Democratic Underground Latest Greatest Lobby Journals Search Options Help Login
Google

SO IT COMES TO THIS: DO YOU BELIEVE 13,047 OR 13,660?

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is archived.
Home » Discuss » Topic Forums » Election Reform Donate to DU
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Feb-21-05 09:05 PM
Original message
SO IT COMES TO THIS: DO YOU BELIEVE 13,047 OR 13,660?
Edited on Mon Feb-21-05 09:40 PM by TruthIsAll
Let's analyze the last 3 National exit polls

1996
PARTY ID WEIGHTS:
DEM 39.4%
REP 34.7%
IND 25.9%

Clinton won the NEP: 48.76-40.58-8.50%
CLINTON WON THE VOTE: 49.20-40.70-8.40%
RIGHT ON THE MONEY. WITHIN 0.44%
.......................
2000
PARTY ID WEIGHTS:
DEM 39%
REP 35%
IND 26%

Gore won the NEP: 48.29-48.09-2.44%
GORE WON THE VOTE: 48.38-47.87-2.74%
RIGHT ON THE MONEY. WITHIN .09%
..............................
2004
PARTY-ID WEIGHTS:
FROM THE WP 13,047 RESPONDENT POLL
DEM 38%
REP 35%
IND 27%

Kerry won the NEP: 50.74-47.93-1.00%
BUSH WON THE VOTE: 50.73-48.28-0.99%
WHAT HAPPENED?
................................................

Here are links to the National Exit Poll demographics.
1996 CNN National Exit Poll 11/06/96
http://www.cnn.com/ALLPOLITICS/1996/elections/natl.exit ...

2000 MSNBC
http://www.msnbc.com/m/d2k/g/polllaunch.asp


2004 WP (13047 respondents)/ CNN (13660)

Do you believe this poll of 13660 respondents?
CNN poll of 13660 respondents 11/03/04 Updated: 2:04 p.m.
http://www.cnn.com/ELECTION/2004/pages/results/states/U ...

Or do you believe this poll of the first 13047 respondents?
2004 Washington Post, 13,047 respondents, created 11/03/04

http:
//media.washingtonpost.com/wp-srv/politics/elections/2004/graphics/exitpolls_us_110204.gif

WP (13047)
The exit poll shows Kerry winning all the demographics, by an average of 50.8-47.9%. The most critical demographics appeared on the NEP/Washington Post site on Nov 3 (shown below).

CNN (13660)
The 13047 poll was later expanded to include 613 additional respondents, bringing the total to 13,660 - and Bush magically won it by 51-48%.

The WP Poll had Kerry winning 51-48.
It included 95.5% (13,047) of the final CNN Poll total of 13,660 respondents.

Therefore, Bush must have won virtually ALL of the remaining 4.5% polled (613) to win the CNN Poll of 13660 by 51-48.

Something fishy is going on here.

Let's do the math:
In the WP Poll, Kerry won 6654 of the 13,047 polled (51-48%).
In the CNN Poll, he won 6693 of the 13,660 polled, a gain of just 39 votes in the additional 613 polled.

In the WP Poll of 13047, Bush won 6263 votes.
In the CNN Poll of 13660, he won 6967 (a gain of 704 votes).
It's impossible, since there were only 613 new votes (13660- 13047).

How do you explain it?

........Polled..................Kerry Bush
Bush 13660 51% 48% 6693 6967
Kerry 13047 51% 48% 6654 6263
...............................Gain 39 704






Printer Friendly | Permalink |  | Top
 
IndyPriest Donating Member (685 posts) Send PM | Profile | Ignore Mon Feb-21-05 09:51 PM
Response to Original message
1. What an AMAZING cohort of voters those 613 are!
Among other things, they must have been the ones who KNEW the exit poll takers were all Kerry supporters and decided to set the record straight!
Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Feb-21-05 10:51 PM
Response to Original message
2. 3 NATIONAL EXIT POLLS: DEMOGRAPHIC-WEIGHTED VOTE% STATS
Edited on Mon Feb-21-05 11:27 PM by TruthIsAll
NATIONAL EXIT POLL - WEIGHTED AVERAGE VOTE

Split  Female/Male   Dem/Rep/Ind
1996     52/48       39/35/26
2000     52/48       39/35/26 
2004     54/46       38/35/27
												
		2004				2000				1996		
DEMOGRAPHIC	Kerry	Bush	Nader		Gore	Bush	Nader		Clinton	Dole	Perot
												
Gender		50.78%	48.22%	1.00%		48.24%	47.80%	2.48%		48.72%	40.88%	8.44%
Race		50.94%	47.86%	1.00%		48.46%	47.91%	2.73%		48.76%	41.12%	8.34%
Age		50.53%	48.17%	1.00%		48.66%	48.05%	2.51%		48.85%	40.53%	8.69%
Education	50.21%	48.05%	1.17%		47.59%	48.72%	2.48%		49.02%	39.64%	9.41%
Income		51.42%	48.12%	0.95%		48.19%	48.28%	2.30%		48.89%	40.79%	7.45%
												
