voters responded MORE vigorously to exit pollsters then those Kerry voters- by the same 3% margin as the preliminary exit poll.

THE FINAL EXIT POLL WAS MODIFIED TO A 43/37 BUSH/GORE 2000 VOTER SPLIT. THEY HAD TO DO THIS IN ORDER TO MATCH THE RECORDED VOTE.

UNEXPLAINED EXIT POLL DISCREPANCIES? NO LONGER.
WHAT WE NOW HAVE ARE CONTRADICTIONS IN THE DATA.
IT WAS INEVITABLE.

AND THERE IS JUST ONE LOGICAL EXPLANATION TO THESE CONTRADICTIONS:
IT'S A FIVE-LETTER WORD BEGINNING WITH F.

CONTRARY TO WHAT A NAYSAYER HAS CLAIMED TODAY ON THIS FORUM (THAT MATHEMATICS IS AN ART), MATHEMATICS HAS ALWAYS BEEN THE PUREST OF SCIENCES.

AND SCIENTIFIC ANALYSIS IS ANATHEMA TO THE MSM.
JUST ASK WINSTON SMITH.

[kster]

13. Kick

[wiggs]

15. My question remains....

how does the final exit poll party ID of 37/37/26 compare with party ID of the actual vote? THERE MUST BE DOCUMENTATION OF PARTY ID OR GENDER IN THE ACTUAL VOTE, OTHERWISE THERE WOULD BE NO BASIS FOR THE WEIGHTING, RIGHT? That's the crux of it...if the exit polls were adjusted to better approximate the demographics of the actual vote, this seems reasonable. So we need to find out if that was the goal of the weighting or not.

Even now, I can't believe Kerry wouldn't beat the worst administration ever by a lot more than 51-48 (13047 poll) so I'm with you in that regard. I just want to know how we can pin down the weighting of exit polls as evidence of someone trying to cover up election fraud. No...I don't believe the majority of new voters were republican. No way. But let's not just assume...let's find out what the party ID of actual voters were on election day. Wouldn't poll books tell us?

I would like to know what is happening to all your dogged analysis...is it being used as part of a developing investigation? Is anyone in authority looking at the same stuff?

[berniew1]

14. The Exit Polls were heavily biased against Kerry- here's why.

Hundreds of thousands of Kerry voters in many states like Florida, Ohio, Pennsylvania, etc. and many thousands in most states were prevented from voting or having their vote count by well documented suppression of minorities and students. These are eligible voters who should have been able to vote but weren't allowed to, and all together they amount to millions. And they were excluded from both the official vote totals and the Exit Poll numbers.

minorities: http://www.flcv.com/summary.html
or http://www.flcv.com/ussumall.html

students: http://www.flcv.com/studentv.html