												
PartyId		50.69%	47.77%	0.92%		48.04%	48.36%	2.69%		48.74%	40.77%	8.46%
Ideology	49.85%	48.15%	1.00%		47.73%	48.22%	2.55%		48.99%	41.14%	8.27%
PrevVote	50.90%	47.09%	1.19%		48.64%	47.81%	2.37%		48.75%	41.18%	8.16%
Decided		51.23%	47.95%	0.54%		48.13%	48.55%	2.20%		48.63%	39.97%	8.58%
Religion	50.85%	47.90%	1.18%		48.55%	48.07%	2.71%		48.22%	39.76%	9.23%
												
Average		50.74%	47.93%	1.00%		48.22%	48.18%	2.50%		48.76%	40.58%	8.50%
Actual		48.28%	50.73%	0.99%		48.36%	47.87%	2.74%		49.20%	40.70%	8.40%
Diff		-2.46%	2.80%	-0.01%		0.14%	-0.31%	0.24%		0.44%	0.12%	-0.10%
Printer Friendly | Permalink |  | Top
 
kiwi_expat Donating Member (526 posts) Send PM | Profile | Ignore Mon Feb-21-05 11:40 PM
Response to Original message
3. We need to compare raw data with raw data.
As I said in my post on your Feb-21 8:29am thread, your 1996 CNN and 2000 MSNBC data do not appear to be RAW data.

(Teixeira claims that his 1996 and 2000 exit poll data is actual raw data. See http://www.tcf.org/publications/pow/nov17_2004.pdf )

The discrepancy between the 2004 WP data and the 2004 CNN data is obviously due to the 2004 CNN data being less "raw" (more "weighted").

I believe that election fraud occurred, big time. I'm still not convinced that the raw exit poll data proves it.

Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Feb-22-05 12:43 AM
Response to Reply #3
4. The RAW pristine poll data should already be weighted by design.
Edited on Tue Feb-22-05 12:44 AM by TruthIsAll
The CNN 13660 poll is contaminated.

The RAW data is already weighted - if the pollsters did their job correctly in the first place.

The 2004 demographics are correct, by comparing to prior elections, they are right on the money. Focus on PartyID, gender and the new voters (17%).

The misrepresentation and confusion regarding raw vs. weighted data is astounding.

My take:
Good data = Sample (Raw data)
Good votes = Fix (bad votes)
Weighted data = Fix (Raw data).
Weighted data = Good votes


Printer Friendly | Permalink |  | Top
 
jkd Donating Member (151 posts) Send PM | Profile | Ignore Tue Feb-22-05 01:52 AM
Response to Reply #4
5. Would Mitofsky really lose their credibility in this manner?
That early data was not for public use. NEP calls it preliminary results. How do we know it was weighted? I wish NEP was more forthcoming.
Printer Friendly | Permalink |  | Top
 
kiwi_expat Donating Member (526 posts) Send PM | Profile | Ignore Tue Feb-22-05 06:25 AM
Response to Reply #4
6. Why compare with "contaminated" 1996 CNN and 2000 MSNBC final data?
In your other thread you appear to agree that the 1996 CNN and 2000 MSNBC poll results have been "manipulated to match the votes".

"That's why the Dems always win the exit polls and lose the votes.
And why the final exit polls look so good - because they're manipulated to match the votes.
Even Mitofsky admits it.
He claims it's SOP."]

So why should we be impressed if the 2004 Washington Post data's PARTY ID. WEIGHTS match those in the manipulated("contaminated") 1996 CNN and 2000 MSNBC final exit poll data?

Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Feb-22-05 08:22 AM
Response to Reply #6
7. Here is why...
Edited on Tue Feb-22-05 09:20 AM by TruthIsAll
Notice that percentages are rounded to the nearest 1%?
That leaves room to fudge.

1) The 2000 exit poll was very close to the actual. There was no need to massage the exit poll data more than 1% to match the actuals. Bush only needed to stop the recount in FL.

2) Because Clinton won handily in 1996, despite the exit poll deviation from the actual. There was no need to massage the exit poll data. They could not steal it.


3) There are more Democrats than Republicans.
Do you believe the 37/37/36 PartyID weights in the CNN poll?
Do you believe the Republicans registered more voters tan Dems in 2004?
We know the opposite took place.

They had to revise the numbers in going from 13047 to 13660.
Bush did not win all the additional voters.
The PartyID weights had to be revised accordingly.

The game goes for the gender weights.
More women voted this time.
They were young, single and pro-abortion.

So what are we left with?
The party-ID tells the story.
Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Tue Feb-22-05 09:51 AM
Response to Reply #7
8. What are we left with ................ Lots of speculation
Edited on Tue Feb-22-05 10:16 AM by Aussie_expat
1."The 2000 exit poll was very close to the actual. There was no need to massage the exit poll data more than 1% to match the actuals. Bush only needed to stop the recount in FL."

Am I to believe that Bush was to know in the afternoon of election day that he only needed to stop the recount?

2."Because Clinton won handily in 1996, despite the exit poll deviation from the actual. There was no need to massage the exit poll data. They could not steal it."

This does not address the exit poll deviation. Why did they massage the data anyway then TIA? and shouldn't this type of "pristine" data that should be used to compare with the 2004 "pristine" data that you use in you calculations?

3."There are more Democrats than Republicans."

What are the percentages?

4."Do you believe the 37/37/36 PartyID weights in the CNN poll?"

Seems hard to believe based on past exit poll data.

5."Do you believe the Republicans registered more voters tan Dems in 2004?" We know the opposite took place.

How many new Dem's registered to vote in 2004? How many new Republicans?

If the opposite took place then what was the spilt?



Printer Friendly | Permalink |  | Top
 
NationalEnquirer Donating Member (571 posts) Send PM | Profile | Ignore Tue Feb-22-05 01:01 PM
Response to Reply #8
9. Guess you are right.
Bottom line is alot of educated guessing, which is probably why in the end, this is going nowhere.

Sorry folks, was off the board here for several days, and my mood has soured, I need a DU pick-me-up!
Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Feb-22-05 02:26 PM
Response to Reply #8
10. No speculation. Eliminate the impossible, you are left with the truth.
Edited on Tue Feb-22-05 02:57 PM by TruthIsAll
"Am I to believe that Bush was to know in the afternoon of election day that he only needed to stop the recount?"

Did I say that?
Of course not.

But they knew about the Palm Beach Butterfly ballot, and the statewide 110,000 overvotes and the 75,000 undervotes.
They knew about Diebold (Volusia 16,022 vote).
And they knew they had friends on SCOTUS.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>


2."Because Clinton won handily in 1996, despite the exit poll deviation from the actual. There was no need to massage the exit poll data. They could not steal it."

This does not address the exit poll deviation. Why did they massage the data anyway then TIA? and shouldn't this type of "pristine" data that should be used to compare with the 2004 "pristine" data that you use in you calculations?
.....


Here goes:

Mitofsky has always said that the exit polls have always been adjusted to match the votes - but he claims the votes are always more accurate and therefore must re-weight the "raw" data accordingly. You would think that his pre-designed sample, based on years of experience, would give him accurate numbers from the get-go. The irony is that they do just that. But he still cannot use them.

The real reason for re-weighting the exit poll is never mentioned, but Greg Palast has indirectly provided the answer: The discrepancy between the votes as cast and the votes counted is primarily due to endemic ballot spoilage, mostly in Democratic precincts. And we are talking about millions of votes here.

In other words, ballot spoilage is anticipated, so by definition the exit polls can never match the vote count unless the exit poll "raw" data are adjusted accordingly. Mitofsky does not want to be known as a lousy exit pollster; after all, he has a reputation to uphold. So he matches the exit polls to the actual recorded votes - but under the guise of re-weighting for exit poll error. In a sense, he is right. Exit polls do not agree with the recorded, final vote.
But that does not mean the votes are correct. Of course, as Americans, we have been brought up to think the votes are correct.
How naive of us.

That was then. This is now. The practice of matching the votes continues. We still have natural ballot spoilage. But now there are other anomalies, like touch screens which register Kerry votes for Bush. And missing punched cards.

Mitofsky is just doing what he's always done: assuming the vote counts are "correct" and the exit polls are "raw" and unadjusted. So he continues to match the "bad" exit polls to the "good" votes, even if it means that the final weightings must be changed accordingly (see the final NEP Exit Poll of 13660) .

That's how he goes from the "preliminary" WP 13,047 poll with a 38/35/27 PartyID mix, to the "final" 13660 poll with a 37/37/26 mix. Of course, the component percentages must change as well.

The two sets of exit poll adjustments (the demographic weights and the percentage increase in Bush's share) are a powerful mix.

That's why the "Preliminary" WP poll was correct to begin with: it reflected the voter's true intentions - which the recorded vote did not.

And that is why the Democrats always win in the exit polls but only sometimes win elections.

Just my 2c.
And that of Gauss and Bernoulli.
Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Tue Feb-22-05 03:26 PM
Response to Reply #10
12. Let me try to be clearer .........
Here's my issue and it seem's that we are talking in circles.

You state that the "pristine" Exit Polls ie: WP 13,047 are always right or within a slim MOE.

Is that correct?

If so then why do you use the "massaged" exit poll figures from the 1996, and 2000 elections for your spreadsheets married "pristine" 2004.

1) a-"Clinton won NEP: 48.76-40.58" = "massaged 1996
b-"Clinton: 52.2% Dole: 37.5%" = "pristine" 1996

2) a-"Gore won the NEP: 48.29-48.09" = massaged 2000
b-"Gore won 48.5% Bush: 46.2%" = "pristine 2000

3) a-"Bush won the NEP 51% Kerry 48%" = "massaged" 2004
b-"Kerry won 51% Kerry:48%" = "pristine" 2004

You are using 1a 2a and 3b for your formulas. or apples and oranges.

This is the problem that I have with your spreadsheets.

IMHO by using supposition and poetic licensing the results are rendered near useless.

This is why I believe that your work can get no traction outside of a few here at the DU.

Remember that I am not critical of your math but just the original figures that you have been using.








Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Feb-22-05 05:11 PM
Response to Reply #12
13. Yes, that is true..
Edited on Tue Feb-22-05 05:24 PM by TruthIsAll
Comparing the "Preliminary" 2004 exit poll to the "final", massaged polls of 1996 and 2000, was mixing apples and oranges. So now let's put that one to rest and focus on the crux of the issue. It was still very instructive to do the comparison. A number of important issues were brought into sharp relief.

Prior to these comparative posts, I focused on the 2004 National and State exit polls, since there has been no controversy over who won the popular vote in prior elections.

The crux of the issue is really just this:
If you believe that the final recorded votes accurately reflect voter intent, then the preliminary exit polls (before adjustment to the correct votes) must necvessarily be wrong.

However, if you believe the recorded votes do not accurately reflect voter intent, then the preliminary exit polls (before adjustment to the recorded/contaminated votes), may very well be right.

Many naysayers have assumed the former proposition. The exit polls are consistently wrong. Just compare them to the actual votes.
That argument is a logically bankrupt.

In fact, final recorded votes do not match voter intent - but for a number of reasons, some of which are universally accepted.

Much confusion has arisen from this simple fact: the vast majority of citizens, including myself, have assumed that prior to 2004, the election results were fairly accurate reflections of voter intent. We now know otherwise.

That's why the "preliminary" 13,047 exit poll is a Smoking Gun. It reflects what everyone believed the evening of Nov. 2 - a 51-48% Kerry win.

The Poll's 38/35/27 PartyID mix agreed with the final, massaged 1996 and 2000 exit polls - and the Dems won the popular vote in both.

Can we agree on the following:
1) According to Mitofsky, exit polls have always been "weighted" to match the actual vote.
2) Punched card/lever ballot vote spoilage, primarily in minority precincts, has always been a factor - and has cost democrats millions of votes.
3) There is no disputing the fact that touch screen computers are vulnerable to hacking.
4) Touch screens switched Kerry votes to Bush in 86 of 88 documented incidents.
5) There is no verified paper ballot for the touch screens.
6) Tabulation of optical scanned ballots uses proprietary software.
7) Recorded votes do not accurately reflect total voter intent.
8) As a result of the above, exit polls can never match the recorded votes, unless they are manipulated.
9) Exit polls reflect voter intent to within a 1% margin of error, as stipulated by Edison/Mitofsky in the notes accompanying the demographic summary of the "preliminary" poll of 13,047 respondents.
10) Exit polls have been very accurate worldwide in predicting the final vote and are used to monitor for fraud - but not in the U.S.


Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Tue Feb-22-05 07:10 PM
Response to Reply #13
15. I'll take a few of these at a time .........
Edited on Tue Feb-22-05 07:16 PM by Aussie_expat
I know I'm setting myself up for proper flaming but ........

1) According to Mitofsky, exit polls have always been "weighted" to match the actual vote.

I absolutly agree.

2) Punched card/lever ballot vote spoilage, primarily in minority precincts, has always been a factor - and has cost democrats millions of votes.

I agree on the spoilage. I have no idea how many votes have been lost over what period of time. I will stipulate that the Dem's have been on the short end of the stick.
I may get flamed for this but it has always bothered me that the bulk of problems associated with spoilage etc. comes from Dem controlled precincts. I know there are a million reasons given for this but I have yet to find one that is reasonable, at least IMHO.

3) There is no disputing the fact that touch screen computers are vulnerable to hacking.

Yes, any computer is vulnerable to hacking. Absolutely.

4) Touch screens switched Kerry votes to Bush in 86 of 88 documented incidents.

Yes, if by documented you mean that someone said so. IMHO documented means multiple unbias witnesses. Which I will agree was unlikely to happen on election day.

5) There is no verified paper ballot for the touch screens.

This is the biggest problem of all. I don't see whats so hard about marking and counting a piece of paper.

6) Tabulation of optical scanned ballots uses proprietary software.

I need to look more into this. But I will say that I am not impressed with computer voting.

7) Recorded votes do not accurately reflect total voter intent.

That statement is so broad I don't know what to make of it. If you mean that some votes were not counted how the voter intended then I would have to agree........... but in and of itself that means nothing as it works both ways.

8) As a result of the above, exit polls can never match the recorded votes, unless they are manipulated.

Change this "As a result of the above," to this "As a result of the above and other contributory factors," and I would be more inclined to agree.

9) Exit polls reflect voter intent to within a 1% margin of error, as stipulated by Edison/Mitofsky in the notes accompanying the demographic summary of the "preliminary" poll of 13,047 respondents.

No, I don't agree. but we can discuss this later.

10) Exit polls have been very accurate worldwide in predicting the final vote and are used to monitor for fraud - but not in the U.S.

For later too.
Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Wed Feb-23-05 09:31 AM
Response to Reply #13
18. Are you finished with this thread after that gem above? n/t
.
Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Feb-22-05 06:55 PM
Response to Reply #12
14. What supposition? What poertic license,
Edited on Tue Feb-22-05 06:56 PM by TruthIsAll
You are reverting to your former self.
You are mis-characterizing the facts.

The only thing you are focusing on are the 3 National exit polls.
Ok, you are right about that.

You keep insisting that the apples and oranges refutes my work.
You are wrong about that.

But you are avoiding the issues that really matter.
Are you building straw men?

Fact 1:
Democrats have always lost from the exit polls to the vote.

Fact 2:
Many Democratic voters exit polled do not realize that their votes have been spoiled.

Fact 3:
Exit polls are matched to the actual votes.

Therefore:
Fact 4:
The actual votes cannot reflect actual voter intent.

Therefore:
Fact 5:
The final, "weighted" exit polls do not reflect voter intent.

So...
Fact 6:
The preliminary exit polls may very well reflect voter intent.

Fact 7:
The 2004 Preliminary National Exit Poll showed that Kerry won.


"IMHO by using supposition and poetic licensing the results are rendered near useless.
This is why I believe that your work can get no traction outside of a few here at the DU.
Remember that I am not critical of your math but just the original figures that you have been using".

Let me return the favor and spell it out as clear as I can.

I have expressed 10 points in the prior post.







Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Tue Feb-22-05 07:13 PM
Response to Reply #14
16. Relax.......and remember
Edited on Tue Feb-22-05 07:17 PM by Aussie_expat
I never said that there was no fraud.

I just said that the data you were posting as "proof" was flawed.

Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Feb-23-05 08:12 PM
Response to Reply #16
22. As it turns out, the "flawed" data actually proves the case.
You are right, the "flawed" exit poll data is misleading, but not for the reason you give us. It's "flawed" because it is based on a based on a "flawed" vote count.

And the original, "flawed" 13,047 exit poll, sweet irony, is the one that is correct... and it shows that Kerry won.

Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Wed Feb-23-05 09:56 PM
Response to Reply #22
23. You are putting words into my mouth now .........
I said your calculations are "FLAWED".

You admitted as much here:

http://www.democraticunderground.com/discuss/duboard.ph...

And here:

http://www.democraticunderground.com/discuss/duboard.ph...

I understand your position and agree that it is a reasonable theory for the most part.
Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Feb-24-05 12:09 AM
Response to Reply #23
25. I never said the calculations were flawed. YOU are putting words
Edited on Thu Feb-24-05 12:13 AM by TruthIsAll
in my mouth.

I admitted that comparing final "weighted" polls (1996,2000) to 2004 was a mistake on my part. That does NOT invalidate my calculations one iota.

It's now very clear why the final exit polls have been "weighted" to agree with the votes: to avoid the issue of ballot spoilage AND fraud as the cause of the discrepancies.

And I have been proven right. The INITIAL exit polls, both state and national, are correct. The FINAL, adulterated exit polls, are not.

When you say "TIA admitted as much" it reminds me of the banned DU naysayer who once said "Even TIA admits he's not an expert".

What a cheap shot. What a strawman.

Coincidence...or conspiracy?

Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Thu Feb-24-05 12:34 AM
Response to Reply #25
26. What would you call it then .....
Your base assumtions were wrong. Apples and Oranges.
Printer Friendly | Permalink |  | Top
 
kiwi_expat Donating Member (526 posts) Send PM | Profile | Ignore Tue Feb-22-05 09:10 PM
Response to Reply #14
17. Yes, Kerry "may very well" have won the popular vote.
"Fact 6:
The preliminary exit polls may very well reflect voter intent.
Fact 7:
The 2004 Preliminary National Exit Poll showed that Kerry won."




Assuming that "preliminary exit polls" are demographically-weighted (and jkd points out that they might not be), I agree with Fact 6, which leads to the conclusion that Kerry "may very well" have won (the popular vote). But doesn't prove he did.

Of course, the only exit polls that really matter are those for Ohio.

Printer Friendly | Permalink |  | Top
 
Chi Donating Member (921 posts) Send PM | Profile | Ignore Wed Feb-23-05 02:10 PM
Response to Reply #12
19. I have a question...
Heya Aussie <waves>

Where did you get the 'pristine' numbers for 2000 and 1996?

"b-"Clinton: 52.2% Dole: 37.5%" = "pristine" 1996

b-"Gore won 48.5% Bush: 46.2%" = "pristine" 2000"

I had assumed (I know, my bad) they were not available.
I figured since pollsters cover their butt by adjusting to the vote tally, why show everyone how far off they were initially.

Thanx in Advance
Chi
Printer Friendly | Permalink |  | Top
 
kiwi_expat Donating Member (526 posts) Send PM | Profile | Ignore Wed Feb-23-05 07:09 PM
Response to Reply #19
20. The "pristine" exit poll numbers are from Ruy Teixeira's analysis...

www.washingtonmonthly.com/archives/individual/2004_11/0...

Year/Exit Poll Results/Dem Lead/Dem Actual

1988/Dukakis: 50.3% Bush: 49.7%/+0.6%/ -7.7%

1992/Clinton: 46% Bush: 33.2%/+12.8%/+5.6%

1996/Clinton: 52.2% Dole: 37.5%/+14.7%/+8.5%

2000/Gore: 48.5% Bush: 46.2%/+2.3%/+0.5%


Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Feb-23-05 07:59 PM
Response to Reply #19
21. That's the question, now, isn't it? WHY? Why not just say ...
Edited on Wed Feb-23-05 08:02 PM by TruthIsAll
the discrepancy was due to "spoiled", mostly Democratic, votes?
Without suggesting HOW it happened.
After all, it happens every time.

WHY DON'T THE EXIT POLLSTERS JUST SAY IT?
Printer Friendly | Permalink |  | Top
 
Chi Donating Member (921 posts) Send PM | Profile | Ignore Thu Feb-24-05 02:25 PM
Response to Reply #21
28. A good test of that theory might be.....
A side-by-side comparison of vote spoilage (in # of votes) to the actual adjustment of the exit polls (in # of votes).
Too bad that info isn't available for previous years.

Just a thought.
Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Wed Feb-23-05 10:47 PM
Response to Reply #19
24. What Kiwi said was true.... and
Mistwell pointed them out to TIA several weeks ago.

They are easily obtainable on the web and I seems that there has been no effort made to hide them.

IMHO, there are far to many unknown variables relating to the exit polling practices and procedures to offer anything but speculation.

Mistwell posted this a few weeks ago and I thought it was straight forward, contained tons of information, and that it was presented in what seemed to be an unbiased fashion:

http://www.mysterypollster.com/main /
Printer Friendly | Permalink |  | Top
 
Chi Donating Member (921 posts) Send PM | Profile | Ignore Thu Feb-24-05 02:17 PM
Response to Reply #24
27. Ahhh...we're talking 'raw' numbers...
I did not know what you meant by 'pristine' (not criticizing).
I thought you were saying they were the late night numbers, the ones Mystery Pollster says are the best to use (post election, not yet adjusted to the vote tally).
Which are not available, that I know of, for previous years.

Actually Mistwell had posted them before that, and insisted that they were comparable to weighted/un-adjusted numbers 0f 2004. It took me a few attempts to get her to read Ruy Teixeira's analysis that I kept quoting for her.
"Consider this. The unweightedcompletely unweighteddata from the last four presidential
elections before this year are as follows:"
<shrug>

Thanx for the link, I ended up reading quite a bit there. But I did have a question on something that stuck out while I was reading.
Here it is....

"Warren Mitofsky, who ran the 2004 exit poll operation along with partner Joe Lenski, wrote the following in the Spring 2003 issue of Public Opinion Quarterly (p. 51):

An inspection of within-precinct error in the exit poll for senate and governor races in 1990, 1994 and 1998 shows an understatement of the Democratic candidate for 20 percent of the 180 polls in that time period and an overstatement 38 percent of the time...the most likely source of this error is differential non-response rates for Democrats and Republicans:"

Then I read this...

(MPollster is taking about the recently released raw data from Mitofski)
"The data do no disclose the actual precincts sampled.

In response to an email query, an Edison Mitofsky spokesperson referred me to the following passage from the Code of Professional Ethics and Practices of the American Association for Public Opinion Research (AAPOR):

"Unless the respondent waives confidentiality for specified uses, we shall hold as privileged and confidential all information that might identify a respondent with his or her responses.""

To me, I see Mitofski willing to access for his report, the same information (different year) he later says his 'code of ethics' does not allow to be disclosed (10 months later).

I was just wondering what your thoughts on that might be.

Thanx in Advance
Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Thu Feb-24-05 03:37 PM
Response to Reply #27
30. Well ....
The process is not transpaent in the least.

Trying to get a handle on this combined with all of these diversions, like TIA's threads, drive me crazy.

Mitifsky's withholding of "proprietary" information makes it difficult to get a handle on the process...... just like the programmers of the electronic voting machines.

The limited nature if this information leaves the door open to rampant speculation.

That's my opinion only ....... but I would like to see a rational thread started discussing exit polling however I don't have enough posts to open a thread.



Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Feb-24-05 05:17 PM
Response to Reply #30
31. So you would like to start a rational thread regarding exit polling?
Edited on Thu Feb-24-05 05:25 PM by TruthIsAll
That is a very presumptuous statement on your part.
And you call my threads diversions?

I suggest you read the papers of Freeman, Baiman and Uscountvotes.org. You will find that their analyses incorporates equivalent mathematics as I've used to compute exit poll probabilities.

HAVE YOU READ THEIR 5-PAGE PAPER WHICH DEBUNKS THE MITOFSKY RELUCTANT BUSH RESPONDER THEORY?

DO YOU APPRECIATE THAT THE PAPER WAS PEER-REVIEWED BY MATHEMATICS/STATISTICS DEPT. HEADS FROM RENOWNED UNIVERSITIES WHO HAVE PLACED THEIR REPUTATIONS ON THE LINE FOR DEMOCRACY?

No one is preventing you from posting analysis right now. So take the next step. Show us what you've got.

I for one am not convinced that your level of mathematics is up to snuff. The esteemed Mystery Pollster, worshiped by the naysayers, a DEMOCRAT no less, asked for help regarding the probabilty calculations in one of my posts. He has never been able to refute anything.

So why don't you tear apart my posts, whether it involves pre-election state, pre-election national, state exit or national exit poll?

And could you give us your thoughts on the following:

1) What do you believe is the appropriate MOE for each of the poll categories?

2)The Reluctant Bush responder theory.

3) Why would 99%+ of voting machine anomalies all favored Bush?

4) Ballot spoilage. What is your understanding of the average number of ballots which are never counted each election cycle?

5) Florida 2000. What can you tell us about the 70,000 undervotes and the 110,000 overvotes? Any idea how many voters were disenfranchised? The Palm Beach butterfly ballot? What about those 16,022 lost Gore votes in Volusia? Do you know of the Diebold connection?







Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Thu Feb-24-05 05:45 PM
Response to Reply #31
32. Answers.....
Your threads are nearly impossible to post in because you take such a ridged stance trying to defend your spreadsheets relevant or not.

Now you are using these questions to divert the attention (strawman)from your now admittedly biased calculations.

You pick the result you want to achieve and then work the math that assures the closest outcome. Then you act like you have solved a big mystery. WOW.

Now I will address your strawmen:

"YOU HAVE NEVER ANSWERED MY DIRECT QUESTION:
HAVE YOU READ THEIR 5-PAGE PAPER WHICH DEBUNKS THE MITOFSKY RELUCTANT BUSH RESPONDER THEORY?"

Yes.

"DO YOU APPRECIATE THAT THE PAPER WAS PEER-REVIEWED BY MATHEMATICS/STATISTIC DEPT. HEADS FROM RENOWNED UNIVERSITIES WHO HAVE THEIR REPUTATIONS ON THE LINE FOR DEMOCRACY?"

Yes. Do they sancition your calculations?

"SO TELL US, WHAT WILL YOU BRING TO THE TABLE IN THIS EFFORT?"

Certainly not flawed calculations.

"I, for one, am not convinced that your level of mathematics is up to snuff."

Never said it was. I do know that if you start with flawed assumptions then your work is worthless.

"1) What are your thoughts on the appropriate MOE for each of the poll categories?"

I would like to learn more about the mechanics used during the exit polling process.

"2)Tell us about the Reluctant Bush responder theory."

It's a theory that states repubs were less likely to participate in the exit poll. Pretty simple concept.

"3)Tell us why 99%+ of voting machine anomalies all favored Bush."

I don't support electronic voting. However, 89 votes do not make or break an election.

"4) Then tell us your thoughts regarding ballot spoilage. What is your understanding of the average number of ballots which are never counted each election cycle?"

Many ballots are spoiled for a wide range of reasons.

"5) Then there's Florida 2000. What can you tell us about the 70,000 undervotes and the 110,000 overvotes? Any idea how many voters were disenfranchised? The Palm Beach butterfly ballot? What about those 16,022 lost Gore votes in Volusia? Do you know of the Diebold connection?"

Certainly valid issues.






Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Feb-24-05 06:18 PM
Response to Reply #32
33. Ok, you've told is what you believe.
Edited on Thu Feb-24-05 07:15 PM by TruthIsAll
My gut feeling is correct.

You may very well be a reincarnation of those who have taken your stance before.

"Your threads are nearly impossible to post in because you take such a ridged stance trying to defend your spreadsheets relevant or not"

Me.
OK, be specific and point them (all of them) out.

"Now you are using these questions to divert the attention (straw man)from your now admittedly biased calculations".

Me.
Biased calculations? Now you really give yourself away.
Name them.
Is it the MOE?
Is it the exit poll data I have used?
Is it the calculations?

Just where is the bias?
You are the one whose bias is now clear.


"You pick the result you want to achieve and then work the math that assures the closest outcome. Then you act like you have solved a big mystery. WOW."

Me.
Is that so?
Seems I've heard that one before.
In point of fact, I have only sought to improve my analyses, by using the latest available data and asking mathematicians at DU to confirm or refute the probabilities . Where an error was found, either by myself or others, I have immediately fixed and posted it.

On the contrary, you appear to be the one who has an agenda.
You will find that it will be quickly exposed, not just by me, but by others at DU, who have extensive experience to draw upon.
We have heard all the excuses, seen all the straw men, waded through the fog.

Do you realize what you are getting yourself into?


You:
Now I will address your strawmen:

"YOU HAVE NEVER ANSWERED MY DIRECT QUESTION:
HAVE YOU READ THEIR 5-PAGE PAPER WHICH DEBUNKS THE MITOFSKY RELUCTANT BUSH RESPONDER THEORY?"

Yes.

Me.
A one line response.
This is NOT a court of law.
This is a give and take discussion.

You:
"DO YOU APPRECIATE THAT THE PAPER WAS PEER-REVIEWED BY MATHEMATICS/STATISTIC DEPT. HEADS FROM RENOWNED UNIVERSITIES WHO HAVE THEIR REPUTATIONS ON THE LINE FOR DEMOCRACY?"

Yes. Do they sanction your calculations?

Me.
They don't need to sanction them. The fact that they come up with the same probabilities is enough.
Do you sanction their results?


"SO TELL US, WHAT WILL YOU BRING TO THE TABLE IN THIS EFFORT?"

Certainly not flawed calculations.

Me.
Of course you won't.
BECAUSE YOU CAN'T OFFER ANY ANALYSIS.
YOU CAN'T DO THE MATH.


"I, for one, am not convinced that your level of mathematics is up to snuff."

Never said it was. I do know that if you start with flawed assumptions then your work is worthless.

Me.
Oh, boy, another one.
Just what I thought.
Your mask is off.
What is it that you find in my any of my posts to be flawed?
The 3 national exit polls. Mixing apples and oranges?

I have already admitted my error in comparing them, and when I became aware of the pre-adjusted polling data, they only served to strenghten my analytical sense - and which has never wavered.

THE ORIGINAL, "PRISTINE" EXIT POLL OF 13,047, WHICH PRECEDED THE CONTAMINATED COOK-BOOK POLL OF 13,660 (MATCHED TO BOGUS VOTE COUNTS), REPRESENTS THE TRUTH.

DO YOU GET IT YET?

Be specific. Review my posts.
It's very easy to say it.
NOW DO IT.
ONCE AGAIN: BE SPECIFIC.
POINT THEM OUT. ALL OF THEM.

"1) What are your thoughts on the appropriate MOE for each of the poll categories?"

I would like to learn more about the mechanics used during the exit polling process.


Me.
You would like to learn more?
At this stage of the game?
And just how will you go about doing that?
Are you going to put on your training wheels now?



"2)Tell us about the Reluctant Bush responder theory."

It's a theory that states repubs were less likely to participate in the exit poll. Pretty simple concept.

Me.
Pretty simple concept.
Is that all you have to say?
No opinion.
Just a simple concept.
I see you are really taking a stance here.

Why are you not critical of Mitofsky's assumptions, as you are of mine?

Mitofsky has not a modicum of data to support that theory.
In fact, the data in his report reveals just the opposite.
Republicans were MORE inclined to respond, not less.

I thought you just said that you read the report from Uscountvotes.org.
Did you?

"3)Tell us why 99%+ of voting machine anomalies all favored Bush."

I don't support electronic voting. However, 89 votes do not make or break an election.

Me.
Once again, you reveal yourself as mathematically weak.
89 votes? Is that your answer?
It shows that you are unaware of the 50,000 recorded election "incident"s which are available online.

Do you know the probability that 89 of 91 votes would switch from Gore to Bush? Would you even know how to compute it?
Or do you even care?

And what about the other 49,900 incidents, of which 99% worked in favor of Bush?
Any comment?


"4) Then tell us your thoughts regarding ballot spoilage. What is your understanding of the average number of ballots which are never counted each election cycle?"


Many ballots are spoiled for a wide range of reasons.

Me.
There you go again.
I asked you for a number.
Avoiding the question.

"5) Then there's Florida 2000. What can you tell us about the 70,000 undervotes and the 110,000 overvotes? Any idea how many voters were disenfranchised? The Palm Beach butterfly ballot? What about those 16,022 lost Gore votes in Volusia? Do you know of the Diebold connection?"

Certainly valid issues.

Me.
Another one liner.
Don't you enjoy the give and take?
You call them valid "issues" as if they warrant discussion/debate.
These are facts which are not in dispute.

Thanks for your responses, as limited as they may be.
They speak volumes.
We now know each other much better.
Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Thu Feb-24-05 06:28 PM
Response to Reply #33
34. Thank you for providing an example so quickly ...
Edited on Thu Feb-24-05 06:28 PM by Aussie_expat
The above post illustrates my point.

As I said before:

Your threads are nearly impossible to post in because you take such a ridged stance trying to defend your spreadsheets relevant or not.

Now you are using these questions to divert the attention (strawman)from your now admittedly biased calculations.



Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Feb-24-05 06:45 PM
Response to Reply #34
35. Very quick response. No, I am using these questions to expose
your lack of credibility.

And I have succeeded.
Printer Friendly | Permalink |  | Top
 
Aussie_expat Donating Member (124 posts) Send PM | Profile | Ignore Thu Feb-24-05 07:26 PM
Response to Reply #35
36. What do you want me to do?
Edited on Thu Feb-24-05 07:37 PM by Aussie_expat
Scream fraud in every breath?

I have said repeatedly that I believe fraud has occured.

I believe that electronic voting is a tragic option.

Since you are painting a picture of me to suit your beliefs let me tell you a few other things that I believe:

I oppose the "War on Terror"

I support nearly any legislation that looks after the working class.

I am for tax reform that places a fair burden on the wealthy.

I am for strong unions.

I think that abortion is a decision that a woman has to make.

I think that corporate profits should not be more important than workers rights.

Just because I don't like agree with your work doesn't give you the right to attempt to label me.





Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Feb-24-05 02:30 PM
Response to Reply #24
29. Mull over this
	Initial 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

Est. "spoiled" votes	3.00%				
Avg Rep Dev.		3.42%					
Avg Dem Dev.		-2.61%						
Printer Friendly | Permalink |  | Top
 
BillyDoc Donating Member (115 posts) Send PM | Profile | Ignore Tue Feb-22-05 03:02 PM
Response to Original message
11. Actually, there were probably more Bush responders
The final report from Edison/Mitofsky is out, and you can find it by going to ElectionFraudBounty.org and clicking on the "Evidence" page. It is linked about half-way down the page.

More interestingly, there is a response to the Edison/Mitofsky report that is linked to on the same page, along with a graph that shows that their claim that "More Kerry supporters responded" is simply not supported by their own data. In fact, it was the opposite.

Which leaves, what? FRAUD, do you think?
Printer Friendly | Permalink |  | Top
 
DU AdBot (1000+ posts) Click to send private message to this author Click to view 
this author's profile Click to add 
this author to your buddy list Click to add 
this author to your Ignore list Fri Jul 28th 2017, 03:26 PM
Response to Original message
Advertisements [?]
 Top

Home » Discuss » Topic Forums » Election Reform Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators


Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC