Pre-election and Exit Poll Summaries - Democratic Underground

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 10:25 AM
Original message

Pre-election and Exit Poll Summaries

Edited on Mon Jan-24-05 11:01 AM by TruthIsAll

		                 Total	Total	2-Party Prop. Alloc.		
Pre-election Polls		Kerry	Bush	Kerry	Bush	
National 18 Avg		        47.17	46.89	50.15	49.85	
National 9 Indep Avg		47.11	46.56	50.29	49.71	
Weighted State Polls		47.88	46.89	50.52	49.48	
						
Average		                47.39	46.78	50.32	49.68	
						
						
Post-election Exit polls	Kerry	Bush	Kerry	Bush	
WP/NEP 13074 		         50.8	48	51.42	48.58	
State		                 NA	NA	50.39	49.61	
						
Average				               50.90	49.10	
						
Bush 11 poll average pre-election approval		48.5

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 11:45 AM
Response to Original message

1. Of the last 50 public opinion polls conducted, Kerry only led in 6

(courtesy of tritsofme)

GW/Battleground (1000 LV) 10/31 - 11/1 50% 46% Bush +4
Rasmussen (3,000 LV) 10/30 - 11/1 50.2% 48.5% Bush +1.7
TIPP (936 LV) 10/30 - 11/1 46.9% 44.3% Bush +2.6
FOX News (1200 LV) 10/30 - 10/31 46% 48% Kerry +2
CNN/USAT/Gallup (1573 LV) 10/29 - 10/31 49% 47% Bush +2
CBS/NY Times (643 LV) 10/28 - 10/30 50% 47% Bush +3
ARG (1258 LV) 10/28 - 10/30 48% 49% Kerry +1
Newsweek (882 LV) 10/27 - 10/29 51% 45% Bush +6
Battleground (1000 LV) 10/25 - 10/28 51% 46% Bush +5
CNN/USAT/Gallup (1195 LV) 10/22 - 10/24 52% 46% Bush +6
Los Angeles Times (881 LV) 10/21 - 10/24 49% 48% Bush +1
Newsweek (880 LV) 10/21 - 10/22 48% 47% Bush +1
Time (803 LV) 10/19 - 10/21 52% 47% Bush +5
GW/Battleground (1000 LV) 10/18 - 10/21 49% 45% Bush +4
Rasmussen (3,000 LV) 10/17 - 10/19 48% 47% Bush +1
FOX News (1000 LV) 10/17 - 10/18 48% 43% Bush +5
CBS News (678 LV) 10/14 - 10/17 47% 46% Bush +1
CNN/USAT/Gallup (788 LV) 10/14 - 10/16 52% 44% Bush +8
Time (865 LV w/leaners) 10/14 - 10/15 48% 48% TIE
Newsweek (LV) 10/14 - 10/15 50% 45% Bush +5
GW/Battleground (1000 LV) 10/11 - 10/14 49% 46% Bush +3
CBS News (760 LV) 10/9 - 10/11 47% 46% Bush +1
ICR (763 LV) 10/9 - 10/11 49% 46% Bush +3
CNN/USAT/Gallup (793 LV) 10/9 - 10/10 48% 50% Kerry +2
Rasmussen (3,000 LV) 10/7 - 10/9 50% 46% Bush +4
Time (886 LV w/leaners) 10/6 - 10/7 47% 46% Bush +1
GW/Battleground (1250 LV) 10/3 - 10/7 49% 46% Bush +3
Fox News (1000 LV) 10/3 - 10/4 48% 45% Bush +3
ICR (762 LV)** 10/1 - 10/5 51% 46% Bush +5
ARG (800 LV) 10/2 - 10/4 46% 47% Kerry +1
CBS/NYT (561 LV) 10/1 - 10/3 48% 47% Bush +1
Zogby (1036 LV) 10/1 - 10/3 46% 45% Bush +1
CNN/USAT/Gallup (772 LV) 10/1 - 10/3 49% 49% TIE
Newsweek (1013 RV) 9/30 - 10/2 46% 49% Kerry +3
Battleground (1000 LV) 9/27 - 9/30 51% 44% Bush +7
LA Times (1100 LV) 9/25 - 9/28 51% 46% Bush +5
CNN/USAT/Gallup (758 LV) 9/24 - 9/26 52% 44% Bush +8
IBD/TIPP (649 LV) 9/22 - 9/27 45% 46% Kerry +1
Time (877 LV) 9/21 - 9/23 49% 43% Bush +6
FOX News (1000 LV) 9/21 - 9/22 45% 43% Bush +2
Battleground (1000 LV) 9/20 - 9/23 50% 45% Bush +5
CBS News (931 LV) 9/20 - 9/22 50% 41% Bush +9
Zogby (1066 LV) 9/17 - 9/19 47% 44% Bush +3
IBD/TIPP (650 LV) 9/14 - 9/18 46% 43% Bush +3
CNN/USAT/Gallup (767 LV) 9/13 - 9/15 55% 42% Bush +13
CBS News (1088 RV) 9/12 - 9/16 50% 42% Bush +8
Battleground (1000 LV) 9/12 - 9/15 49% 45% Bush +4
IBD/TIPP (674 LV) 9/7 - 9/12 47% 47% TIE
Newsweek (1003 RV) 9/9 - 9/10 50% 45% Bush +5
Zogby (1018 LV) 9/8 - 9/9 47% 45% Bush +2
Time (857 LV) 9/7 - 9/9 54% 42% Bush +12

Printer Friendly | Permalink | | Top

FogerRox

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 11:49 AM
Response to Reply #1

2. An incumbant gets his May/june approval rating in NOV

Bush was at 44%. Peace time--war time, dont matter it always works out.

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 11:52 AM
Response to Reply #2

3. Have you ever seen the warning on

things like mutual funds: "Past history does not guarantee future results"? A lot of people seem to forget it.

Printer Friendly | Permalink | | Top

FogerRox

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 12:06 PM
Response to Reply #3

4. oh you mean that casino--- Head to head polls are problematic

Edited on Mon Jan-24-05 12:09 PM by FogerRox

Untill 'tween the last debate and election day. May june approval always works--untill now. statistically improbable, me thinks.

The sun rises every morning---does past performance indicate---... JOKING

Printer Friendly | Permalink | | Top

Carolab

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 03:45 PM
Response to Reply #3

15. Mutual funds & voting. I don't get the connection. Poor analogy. n/t.

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 05:52 PM
Response to Reply #15

27. In anything that is "fuzzy science" -

such as statistics, meteorology, stock market predictions, or election predictions, the statements such as "if the candidate leads in july, he will win in november" or "if the candidate's approval rating is below 50, he loses", or "if the AFC football team wins the superbowl, the incumbent wins" etc. are not physical laws. They are based on the past. And they are true statements only until they are not true. That is why "the past performance does not predict the future results" is the statement that you and TIA should pay a bit of attention to.

Printer Friendly | Permalink | | Top

Carolab

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 09:25 PM
Response to Reply #27

48. How about:

Edited on Mon Jan-24-05 09:26 PM by Carolab

"Insanity is doing the same thing over and over again and expecting a different result."

Like voting for George Bush, TWICE.

This election was more indicative of past performance not predicting future behavior, from that standpoint.

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 09:29 PM
Response to Reply #48

49. Ah the "how can they be such idiots" argument -

according to which the fact that official results say Bush won is a proof of fraud all by itself.

At least that argument does not need poll cherrypicking to back it up.

Printer Friendly | Permalink | | Top

Carolab

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 09:54 PM
Response to Reply #49

54. Right.

More a priori than a posteriori.

Printer Friendly | Permalink | | Top

RaulVB

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 04:01 PM
Response to Reply #3

19. That was a bad talking point, "friend." (n/t)

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 01:04 PM
Response to Reply #1

5. Once again, tritsofme..totally biased to LV's only. I stand by my numbers

Edited on Mon Jan-24-05 01:08 PM by TruthIsAll

All I see is clutter and fog. As if duplicate polls count.
I used the FINAL 18 individual National Polls, a mix of RV's
and LV's.

For the upteenth time, RV's made more sense in this election
due to the large number of new (young, Democratic) women
voters for Kerry. These were RV's, not LV's.

Why is it so hard for you to accept that?
Why are you trying so hard to camouflage that fact?

Try as you might, you can't change the TRUTH.

I smell agenda here.


				LATEST NATIONAL POLLS				
								
						Kerry projections	
Nat.	Poll	Poll	Curr. 	Curr. 	Curr	Undecided alloc. % 		
Poll	Date	MoE	Kerry	Bush	K/K+B	60.0	67.0	75.0
								
AP	1020	3.50	49	46	51.6	52.0	52.4	52.8
Time	1021	3.00	46	51	47.4	47.8	48.0	48.3
LAT	1024	4.00	48	47	50.5	51.0	51.4	51.8
Harris	1025	2.50	48	47	50.5	51.0	51.4	51.8
ICR	1026	3.10	44	46	48.9	50.0	50.7	51.5
								
Econ	1027	2.00	49	45	52.1	52.6	53.0	53.5
Nwk	1029	4.00	45	48	48.4	49.2	49.7	50.3
Pew	1030	3.50	46	45	50.5	51.4	52.0	52.8
Zogby	1030	3.10	47	48	49.5	50.0	50.4	50.8
ARG	1030	3.50	49	48	50.5	50.8	51.0	51.3
								
ABC	1030	3.00	48	47	50.5	51.0	51.4	51.8
Marist	1031	2.50	49	48	50.5	50.8	51.0	51.3
TIPP	1031	3.50	44	45	49.4	50.6	51.4	52.3
CBS	1031	3.00	46	47	49.5	50.2	50.7	51.3
DemC	1031	3.10	48	47	51.6	52.2	52.7	53.3
								
FOX	1031	3.40	48	45	50.5	51.0	51.4	51.8
Gallup	1031	4.00	48	46	51.1	51.6	52.0	52.5
NBC	1031	3.00	47	48	49.5	50.0	50.4	50.8
								
	Average	0.73	47.17	46.89	50.15	50.73	51.15	51.63
								
	Kerry Win Probabilities		65.17	97.55	99.90	100.00

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 01:07 PM
Response to Reply #5

6. You cherrypick the polls -

then you can get whatever conclusions you want from them.

Printer Friendly | Permalink | | Top

Imnottelling

(22 posts) Send PM | Profile | Ignore

Mon Jan-24-05 05:20 PM
Response to Reply #6

23. Do us a favor

Go over to www.electoral-vote.com and check out the guy's website. There is a lot of information on all the pre-election polls and pre-election polling in general.

I can't stand reading these pointless "my data is better than your data" arguments.

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 06:13 PM
Response to Reply #23

31. Ok, I went to the guy's web site:

Ok, let's see: from http://www.electoral-vote.com/info/polls-2004.html

Only Registered Voters polls:



Poll                Date                Kerry   Bush    
=====================================================
American Res. Grp   Oct 28 - Oct 30     49      48    	
Gallup              Oct 29 - Oct 31     48      46      
Opinion Dynamics    Oct 30 - Oct 31     47      45      
Los Angeles Times   Oct 21 - Oct 24     47      47      
Marist Coll.        Oct 31 - Oct 31     48      48      
Newsweek            Oct 27 - Oct 29     44      48      
Pew                 Oct 27 - Oct 30     45      45      
Time                Oct 19 - Oct 29     43      50      
=====================================================
Average                                46.375  47.125 


Only Likely Voters:

Poll                Date                Kerry   Bush
=====================================================
American Res. Grp   Oct 28 - Oct 30     49      48  
CBS News            Oct 28 - Oct 31     46      47  
CBS/NYT             Oct 28 - Oct 30     47      50
Gallup 	USA Today   Oct 29 - Oct 31     47      49
Opinion Dynamics    Oct 30 - Oct 31     48      46
Greenberg           Oct 26 - Oct 28     49      46
Harris              Oct 29 - Oct 31     45      49
Los Angeles Times   Oct 21 - Oct 24     48      48
Marist Coll.        Oct 31 - Oct 31     49      48
NBC/WSJ             Oct 29 - Oct 31     47      48
Newsweek            Oct 27 - Oct 29     44      50
Zogby Reuters       Oct 11 - Oct 13     45      46
Zogby Time          Oct 11 - Oct 13     46      51
YouGov              Oct 29 - Nov 1      50      47
=====================================================
Average                                47.142  48.07


Either way you take it, Bush is ahead in pre-election polls - IF you do not cherrypick the polls, like TIA does.

Printer Friendly | Permalink | | Top

righteous1

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 07:45 PM
Response to Reply #31

37. My findings as well. "Truth is all" as long as it can be tortured to

make a (his) point

Printer Friendly | Permalink | | Top

righteous1

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 07:57 PM
Response to Reply #31

38. Here's another of interest

Edited on Mon Jan-24-05 08:02 PM by righteous1

http://www.pollingreport.com/2004.htm B 49.54% K 48.76% additionally Final Trial Heat B 48.47% K 47.02% Final total of all major polls accept Rassmussen B 48.72 K 47.80

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 08:36 PM
Response to Reply #31

40. Nice try, but...

Edited on Mon Jan-24-05 08:37 PM by TruthIsAll

1. You don't match my 18 polls. 
You conveniently forgot Fox and AP							

2. Nine (9) polls don't agree. 
Are you point shaving?				

3. At least 6 of mine are LV polls 
Don't use that RV strawman.

4. You get an incomplete
Maybe you should get more help next time.
	
								My 18 Polls		
			Poll	Kerry	Bush			Kerry	Bush	Agree?
1	NA	AP	-	-	-		1020	49	46	
2	NA	TIPP	-	-	-		1031	44	45	
3	NA	FOX	-	-	-		1031	48	45	
4	NA	ABC	-	-	-		1030	48	47	
5	1031	NBC	LV	47	48		1031	47	48	
6	1030	ARG	LV	49	48		1030	49	48	
7	1031	GallupRV	48	46		1031	48	46	
8	1031	Op		47	45		1026	44	46	NO
9	1024	LAT	LV	47	47		1024	48	47	NO
10	1031	Marist            49   48		1031	49	48	
11	1029	Nwk	RV	44	48		1029	45	48	NO
12	1030	Pew		45	45		1030	46	45	NO
13	1021	Time		43	50		1021	46	51	NO
14	1028	Greenberg        49	46		1031	48	47	NO?
15	1031	CBS	LV	46	47		1031	46	47	
16	1031	Harris           45   49		1025	48	47	NO
17	1013	Zogby	LV	45	46		1030	47	48	NO
18	1101	YouGovLV	50	47		1029	49	45	NO
				46.69	47.08			47.17	46.89

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 08:39 PM
Response to Reply #40

41. You cherry-picked your 18 polls -

Edited on Mon Jan-24-05 08:41 PM by qwghlmian

of course they don't agree.

I just took the whole page http://www.electoral-vote.com/info/polls-2004.html

If you think the guy is fudging the numbers on that page, say so.

And as for LV vs RV, as you may have noticed have you read my post, I calculated the average of RVs and of LVs separately, and both disagree with your conclusions.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 09:33 PM
Response to Reply #41

50. YOU CONVENIENTLY FORGOT FOX AND AP, CHERRY-PICKER

Jeb Bush can sure use you in 2008.

You are good at hiding the votes.

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 09:35 PM
Response to Reply #50

52. Once again - someone suggested the

www.electoral-vote.com website, so I took his page there

http://www.electoral-vote.com/info/polls-2004.html

at least I took the whole thing and did not try to pick and choose the polls on that page. Something you should learn from.

Printer Friendly | Permalink | | Top

tritsofme

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 09:37 PM
Response to Reply #50

53. Bbbut I thought Fox was one of the "whores"

Wasn't AP-Ipsos in that group too, depending on what their data said?

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 01:10 PM
Response to Reply #5

7. You cherry-picked the LV's. I don't see ONE RV on your list. Just dupes.

Edited on Mon Jan-24-05 01:12 PM by TruthIsAll

What's the matter?
Can't handle the truth?

FYI, I calculated the average of these polls in The Election Model BEFORE the election. I forecast using these polls.

And I was RIGHT.

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 01:12 PM
Response to Reply #7

8. Ok - go ahead and post all the RV numbers for

Edited on Mon Jan-24-05 01:14 PM by qwghlmian

the same polls - since you're not afraid of "the truth".

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 01:18 PM
Response to Reply #8

9. OK, you asked for it

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 01:35 PM
Response to Reply #9

11. I asked for a simple thing -

you seem not to have understood the question.

Please give the RV numbers corresponding to the LV numbers that I posted.

I hope this is clearer.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 03:55 PM
Response to Reply #11

17. You have the list. What else do you need? Just compare them.

Why should I do your work for you?

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 06:14 PM
Response to Reply #17

32. I gave you a long list of LV polls -

if you claim that RV is a better way to go, provide an RV number for every one of those polls (which you did not) and we will compare.

See post #31 - that is what happens when you do not cherrypick the polls.

Printer Friendly | Permalink | | Top

Karenca

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 07:01 PM
Response to Reply #32

36. Oh jeez......

you are so wrong!!

Anyway, what's the motive behind posting that false information, when the truth is that Kerry was leading in all the polls, before the election AND the exit polls?

KERRY!!! KERRY!!!!

KERRYY!! KERRY!!!

KERRY!!! KERRY!!!

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

Btw, B*sh is a really some piece of shit. :evilfrown: :evilgrin: :evilfrown:

Printer Friendly | Permalink | | Top

tritsofme

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 08:43 PM
Response to Reply #36

42. What false information?

Post #1 shows that you are clearly wrong.

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 08:44 PM
Response to Reply #42

43. You don't understand the "code", you see -

"false information" in the code means "anything that contradicts TIA's numbers". You have to know the code to understand these posts.

Printer Friendly | Permalink | | Top

Karenca

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 09:11 PM
Response to Reply #42

46. No, You've got it wrong....

Who cares about the Oct. and Sept. polls????? :boring:
We all knew what the polls were after that obscene display of false patriotism in the name of god (really in the name of oi) after the RNC.
Those are not relative, so why bother???

KERRY
KERRY
KERRY!!!!!:dem: :dem: :dem:

KERRY !
KERRY !
KERRY !!!!!

Printer Friendly | Permalink | | Top

tritsofme

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 09:34 PM
Response to Reply #46

51. Ok, of the last 12 polls conducted in the 10 days before the election

Kerry only led in 2.

GW/Battleground (1000 LV) 10/31 - 11/1 50% 46% Bush +4
Rasmussen (3,000 LV) 10/30 - 11/1 50.2% 48.5% Bush +1.7
TIPP (936 LV) 10/30 - 11/1 46.9% 44.3% Bush +2.6
FOX News (1200 LV) 10/30 - 10/31 46% 48% Kerry +2
CNN/USAT/Gallup (1573 LV) 10/29 - 10/31 49% 47% Bush +2
CBS/NY Times (643 LV) 10/28 - 10/30 50% 47% Bush +3
ARG (1258 LV) 10/28 - 10/30 48% 49% Kerry +1
Newsweek (882 LV) 10/27 - 10/29 51% 45% Bush +6
Battleground (1000 LV) 10/25 - 10/28 51% 46% Bush +5
CNN/USAT/Gallup (1195 LV) 10/22 - 10/24 52% 46% Bush +6
Los Angeles Times (881 LV) 10/21 - 10/24 49% 48% Bush +1
Newsweek (880 LV) 10/21 - 10/22 48% 47% Bush +1

Printer Friendly | Permalink | | Top

Karenca

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 09:56 PM
Response to Reply #51

55. The ONLY pre-election polls that mattered were the

ones taken immediately before election day.
The others don't mean a damn fucking thing.
So, Why bother?

KERYY KERRY KERRY
KERRY KERRY KERRY!!!!!!
:dem: :dem: :dem: :dem: :dem: :dem: :dem: :dem: :dem: :dem: :dem:

Printer Friendly | Permalink | | Top

tritsofme

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 10:09 PM
Response to Reply #55

57. Are you for real?

90%+ of voters decided who they were going to vote for more than a week before the election.

Printer Friendly | Permalink | | Top

Karenca

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 10:57 PM
Response to Reply #57

60. I am

for real.
And you're all wrong about your polls......Since you have a need to go way back...From Electoral-Vote.com, A true compilation of ALL the entire polling places. KERRY 281 ELECTORAL VOTES
BUSH 257 NOV 2nd 2004.

State,EV,Kerry,Bush,Nader,Date,=10,5-9,<5,Tie,<5,5-9,=10,Poll source-duration
Alabama,9,40,58,1,Oct 27,,,,,,,9,Kerry 1% Bush 1%
Alaska,3,37,61,1,Sep 11,,,,,,,3,Kerry 7% Bush 4%
Arizona,10,42,57,0,Oct 30,,,,,,,10,Kerry 1% Bush 1%
Arkansas,6,47,51,1,Nov 1,,,,,6,,,Kerry 1% Bush 0%
California,55,55,44,0,Oct 31,55,,,,,,,Kerry 1% Bush 1%
Colorado,9,48,50,1,Nov 1,,,,,9,,,Kerry 1% Bush 0%
Connecticut,7,55,43,1,Oct 28,7,,,,,,,Kerry 3% Bush 1%
Delaware,3,55,43,1,Sep 25,3,,,,,,,Kerry 10% Bush 5%
D.C.,3,84,14,1,Sep 13,3,,,,,,,Kerry 6% Bush 3%
Florida,27,52,46,1,Oct 31,,27,,,,,,Kerry 3% Bush 2%
Georgia,15,43,56,0,Oct 31,,,,,,,15,Kerry 2% Bush 1%
Hawaii,4,50,49,0,Oct 20,,,4,,,,,Kerry 5% Bush 3%
Idaho,4,37,62,0,Sep 10,,,,,,,4,Kerry 7% Bush 3%
Illinois,21,55,44,0,Oct 31,21,,,,,,,Kerry 2% Bush 1%
Indiana,11,40,59,0,Oct 29,,,,,,,11,Kerry 1% Bush 1%
Iowa,7,51,47,1,Nov 1,,,7,,,,,Kerry 1% Bush 0%
Kansas,6,38,60,1,Oct 27,,,,,,,6,Kerry 1% Bush 0%
Kentucky,8,39,59,1,Oct 30,,,,,,,8,Kerry 1% Bush 0%
Louisiana,9,37,61,1,Oct 22,,,,,,,9,Kerry 5% Bush 3%
Maine,4,53,45,1,Oct 30,,4,,,,,,Kerry 1% Bush 1%
Maryland,10,55,43,1,Oct 29,10,,,,,,,Kerry 1% Bush 0%
Massachusetts,12,59,40,0,Oct 5,12,,,,,,,Kerry 9% Bush 4%
Michigan,17,52,46,1,Nov 1,,17,,,,,,Kerry 0% Bush 0%
Minnesota,10,52,46,1,Nov 1,,10,,,,,,Kerry 1% Bush 1%
Mississippi,6,45,53,1,Sep 17,,,,,,6,,Kerry 3% Bush 2%
Missouri,11,47,52,0,Oct 31,,,,,,11,,Kerry 0% Bush 0%
Montana,3,39,59,1,Oct 20,,,,,,,3,Kerry 3% Bush 2%
Nebraska,5,35,63,1,Oct 20,,,,,,,5,Kerry 3% Bush 2%
Nevada,5,45,53,1,Nov 1,,,,,,5,,Kerry 0% Bush 0%
New Hampshire,4,50,48,1,Oct 31,,,4,,,,,Kerry 1% Bush 0%
New Jersey,15,51,47,1,Oct 31,,,15,,,,,Kerry 9% Bush 5%
New Mexico,5,51,48,1,Nov 1,,,5,,,,,Kerry 0% Bush 0%
New York,31,58,40,1,Oct 28,31,,,,,,,Kerry 1% Bush 1%
North Carolina,15,46,53,0,Oct 31,,,,,,15,,Kerry 1% Bush 0%
North Dakota,3,40,58,1,Oct 19,,,,,,,3,Kerry 5% Bush 3%
Ohio,20,48,51,0,Oct 31,,,,,20,,,Kerry 1% Bush 1%
Oklahoma,7,35,64,0,Oct 30,,,,,,,7,Kerry 1% Bush 0%
Oregon,7,53,46,0,Oct 29,,7,,,,,,Kerry 3% Bush 2%
Pennsylvania,21,50,49,0,Nov 1,,,21,,,,,Kerry 1% Bush 1%
Rhode Island,4,56,42,1,Oct 27,4,,,,,,,Kerry 2% Bush 1%
South Carolina,8,44,54,1,Oct 31,,,,,,,8,Kerry 1% Bush 0%
South Dakota,3,40,58,1,Oct 24,,,,,,,3,Kerry 7% Bush 3%
Tennessee,11,40,58,1,Oct 30,,,,,,,11,Kerry 0% Bush 0%
Texas,34,39,60,0,Oct 28,,,,,,,34,Kerry 2% Bush 1%
Utah,5,27,71,1,Oct 28,,,,,,,5,Kerry 3% Bush 2%
Vermont,3,56,42,1,Oct 12,3,,,,,,,Kerry 3% Bush 2%
Virginia,13,48,51,0,Oct 29,,,,,13,,,Kerry 1% Bush 0%
Washington,11,52,46,1,Oct 31,,11,,,,,,Kerry 1% Bush 0%
West Virginia,5,46,52,1,Oct 29,,,,,,5,,Kerry 3% Bush 1%
Wisconsin,10,48,50,1,Oct 31,,,,,10,,,Kerry 3% Bush 2%
Wyoming,3,32,66,1,Sep 11,,,,,,,3,Kerry 3% Bush 1%
,,,,,Sums,149,76,56,0,58,42,157,538

Updated Nov. 02,,,,,,,,,,,,,
,,,Projected,,,Kerry,,,,Bush,,,Sum
,,,electoral votes,,,281,,,,257,,,538

http://www.electoral-vote.com/pred/nov02x.csv

Printer Friendly | Permalink | | Top

tritsofme

(1000+ posts) Send PM | Profile | Ignore

Tue Jan-25-05 12:58 AM
Response to Reply #60

66. I didn't have time go through all of that

But it looks the only thing that flipped from his prediction was Florida.

And IIRC Electoral-Vote.com just took the last poll conducted and used that on their webpage. (correct me if I'm wrong here)

An average of polls conducted the week before the election a dead heat with a slight advantage toward Bush.

RCP Average | 10/25-11/1 - - 48.2 47.6 1.0 Bush +0.6
ARG | 10/30-11/1 600 LV 4.0 48 50 1 Kerry +2
Zogby | 10/29-11/1 601 LV 4.1 48 48 - TIE
FOX News | 10/30-31 700 LV 3.0 44 49 1 Kerry +5
SurveyUSA | 10/29-31 742 LV 3.7 49 48 - Bush +1
Insdr Adv | 10/29-31 400 LV 5.0 48 48 1 TIE
Quinnipiac | 10/27-31 1098 LV 3.0 51 43 1 Bush +8
CNN/USAT/Gallup | 10/28-30* 1138 LV 4.1 47 50 - Kerry +3
Rasmussen | 10/25-31 600 LV 4.0 50 47 - Bush +3
Mason-Dixon | 10/27-29 625 LV 4.0 49 45 - Bush +4

Printer Friendly | Permalink | | Top

Karenca

(1000+ posts) Send PM | Profile | Ignore

Tue Jan-25-05 10:31 AM
Response to Reply #66

73. If you DO take the time to look, you will see that KERRY was leading.

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

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

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 11:16 PM
Response to Reply #51

61. Again. Cherry-picking. All LV polls. Of course. Jeez, you are too much.

Edited on Mon Jan-24-05 11:55 PM by TruthIsAll

In my list of 18 polls, which includes at least 7 LV's, Kerry led in ELEVEN (11).

In your list of 12, which includes NOT ONE RV, Kerry led in TWO.

You are so transparent.
You are making yourself look ridiculous.
Do you never get tired of your machinations?

All your contortions to make it appear that Bush was leading in the final polls when in reality Kerry was pulling ahead.

You choose LV polls only. And Kerry was ahead in 2 of 12.
Yet, in virtually ALL RV polls at the end, Kerry was leading.
And you don't show any of them them.

I included a number of LV's in my group of 18.
But you don't include a single ONE in your 12.
Why?

Tell me. Is it just a coincidence?
Why all LV polls, and no RV polls?

Can you answer that?

Printer Friendly | Permalink | | Top

tritsofme

(1000+ posts) Send PM | Profile | Ignore

Tue Jan-25-05 12:51 AM
Response to Reply #61

65. No coincidence at all

I don't see the relevance of including RV polls, and I have always found LV polls to be superior, both before and after the election. I haven't looked very closely at the polls you have chosen, but I didn't choose polls that included Nader either because his support turned out to be negligible.

Most LV models are now more complex that just asking whether or not you had voted in the last election, and many other factors are taken into consideration. Many assumed 60% voter turnout as well.

In order to believe that RV samples were correct you have to buy your hypothesis of massive fraud.

In an LV sample at least we can be almost positive that everyone included actually went to the polls on election day, in an RV poll there is a substantial chunk of registered voters that did not vote on election day, and it has always been conventional wisdom that non-voters and infrequent voters would tend to favor Democrats, so it would not surprise me that RV polls overstate Democratic support.

Printer Friendly | Permalink | | Top

Name removed

(0 posts) Send PM | Profile | Ignore

Tue Jan-25-05 01:18 AM
Response to Reply #65

67. Deleted message

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

tritsofme

(1000+ posts) Send PM | Profile | Ignore

Tue Jan-25-05 01:23 AM
Response to Reply #67

68. Which polling outfit a week before an election

puts out only an RV poll?

One that is not very credible.

I'll thank you not to insult me personally, as I have never done the same to you.

Printer Friendly | Permalink | | Top

anaxarchos

(963 posts) Send PM | Profile | Ignore

Tue Jan-25-05 04:47 PM
Response to Reply #65

75. OK, I'll bite...

WHICH LV model do you believe is the correct one? How did you use that to select the polls you just chose? Why do the polling companies offer RV polls at all? Where is the crossover? What does the general trending indicate to you? What is the RV/LV trend for the various LV models?

Just to be perfectly clear... I don't think you have the slightest idea of what you are talking about.

Make your case...

Printer Friendly | Permalink | | Top

Yupster

(1000+ posts) Send PM | Profile | Ignore

Thu Jan-27-05 01:49 AM
Response to Reply #65

98. You see tritsofme, you must use RV polls

even though all networks switch to likely voter polls because they consider them more accurate as election day nears.

The reason you have to use RV polls is the PEW Poll.

The PEW poll's final pre-election result was Bush 51 - Kerry 48.

They hit it right on the nose.

Well we can't have that, so we use RV where PEW had Kerry up 46-45.

Then the conclusion is obvious. Kerry was ahead.

See. A poll that hit the result dead on can then be used as evidence of how the polls got it all wrong.

Amazing shit.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 11:36 PM
Response to Reply #51

62. Your nose is growing. In the last 18 polls, Kerry led in ELEVEN.

Edited on Mon Jan-24-05 11:41 PM by TruthIsAll

And of these 18 INDIVIDUAL polls, at least SEVEN were LV's.
Count 'em, cherry-picker.
11 of 18.

EIGHTEEN POLLS.
18 INDIVIDUAL POLLS.
NO DUPES.
I DON'T PULL THAT SLEIGHT OF HAND.
ONLY YOU DO THAT.

NOT TWELVE POLLS.
EIGHTEEN POLLS.
THE FINAL 18 POLLS.
AND KERRY LED IN 11.

HOW DOES IT FEEL, TO HAVE TO DEAL WITH THE TRUTH?

Printer Friendly | Permalink | | Top

qwghlmian

(768 posts) Send PM | Profile | Ignore

Mon Jan-24-05 11:39 PM
Response to Reply #62

63. Cherry-picked truth, no less. n/t

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 11:52 PM
Response to Reply #51

64. KERRY LED IN 7 OF THE LAST 11 POLLS, 2 DAYS BEFORE THE ELECTION.

TRUTH.
TRUTH.

GO AHEAD.

HAS THE JURY REACHED A VERDICT?
WE HAVE, YOUR HONOR.

Printer Friendly | Permalink | | Top

righteous1

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 01:49 PM
Response to Reply #5

12. I have noticed for quite a long time that anything that disagrees with

your numbers is "biased"

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 03:08 PM
Response to Reply #12

13. Did you read the post? I'm responding to his claim that I cherry-picked.

Edited on Mon Jan-24-05 03:44 PM by TruthIsAll

Your statement is total BS.
I don't ask anyone to agree with my analysis.

He is the one calling the use of RV polls cherrypicking.
How come the pre-election polls I picked BEFORE THE ELECTION agreed with the exit polls?
There is nothing to disagree with.

I listed 18 final pre-election, RV and LV polls, prior to the election, which were right on the money compared to the exit polls.

And he comes with his list of RV polls ONLY going back to Sept.
So just who is cherry-picking?
And just who is biased?

The polls/dates I show are factual.
Show me ONE poll that is in error.
If you can, I will note it.
But you will not because you cannot.

Why do you guys get so upset by poll numbers, anyway?

Printer Friendly | Permalink | | Top

righteous1

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 04:38 PM
Response to Reply #13

20. I saw that ,and you left out some. I averaged all the polls as well

and came up with a 1.8% Bush margin prior to the election

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 04:50 PM
Response to Reply #20

21. You did, huh? Then explain THIS. Let's see your spreadsheet of the LATEST

Edited on Mon Jan-24-05 05:17 PM by TruthIsAll

polls only. No dupes, please.

Break out the LATEST RV and LV polls prior to the election while you're at it.

Let's see how you compute your average.
You are missing polls, buddy.

Here is your latest 18:
You show just TEN unique polls from Oct. 14:
I show EIGHTEEN polls from Oct. 20
Talk about cherry-picking!
Tell trit to get a new set.

3 GW Battleground
2 Rasmussen
3 CNN USA Today
2 FOX
2 CBS
2 Newsweek
1 ARG
1 TIPP
1 LAT
1 Time

18 Total

1GW/Battleground (1000 LV) 10/31 - 11/1 50% 46% Bush +4
2Rasmussen (3,000 LV) 10/30 - 11/1 50.2% 48.5% Bush +1.7
3TIPP (936 LV) 10/30 - 11/1 46.9% 44.3% Bush +2.6
4FOX News (1200 LV) 10/30 - 10/31 46% 48% Kerry +2
5CNN/USAT/Gallup (1573 LV) 10/29 - 10/31 49% 47% Bush +2
6CBS/NY Times (643 LV) 10/28 - 10/30 50% 47% Bush +3
7ARG (1258 LV) 10/28 - 10/30 48% 49% Kerry +1
8Newsweek (882 LV) 10/27 - 10/29 51% 45% Bush +6
9Battleground (1000 LV) 10/25 - 10/28 51% 46% Bush +5
10CNN/USAT/Gallup (1195 LV) 10/22 - 10/24 52% 46% Bush +6
11Los Angeles Times (881 LV) 10/21 - 10/24 49% 48% Bush +1
12Newsweek (880 LV) 10/21 - 10/22 48% 47% Bush +1
13Time (803 LV) 10/19 - 10/21 52% 47% Bush +5
14GW/Battleground (1000 LV) 10/18 - 10/21 49% 45% Bush +4
15Rasmussen (3,000 LV) 10/17 - 10/19 48% 47% Bush +1
16FOX News (1000 LV) 10/17 - 10/18 48% 43% Bush +5
17CBS News (678 LV) 10/14 - 10/17 47% 46% Bush +1
18CNN/USAT/Gallup (788 LV) 10/14 - 10/16 52% 44% Bush +8

Printer Friendly | Permalink | | Top

righteous1

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 05:11 PM
Response to Reply #21

22. Better than that I will point out a few of your errors.

Edited on Mon Jan-24-05 05:19 PM by righteous1

TIPP +2.1 B Rassmussen (missing) +1.7 B GWB(missing) +4B Newsweek B+6 Pew B+3 CBS B+2 Fox K+2 ICR B+3 AP poll (2weeks old) ARG TIE LA Times TIE etc etc

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 05:42 PM
Response to Reply #22

24. You failed to 1) point out any errors, 2)explain why your 10 polls duped

Edited on Mon Jan-24-05 05:43 PM by TruthIsAll

Who are you fooling?

Its all right there.
You cannot hide.

You show only 10 unique polls, all duplicated.

You don't do your homework at all.
You are dealing with someone who does.
Everyone at DU knows I do my work without any BS.

Is this the best you and trit can come up with?
I just called you out on your "work".
It's a sham and you know it.

I challenge you once again:
I want to see 18 FINAL, UNIQUE polls.
NO DUPES.

I want you to PROVE my numbers are inaccurate.
Cite dates, sources.

So far, you have only proved that your polls are incomplete and cherry picked LV duplicates.

OK, go back to the drawing board.
If you dare.

Printer Friendly | Permalink | | Top

righteous1

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 05:45 PM
Response to Reply #24

25. I am challenging you #1 on your numbers and #2 on your omissions

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 05:53 PM
Response to Reply #25

28. I guess you won't do your homework. As usual, you avoid the issues.

I was very clear that the 18 polls were my choice. I never claimed they were ALL the polls.

WHO THE HELL ARE YOU TO DEMAND THAT I DISPLAY EVERY DAMN POLL OUT THERE?

Now answer the questions I posed in my last post or just admit defeat.

Printer Friendly | Permalink | | Top

righteous1

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 05:56 PM
Response to Reply #28

29. Well JEEEZ, I can make it look like Ralph Nader won if I can

Edited on Mon Jan-24-05 06:53 PM by righteous1

choose my own perameters for doing so. You obviously are not a big believer in the "scientific method"

Printer Friendly | Permalink | | Top

righteous1

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 05:51 PM
Response to Reply #24

26. here's one CBS

Edited on Mon Jan-24-05 06:08 PM by righteous1

http://www.cbsnews.com/stories/2004/11/01/politics/main652662.shtml there are several more. Your "omissions" speak for themselves. 2 major polls GW Battleground and Rassmussen you just conveniently left out.

Printer Friendly | Permalink | | Top

righteous1

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 05:58 PM
Response to Reply #24

30. I have come to the conclusion that contrary to your screename

you are far less interested in the truth and far more interested in legitimizing your preconceived notions

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 08:50 PM
Response to Reply #30

45. As your hero said once : "Who cares what you think"?

I will leave you to your imagination what I and most DUers think about you.

Printer Friendly | Permalink | | Top

righteous1

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 09:18 PM
Response to Reply #45

47. OK by me ....n/t

Printer Friendly | Permalink | | Top

tritsofme

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 06:47 PM
Response to Reply #5

35. Like I've said a million times before

All I did was grab the last 50 polls conducted before the election.

There is no reason to believe that RV polls are more reliable that LV polls in this election because most likely voter models assumed turnout of around 60% and are more complicated than asking whether or not you voted in the last election.

All you ever did was cherry pick the polls to fit your conclusions.

Those ridiculous assertations before the election that the probability of a Kerry win was 99.99% did a lot more harm than good.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Tue Jan-25-05 07:17 PM
Response to Reply #35

80. HERE IS WHAT THE MYSTERY POLLSTER SAID ABOUT THE FINAL 15 POLLS...

Edited on Tue Jan-25-05 07:23 PM by TruthIsAll

Why don't you naysays take it up with him?
Show him your list.
Tell him TIA told you about it.

http://www.mysterypollster.com/main/2004/11/lessons_likely_.html

Blumenthal (10 days after the election) talking about the "election" of Bush versus the pre-election polls:

"However, that margin of error applies to only to one survey at a time. As Prof. X pointed out, the same logic does not apply to an average of 15 or more polls. In fact, most of the surveys done in the final week (11 of 15) had a “bias” toward Kerry – they showed Kerry ahead or had the margin closer than three points. That result cannot be explained by random variation: My application of the binomial distribution puts the probability of that happening by chance alone (assuming that undecideds broke evenly) at roughly 6%."

Printer Friendly | Permalink | | Top

anaxarchos

(963 posts) Send PM | Profile | Ignore

Wed Jan-26-05 12:50 AM
Response to Reply #80

85. Hah, since you are using this, MP also says 10 of the last 18 LV...

...favored Kerry. This is the pertinent quotation in the same reference:

http://www.mysterypollster.com/main/2004/11/lessons_lik...

"Some likely voter models worked better than others - When I presented the details on likely voter models, I noticed that the pollsters that used a variant of the Gallup likely voter model showed Bush doing consistently better than other surveys. That difference now looks prescient. The following table shows the results of those using the Gallup likely voter model either in the final week (Gallup, Pew, Newsweek) or in the final two weeks (adds Time and the LA Times). In both cases, the Gallup-model showed a Bush margin closer to the actual result (3.2%+) than the average of the other surveys (0.9%). There were three surveys in the “other” category that correctly forecast Bush’s final three-point margin (notably, Pew, TIPP, ICR), but the other 10 showed Kerry doing slightly better."

Of course it is upside down (the ones that "did well" matched the vote count) but the point is that 10 out of the last 18 LV polls favored Kerry.

The irony is that the "Gallup likely voter model" was mostly discredited before this election because it uses a "cutoff" for categories like "young first time voters" and thus fails completely in high turnout elections. Blumenthal himself admits this in his pre-election comments.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 01:25 PM
Response to Reply #1

10. Why not go back to August when Kerry was leading?

Edited on Mon Jan-24-05 01:34 PM by TruthIsAll

Do you really believe you are accomplishing anything by posting all that stuff?

I provided FINAL polls, a MIX of LV's and RV's.
And you provided that worn out LV clutter.

You cannot refute a SINGLE poll that I reference here.
They include the NINE major Independent and NINE Corporate pollsters.

You think you can just splash those all those LV polls from Sept. and gain any credibility around here?

Your response is an exercise in futility.

You and trit seem to take this stuff very personal.
What's your problem?

Printer Friendly | Permalink | | Top

anaxarchos

(963 posts) Send PM | Profile | Ignore

Mon Jan-24-05 03:36 PM
Response to Reply #10

14. There was a lot of LV/RV discussion before the election...

...it was pretty much of a consensus that if turnout exceeded 116-118 million, the RV number was more significant.

I'll see if I can dig up a "learned" quote.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 03:48 PM
Response to Reply #14

16. Of course, I read it also. And it makes sense.

anaxarchos, you are quickly learning that facts mean nothing to these naysayers.

They keep coming back with their same old talking points.
They are on a mission to deflect, not effect, the truth.

Printer Friendly | Permalink | | Top

RaulVB

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 03:58 PM
Response to Reply #1

18. So you are unable to get your own "info" now?

Glad to see it. Running out of talking points, Great!

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Mon Jan-24-05 06:28 PM
Response to Original message

33. Hi Truthisall,

I hope you will be as kind at to answer my questions. Even for someone that have read a few of your posts, it is not always obvious what your numbers represent.

I appreciate that you posted this in regular caps.

What is the weighting method for the state polls?

Have you weighted the 18 avg or 9 avg for sample/MOE? It says indep, does this mean that there isn´t any overlap? Are these weighted? What is the accumulated MOE for each poll?

What merit does this approach to predicting the election
result have?

Have you compared the state polls you used against the state exit polls? How was the match?

What dataset of the Exit polls are you using? Why? I take it you know about the four different sets as reported in the reports from the caltech/MIT voting project? What are the problems with the others?

I think I should state for clarity that I am personally sure there was large scale electronic fraud, although it seems to me it was not coordinated or "singular". I witnessed an election being stolen in Albania a while back where the winner actually didn´t need the fraud he had set up to win. He still cheated, this election was later accepted by the OSSE as "legit with problems". Blanket accepted by the press, this election was good news even if the opposition were silenced and the first demonstration in the recent history of the country was brutally beaten down just before the election. It put what was the Right hand of the communist dictator Enver Hoxha in office, he was chair of what was actually called the "Democratic party". Not such a long time after that he stole about all the savings of his own people in a pyramid game pensions sham, sound familiar?

The actual vote from New Mexico broken down with voting equipment, as well as that report from the Washington race I consider "hard proof", I actually considered it proved the moment I learnt that the code on the machines are not inspected, I am actually surprised there is so much evidence as there is.

Printer Friendly | Permalink | | Top

torque

(167 posts) Send PM | Profile | Ignore

Mon Jan-24-05 06:42 PM
Response to Original message

34. Same MO - Avoid the truth at all costs

Same crap, different day. Their playbook is pretty shallow BECAUSE it's the truth they fear. Grow up repigs.

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Mon Jan-24-05 08:01 PM
Response to Reply #34

39. Is this a response to my post?

Edited on Mon Jan-24-05 08:09 PM by Bouvet_Island

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 08:45 PM
Response to Reply #39

44. No. This is. Read my posts, draw your own conclusions.

Edited on Mon Jan-24-05 08:45 PM by TruthIsAll

Your a smart guy.

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Tue Jan-25-05 09:00 AM
Response to Reply #44

69. I think my questions are relevant.

Edited on Tue Jan-25-05 09:29 AM by Bouvet_Island

If you are not willing to disclose your methods, I take it you are simply an educated guesser, nothing wrong with that. Maybe you should represent the proportion and quality of doubt better in your analysis, from reading your posts I get the impression you are not only sure there was fraud but that you are able to measure it correctly.

This is a mystery to me, I tend to believe there are a particular problem with your approach and that is that the fraud itself is an error source in the exit poll weigthing. We know of equipment that *lost* votes, and I believe turnout are a factor in the weighting. I also think the amount of suppression is a problem, as I got the impression that previous turnout are a factor as well. It is a too complicated debate to get finished before we know more about the methods and sample, but it just seems to me that the conlusion we both believe are correct means that the current exit poll have large error sources that are extremely hard to eliminate.

I believe the poll could be reassembled to show something useful, but I believe that would need people a lot smarter than both me and you.

Your calculations over are incorrect if you didn´t weigh for sample size and MOE, the idea is that we listen more to the guy that won the most earlier when we place our bets on the horses. We don´t have a democratic vote with 12 freshman idiots and 6 clever more experienced guys.

The way uncertainty is usually represented in science is in using ranges instead of exact numbers. If you know the numbers are inaccurate you represent that, you don´t post the extrapolations as if they were actual observation. The press don´t usually feel like doing this, but then they don´t need to be either accountable or responsible.

The European exit polls have perfect previous both exit poll and vote data, and they are working with a uniform standarized election system, not 13000 independent elections with 3 types of ballots measured on 10 different types of equipment. Someone telling me US exit polls would be able to achieve *the same* accuracy in an election with large scale fraud, makes me sceptical.

I believe you could be entirely wrong, in both directions, maybe even at the same time on the state level. You could also be "correct", but I would attribute that to the kind of luck one have in the lottery.

It also makes no sense to me that you want polls from long before the election to be a factor, this election are a rulebreaker of different obvious reasons. The Bush administration have been pushing any and all buttons suited to make people act unpredictably when filling out the ballot, in both directions. I believe it is the same factor as in the Israeli elections that got sharon in Office. It faulted the polls as people didn´t display rational "continuos" behaviour, they became somehow split and unpredictable. I think many of the polls systematically underestimated the Kerry vote ahead of the elections on purpose, many of them didn´t call cell phones at all. I believe they also undersample poor people, and that there was a larger poor vote than usual. The point is, if you can extrapolate from previous elections in general, you can´t do it with this one and it is so obvious that I believe even common people are able to "sense" it. You get that hunch not to trust the polls.

If you represented uncertainties and possible error in your analysis, they would act less as flamebait, and they would also be more efficient in convincing people that are not certain of the conclusion in advance.

* edited for clarity.

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Tue Jan-25-05 09:24 AM
Response to Reply #69

70. Some background.

Edited on Tue Jan-25-05 09:41 AM by Bouvet_Island

Mitofskys faq

How are projections made?
Projections are based on models that use votes from three (3) different sources -- exit poll interviews with voters, vote returns as reported by election officials from the sample precincts, and tabulations of votes by county. The models make estimates from all these vote reports. The models also indicate the likely error in the estimates. The best model estimate may be used to make a projection if it passes a series of tests.
___

(This means that fraud would be an error source in the weighting of the results).
___

How do you select sample precincts?
The polling places were selected as a stratified probability sample of each state. The purpose of stratification is to group together precincts with similar vote characteristics. A recent past election was used to identify all the precincts as they existed for that election. The total vote in each precinct and the partisan division of the vote from this past race are used for the stratification. In addition, counties are used for stratifying the precincts. The total vote also is used to determine the probability of selection. Each voter in a state has approximately the same chance of being selected in the sample.
___

This raises questions about the randomness of the precinct selection, if also previous elections were fraudulent. It also could mean that the precinct sample are not random at in respect to vote equipment type.
___

*Also edited for clarity
*And the furcking link...

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Tue Jan-25-05 10:21 AM
Response to Reply #69

71. I have done a math analysis of deviations, as have Freeman, Baiman et al

YOU are misrepresenting.
In a very clever way.
That is clear.

I calculated the numbers.
They speak for themselves.
Whether or not fraud was involved, the results are beyond random chance.
There is a conclusive pattern here.

Take your pick.
Fraud or polling error (in one direction).
YOU can draw whatever conclusions you wish.

I did the numerical analysis.
Did you ever do one?
If not, why not?

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Tue Jan-25-05 02:48 PM
Response to Reply #71

74. Heh. I´ll have to reply on verse then.

I´ll hope you´ll allow me the pleasure.

YOU are unaccountable
piling insurmountables
Starting counter fog to extinguish the night.

THERE is the accusation
NOONE can penetrate the Night of Mitofsky
The binary night of the he blind ballots
Are the TRUTH still afloat?
The sun will burn its wax wings off.

Suspicions fly with the ghosts
the single candle burns and burns
but is it making the night lighter
or the eyes blind to those dark shades of grey?

Let´s pray the Faith will burn
And a new leaf will grow
on that burned old tree
the sinew still in place
inside.

____

I hope this answers any legitimate questions anyone could have, about anything and that it didn´t hurt anyones eyes or ears.

Printer Friendly | Permalink | | Top

anaxarchos

(963 posts) Send PM | Profile | Ignore

Tue Jan-25-05 06:00 PM
Response to Reply #69

78. Cute logic is not apriori the same as science....

... it is just as easily sophistry. I read a lot of words about exit polls but the point of the thread was pre-election polling. The scope of your provocations is so wide as to be nearly unaddressable and not one in ten of those is directed at a single thing that was mentioned in this thread. The idea that previous election fraud itself induces error is very cute but it is also silly unless you say how, to what extent, etc.

Cut the cute, where's the beef? What's your case?

Printer Friendly | Permalink | | Top

Name removed

(0 posts) Send PM | Profile | Ignore

Tue Jan-25-05 10:01 PM
Response to Reply #78

81. Deleted message

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

anaxarchos

(963 posts) Send PM | Profile | Ignore

Tue Jan-25-05 11:10 PM
Response to Reply #81

83. Uh Huh.... A lot more words...

1) If you know about the archives then you know that the debate about exit polls has raged long and hard. This thread is about the similarity of results between the last set of polls and the exit polls. You are off topic from the get go. If you have a theory about this, state it.

2) You manage to talk about everything but what I asked you. If you have A.D.D., I apologize. Otherwise, how exactly does previous election fraud swing the exit polls to Kerry? Explain it slowly with lots of math if you don't mind. I'll try to keep up.

3) The idea that any error, no matter how contrived, makes the results of any analysis suspect is sophistry... one of the basic tricks of sophistry at that. "Cute" is a nice word used in place of "cheap" or "crude".

4) You are not yet another student of Diogenes are you... looking for a truly honest man? If your life's mission is to make TIA less sure of himself because it could be "misleading", that's a pretty crowded field here (all full up). I would have thought Rice or Rumsfeld or Rove would be a better target but hey,... the three 'R's are overrated (way overrated).

5) If you want to play, I'm your man. But, let's play by my rules. You say what you think (state your case) in a scientifically coherent way and then we'll find out what I know about scientific method or polling or any other relevant issue. Let's hear YOUR theory hotshot.

6) I have never been gay but after that sleazy gay-bashing/gay-baiting "joke" of yours, I'm gay now. We can discuss this as well. Create a thread.

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Wed Jan-26-05 06:53 PM
Response to Reply #83

93. Heh,

I had a hunch that post would possibly get deleted. It is good the system works!

I disagree I am off topic, TIA gave a nice answer to my post at the bottom, and what rule is it anyway that says you have to keep strictly on topic? What is your point?

I actually have a mild adult ADHD, that is not exactly the same but I accept the apology. My mother works with kids with heavy ADD and ADHD, and she is the one I got it from. She get´s paid trips to the US from Norway from the makers of Ritalin.

I like cheap or crude, you can keep cute. If smoking good cigars and eating oysters are sophistry too, I am guilty as hell. I even eat goose liver, which I believe Bush just lifted the import boycott of in the US. Good for you!

I don´t think either you or TIA would be a life´s mission for me. And I hardly doubt I´d be able to stop TIA if he sets his mind on something, maybe push his general direction an inch or two but I don´t hold much illusion there, he has some character.

I don´t want to "play", I was interested in a discussion of certain facts and theories. There isn´t much relevant to that in both your posts. I have made my opinion and it´s reason, I believe clear enough. I am not able to see any incoherence in my assesment. My english is limited so I might not have been able to communicate it perfectly, maybe it is clearer to you in my post at the bottom of this thread. Hot shots was a cool movie at 9.

That wasn´t really a joke or funny, I agree, sleazy that is a compliment. The poem I admit was a joke though, I pray to God every night noone takes that seriously. I hope you don´t take this too personal, in your previous post you somehow didn´t instruct me to treat you very humbly, if you had attacked my points I maybe would have kept more strictly on topic.

I didn´t understand that part about the truly honest man, but I hear the weather is great in ancient Greece so maybe I´ll meet you there some day.

Printer Friendly | Permalink | | Top

anaxarchos

(963 posts) Send PM | Profile | Ignore

Wed Jan-26-05 09:48 PM
Response to Reply #93

95. Fremskrittspartiet? ....................eom.

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Wed Jan-26-05 10:40 PM
Response to Reply #95

96. Well,

"I believe" in the greenhouse effect, I once actually made Carl I. Hagen admit a full defeat which I haven´t seen before or after, I asked him if he had ever thought about the fact that manufacturing a car larger emissions than driving it, and it like hit him and he just said "Eh.. NO.", but then a second and a half and he is back again and he´s like "but if you are thinking about the 1% of carbon dioxide that is emitted by man" and the whole audience where like "Oh crap". That was cool though.

Let´s say I am not a fan of the current government. The PM is a priest, it was revealed that he have sworn an allegiance oath at a neocon christian brotherhood institution, and that he had arranged an unofficial official meeting with Ashcroft there, outside of the Official Program. In like the largest paper in Norway, that was.

We are getting a semblance of partisan politics here in the next elections, that is a fresh breath after a rather confused situation lately. I mean we get Red-green-light red (US fairly progressive dem) coalition against Light blue-dark blue-(brownish blue: Fremskrittpartiet) (republicans super light). I´m about in the center of the red camp, maybe with a dose of globalism or how to put it. I´m not enthustiastic about Norwegian politics because none of the parties keep themselves with any real Vision, thus people are left to vote on their pockets or "values".

I guess you are a Democrat?

Printer Friendly | Permalink | | Top

anaxarchos

(963 posts) Send PM | Profile | Ignore

Wed Jan-26-05 11:02 PM
Response to Reply #96

97. You never know.... I could be Mot Dag........eom.

Printer Friendly | Permalink | | Top

anaxarchos

(963 posts) Send PM | Profile | Ignore

Tue Jan-25-05 11:15 PM
Response to Reply #81

84. self deleted n/t

Edited on Tue Jan-25-05 11:16 PM by anaxarchos

Printer Friendly | Permalink | | Top

rdmccur

(622 posts) Send PM | Profile | Ignore

Mon Jan-24-05 10:03 PM
Response to Original message

56. TIA

Why not send all of you analyses to uscountvotes.org?

Printer Friendly | Permalink | | Top

Carolab

(1000+ posts) Send PM | Profile | Ignore

Mon Jan-24-05 10:11 PM
Response to Reply #56

58. I was just about to say, let' s see what uscountvotes.org has to say. n/t

Printer Friendly | Permalink | | Top

rdmccur

(622 posts) Send PM | Profile | Ignore

Tue Jan-25-05 05:10 PM
Response to Reply #58

76. I'm not looking

for feedback from them. It's just that they may find some of these analyses helpful.

Printer Friendly | Permalink | | Top

jkd

(151 posts) Send PM | Profile | Ignore

Mon Jan-24-05 10:57 PM
Response to Original message

59. Sate exit polls provided by NEP

The National Election Pool(NEP) provided post-election exit poll results for each state. I don't know if they are averaged anywhere, but you could easily do that. http://www.exit-poll.net/

Printer Friendly | Permalink | | Top

NationalEnquirer

(571 posts) Send PM | Profile | Ignore

Tue Jan-25-05 10:29 AM
Response to Original message

72. Wow, great post.

I need info like this.

Printer Friendly | Permalink | | Top

anaxarchos

(963 posts) Send PM | Profile | Ignore

Tue Jan-25-05 05:30 PM
Response to Original message

77. Support for TIA from a most unlikely source....

...Mark Blumenthal (aka the Mystery Pollster) who has become something of a darling for "the exit polls were wrong" right-wing apologist set (even though he is a democrat).

http://www.mysterypollster.com/main/2004/11/lessons_likely_.html

Blumenthal (10 days after the election) talking about the "election" of Bush versus the pre-election polls:

"However, that margin of error applies to only to one survey at a time. As Prof. X pointed out, the same logic does not apply to an average of 15 or more polls. In fact, most of the surveys done in the final week (11 of 15) had a “bias” toward Kerry – they showed Kerry ahead or had the margin closer than three points. That result cannot be explained by random variation: My application of the binomial distribution puts the probability of that happening by chance alone (assuming that undecideds broke evenly) at roughly 6%."

Printer Friendly | Permalink | | Top

dzika

(1000+ posts) Send PM | Profile | Ignore

Wed Jan-26-05 02:58 AM
Response to Reply #77

87. I don't trust Blumenthal's motives

He's slowly blogging himself into a corner. Protecting his "TV Analyst" status is more important than simple truth.

Math comes down to simple truth. That's why TIA runs circles around these other clowns.

Printer Friendly | Permalink | | Top

anaxarchos

(963 posts) Send PM | Profile | Ignore

Wed Jan-26-05 01:39 PM
Response to Reply #87

88. He said a lot of stuff before and immediately after...

... the election that he doesn't want to remember now. You are right about not trusting him. He is the most common right-wing citation for why the exit-polls don't support election fraud. And since Blumenthal has taken every position imaginable, it is easy to do.

Printer Friendly | Permalink | | Top

loudsue

(1000+ posts) Send PM | Profile | Ignore

Tue Jan-25-05 06:15 PM
Response to Original message

79. You're doing great, TIA! The TIA-bashers are out in force...yet again!

If that doesn't tell you something, nothing does! They can't handle the truth!! They just seem to want it to be true that bush won this election fair and square so badly, that they have to come around ganging up on you. LOL!!

Funny that your conclusions are also the ones that the professional academic statisticians come to. There's NO FRIKKIN' WAY bush won this election. Thousands upon thousands of people who have always voted republican, voted AGAINST bush in this past election. And I haven't heard of ANYBODY that voted for Gore or Nader before, voted for bush this time.

Anybody that thinks bush won this fair and square isn't paying attention, NO MATTER WHAT THE POLLS SAID!!

It seems that, prior to the election, we were finding all these polls that were polling something like 15% more republicans, anyway. One poll was polling 23% more republicans, if I remember correctly, just so that they could show he was neck-n-neck with Kerry.... which, of course, he wasn't! Kerry won by a landslide.

The election was STOLEN...yet again.
:kick::kick::kick:

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Tue Jan-25-05 10:09 PM
Response to Reply #79

82. Is it possible to believe the election was stolen,

and that the exit polls are impresise? That it is possible they actually showed less fraud than was present? And that we can´t be that certain about their accuracy, or that they were carried through in the amount they were said to be?

There are an academic debate, there are good scientists that are anti Bush that believe Freeman et al are wrong.

As well as bad scientists, like the MIT/caltech electronic voting people that have produced some of the lousiest and most untimely science I have seen my entire life.

Printer Friendly | Permalink | | Top

loudsue

(1000+ posts) Send PM | Profile | Ignore

Wed Jan-26-05 02:11 AM
Response to Reply #82

86. When it comes right down to it, Bouvet, you just have to look at the big

picture.

The big picture here is that it's OBVIOUS the election was rigged and gerrymandered. Too many pieces of the puzzle are already known.

Is it possible to believe the election was stolen AND the polls were wrong? Sure. When it comes to this election, it's possible to believe most anything....EXCEPT that bush, and a whole slew of lousy, mentally ill, screw-the-people, fanatically right wing Kristian republicans won. THAT part is NOT possible to believe.

:kick::kick::kick:

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Wed Jan-26-05 04:11 PM
Response to Reply #86

90. Good.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Wed Jan-26-05 03:41 PM
Response to Reply #82

89. Of course, the exit polls were wrong...

Edited on Wed Jan-26-05 04:37 PM by TruthIsAll

You ask:Could it be the exit polls were wrong?
Of course they were.
But there is this little thing called margin of error.
And there is this little thing called consistent bias.

If the question is:
Were the polls precisely correct, and did the deviations exactly matched the true votes? Of course not.

But that is NOT the question that needs to be asked.
That question is:
What are the probabilities, based on sound statistical analysis of the best available prior and post-election polling data, that Bush could have achieved the results he reportedly did?

Fact:
Kerry lead in the majority of pre-election polls and was poised to win at least 2/3 of the undecided vote, as ALL challengers have done historically.

Fact:
Bush had to win 80% of the undecided vote to end up with the numbers he did. That was a sheer impossibility. Every reputable pollster, including MP, knows it never happened.

Fact:
43 out of 51 states deviated to Bush from the exit polls.
The odds of this occurrence: 1 in 2 MILLION.

Fact:
15 states deviated BEYOND the MOE to Bush.
The odds:
1 in a TRILLION.

Fact:
The National Exit Poll of 13,047 before becoming contaminated with bogus Bush votes to a "final" 13660, had an MOE of 1%.
The naysayers claimed otherwise.
But Mitofsky claimed it was 1.0% right at the bottom of the Washington Post link.

Fact:
The NEP was a randomly-selected sample.
The naysayers said otherwise.
But Mitofsky claimed it was so at the bottom of the Washington Post link.

Fact:
Kerry led the NEP in ALL the weighted demographic categories.
Depending on the category, the probability of Bush getting his final vote percentage varies from 1 in a million to 1 in 300 million.

Fact:
All machine and voter registration glitches have been in one direction only. To Bush.

Fact:
The lines at polling places in heavily OH democratic districts stretched around the block.

Fact:
There are many more facts of mistabulation. Virtually 99.9% favorable to Bush. That is beyond coincidence.
If it's not a coincidence, then it's massive F-R-A-U-D.

So, to answer your question: could the exit polls have been wrong?
Of course, wherever the numbers come from, you can bet that Kerry did MUCH better - even better than the National AND state exit polls which indicated that it would take divine intervention for Bush to win.

Gather the pure, uncontaminated pre-election and exit polling data.
Do the math. Calculate the probabilities
Study the history (2000/2002).

And you come to just one conclusion.

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Wed Jan-26-05 06:00 PM
Response to Reply #89

92. It is nice we can have civilized discussion.

I will cite Mitofsky againg at http://www.exit-poll.net/faq.html#a15">exit-poll.net

What is the Margin of Error for an exit poll?
Every number estimated from a sample may depart from the official vote count. The difference between a sample result and the number one would get if everyone who cast a vote was interviewed in exactly the same way is called the sampling error. That does not mean the sample result is wrong. Instead, it refers to the potential error due to sampling. The margin of error for a 95% confidence interval is about +/- 3% for a typical characteristic from the national exit poll and +/-4% for a typical state exit poll. Characteristics that are more concentrated in a few polling places, such as race, have larger sampling errors. Other nonsampling factors may increase the total error.

_____

I am sceptical to the MOE here, I believe it would be less and that he is doing this instead of buying some insurance. I think as low as 1% would be a stretch in this particular election on account of unexpected turnout, I am talking about the "pure" sampling error before weighting.

The point of the quote is at the bottom, the point is that you only account for standard sampling error, and you make a few assumptions that are not realistic:

1. There wasn´t a higher non-response from Bush voters.
2. There wasn´t problems with the representation of the different equipment with the precinct sample.
3. There wasn´t problems with the randomness of the precinct sample
4. There wasn´t problems with the precinct samples not adequatly representing voter suppression.
5. The precinct sample as well as the sample at each precinct were *perfectly random*. See under.

6. Fraud this year didn´t affect the weighting.

This is an incredible position, if you hold it. Your conclusion means this, I can´t see how you could escape this.

7. Previous fraud didn´t affect the weigthing.

8. Mitofsky or other people in his organisation didn´t cheat, or manipulate numbers that for example would be so wild as in effect to prove fraud.

You calculate the probability of any of these factors to have any influence to be: 0. not 1: 100000000000000000000000, no, they are 0.

You might see this as nitpicking, the Bias still is suspicious to me and all normal people, but there is not a 0 percent chance that it as an example could have been introduced by fraud. Even if the possibility of error producing bias in the poll is as low as 1%, it means that the possibility is not 1:trillion it is then about 1:99 that the result could have been produced and the poll as well as the election result still would be valid. What you are saying are to a degree correct, but what you "telling people" are not.

If there was fraud. Which I want to stress that I am sure there was, although I am not sure of the proportion. I know of several states with wrong results. The entire popular vote have not been accounted for, but I don´t believe in it. I say if there was fraud, because that is how I would write it in a formula. And I don´t operate with a 0 possibility that the machines were actively fraudulent, although it is under 1%.

If there was fraud, I think we have to accept a high chance for this to have affected the exit polls, and also we have to accept that we, and probably noone, will be able to judge how if we don´t get specific information on the method and quantity of say Diebold fraud, as well as full disclosure from Mitofsky.

It is possible there could be something hard inside to be found by someone picking it apart and putting it back together again, but I really wouldn´t expect anything but floaty "suspicious" as opposed to "impossible result".

The math is about the truth, and your math are saying "impossible". That is not true, it is not impossible.

A really good poll, it could do that but as I think I maybe mentioned we are talking about making like the biggest poll in history, obviously against the will of very powerful people. As I also think I maybe mentioned this is important, and we should not convince the good people here that the Exit polls we currently have is "extremely accurate", the idea that the current exit polls are "proof" is simply working against our objectives.

The idea that the results are "funny", on the other hand, is something we easily should tell people. But we have to *convince* people with the other facts, not only because of the problems with the accuracy but because it is a more effective route to do so. Will Pitts post about his paperclip, talking to people on the street, in a respectful manner. For the scientists, we have excellent, irrefutable numbers from New mexico, complete with coverup, suspect judges, and even a democratic SOS, interestingly. The Republicans will still not want to open that jar, I believe it could be a good show.

I think I have been able to make my point now, If there are concrete questions on facts or clarity I will answer, if else I will leave it at this. I appreciated that answer TIA, I believe you are showing nice, insistent spirit although we disagree some on the details. Have a nice one!

Printer Friendly | Permalink | | Top

CFO6

(22 posts) Send PM | Profile | Ignore

Wed Jan-26-05 05:34 PM
Response to Reply #82

91. I take it you believe the election fraud premise is hooooie

Printer Friendly | Permalink | | Top

Bouvet_Island

(227 posts) Send PM | Profile | Ignore

Wed Jan-26-05 06:55 PM
Response to Reply #91

94. Please elaborate,

it is not clear to me what you are saying.

Printer Friendly | Permalink | | Top

jkd

(151 posts) Send PM | Profile | Ignore

Thu Jan-27-05 02:16 AM
Response to Original message

99. MOE's not a simple function of sample size.

TIA makes the assumption that exit polls are a simple function of sample size. He has computed the MOE for all state NEP polls in spite of the fact that Mitofsky published their margin of error. Why would his MOE be more accurate than the poll’s designer. Professor Howard Christensen, dean of the BYU Statistics Department and designer of the Utah College Exit Polls, states: “In a complex sampling situation such as the exit poll, the stages of sampling--the number of selected counties out of a congressional district, and the number of polling places selected--influence the margin of error along with the number of voters selected at each polling place.”

Mr. Christensen continues: “A simple random sample is one selected such that every possible sample of the same size has the same probability. To implement it requires a "list" of the elements in the population; in the case of an exit poll, it would require a list of all of the voters who turnout. Since this list doesn't exist, a simple random sample cannot be conducted in an exit poll setting. We implement a stratified, two or three stage sample (the number of stages depends on which counties are determined to be a part of the sample of counties with certainty) with a systematic random sample of voters at the polling place.”

The BYU poll interviewed more than 9,000 voters. TIA would assign a MOE of near 1.05% for such a sample based solely on size. Here is what Christensen says about the Utah poll: “Our margin of error with a 99% confidence level, was 3.51%. In typical reporting for ordinary opinion polling issues, attitudes etc., the margins of error reported in the media have a confidence level of 95%. To put ours on that comparative basis gives a MOE of 2.67%. The margin of error for other races was larger or smaller depending on the affects of the design and so forth” Christensen says that the “affects of the design” in exit polls determine MOE. Where does that leave TIA’s computations?

The WPE (within precinct error) is more troubling. The individual precinct polling should match the results with a considerably smaller error. Professor Christensen says this about the NEP polls: “Mitofsky and his colleagues have prepared a report and acknowledge the "within precinct" error was larger than it should have been statistically, indicating there were possible selection biases. You can access his report at www.exit-poll.net.
The problems have to be with the polling process, not the election itself because the polling process is supposed to pick up the features of the election itself.”

I didn’t ask Mr. Christensen about possible fraud, but it seems to me that WPE’s are more suspect than the MOE’s. I believe the good professor’s information explains why raw data must be run through the poll’s model before there is any validity to an exit poll. Someone with more time than myself might compare the final poll results with the actual election numbers and then match it to Mifofsky’s MOE. How many state polls will now exceed the MOE?

Printer Friendly | Permalink | | Top

Carolab

(1000+ posts) Send PM | Profile | Ignore

Thu Jan-27-05 05:32 AM
Response to Reply #99

100. Mitofsky's a friggin LIAR. He lied in 2000 and he's lying now. n/t

Printer Friendly | Permalink | | Top

Nederland

(1000+ posts) Send PM | Profile | Ignore

Thu Jan-27-05 10:41 AM
Response to Reply #100

103. If Mitofsky's a friggin LIAR...

...why do you trust his exit poll numbers?

Printer Friendly | Permalink | | Top

Carolab

(1000+ posts) Send PM | Profile | Ignore

Thu Jan-27-05 08:22 PM
Response to Reply #103

104. He's lying about them being wrong. n/t

Printer Friendly | Permalink | | Top

Nederland

(1000+ posts) Send PM | Profile | Ignore

Fri Jan-28-05 12:38 AM
Response to Reply #104

108. That's my point

You seem to be awfully selective about when you trust him.

Printer Friendly | Permalink | | Top

Carolab

(1000+ posts) Send PM | Profile | Ignore

Fri Jan-28-05 01:28 AM
Response to Reply #108

111. I trust the figures that were coming in from the exit poll interviewers.

Edited on Fri Jan-28-05 01:28 AM by Carolab

I don't trust his "computer crash" and his bogus explanations for why the numbers shifted to Bush's favor in the early morning of 11/3/04.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Thu Jan-27-05 08:16 AM
Response to Reply #99

101. I suggest you check the MOE's of published polls and their sample-sizes

Edited on Thu Jan-27-05 09:15 AM by TruthIsAll

You will find that in each case, the MOE agrees with my calculation.
Here is the MOE for some typical sample-sizes, using the simple formula:

MOE = 1/sqrt(N)

N MOE
600 4.08%
900 3.33%
1000 3.16%
1250 2.83%
1500 2.58%
2000 2.24%
2500 2.00%

Do they look familiar?

I point you to Edison/Mitofsky's own notes at the bottom of the Washington Post/NEP web site of the original 13,047 RANDOMLY-SELECTED sample-size poll (before contamination).

You will see that they themselves claim the MOE is 1.0%.
The calculation gives: MOE = 1/sqrt(13047) = 0.875%
So they probably rounded it UP to an even 1.0%.

Try as you might, you cannot spin the TRUTH.

The historical fact is that Exit Polls are more accurate than standard pre-election polls. This is primarily due to two basic, simple, obvious factors:

1) They sample MORE people (most important)
2) Those sampled are not undecided.

Try as you might, you cannot spin the TRUTH.

Printer Friendly | Permalink | | Top

jkd

(151 posts) Send PM | Profile | Ignore

Thu Jan-27-05 10:24 AM
Response to Reply #101

102. You didn't read the post TIA

NEP is National Election Pool. They have published the MOE for all of the states. I told you that twice already,but you just ignore it, like you do everything else.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Thu Jan-27-05 08:27 PM
Response to Reply #102

105. OK, now it's your turn. What is the NEP National Exit poll MOE?

Edited on Thu Jan-27-05 09:14 PM by TruthIsAll

I will ignore you after this one is resolved, I assure you.
In fact, you will probably ignore me.

Because you have stepped right into it.

I am referring to the national exit poll of 13,047 respondents.
Are you even aware that there was one?

Assuming that you are aware of the NATIONAL EXIT POLL, are you also aware of the MOE for this poll, as provided by Edison/Mitofksky?

Just answer that for now.

Hint: The MOE is mentioned in the notes below the poll's statistical summary on the Washington Post/NEP site.

Whatever your response, let us proceed from there.
I do not want to drag this one out.

I will proceed to state my case.
You have already stated yours.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Thu Jan-27-05 10:42 PM
Response to Reply #101

106. Still waiting. n/t

Printer Friendly | Permalink | | Top

jkd

(151 posts) Send PM | Profile | Ignore

Thu Jan-27-05 11:25 PM
Response to Reply #106

107. I'm referring to state exit polls

Your original post that began this thread has NA for state exit polls. My post # 99 was to make you aware of the National Election Pool’s state exit poll results. http://www.washingtonpost.com/wp-dyn/articles/A3735-2004Oct27.html You might want to replace the NA with them.

I noticed that the MOE’s were higher than those you have provided. A rereading of post#99 may suggest why.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Fri Jan-28-05 12:59 AM
Response to Reply #107

109. I'm glad you replied. But you did not answer the question.

Edited on Fri Jan-28-05 01:08 AM by TruthIsAll

What was the MOE for the national exit poll?

You won't answer, so I will.
Mitofsky said it was 1.0%.
That's for 13,047 respondents.

If the MOE = 1 /sqrt(N), where N= 13,047, what is the MOE?
Well, its 0.88%. Pretty close to 1.0%, I'd say. In fact, Mitofsky probably decided to round up to 1.0%.

So if Mitofsky's MOE agrees with the calculated MOE for the National Poll, then he is implicitly using the same formula, no?

And if the formula is correct for 13,047 randomly-selected sample, then it should also be correct for the 2846 sampled by Mitofsky in Florida, right?

Well, for Florida, the MOE = 1/sqrt(2846) = 1.87%
And that is why I calculate MOE's for each state.

They are all exit polls.
They are randomly-selected samples.
The same formula applies to National and State samples.
Except for ONE caveat.

These are EXIT POLLS, not standard, pre-election polls.
We know that exit polls are even more accurate than standard polls. But the formula reflects a standard poll MOE.

Typically, national pollsters sample 1000. That gives a 3.10% MOE. Yet The Economist (YouGov) sampled 2500, for an MOE of 2.0%.

WE CAN ASSUME THAT EXIT POLL MOE'S ARE EVEN LOWER THAN STANDARD POLLS.

MAYBE, FOR THE NATIONAL EXIT POLL OF 13,047, ITS REALLY 0.75%, NOT 0.87%. AND MAYBE MITOFSKY ADDED A 33% CLUSTER EFFECT TO BRING THE 0.75% MOE UP TO 1.0%.

SO WHY NOT MAKE THE SAME ADJUSTMENT FOR THE STATES?
THE 1.87% FLORIDA MOE MAY REALLY BE EQUIVALENT TO 1.35% FOR AN EXIT POLL. IF WE ADD THE 33% CLUSTER EFFECT, WE ARE BACK WHERE WE STARTED - AT 1.87%.

IT ALL SEEMS VERY LOGICAL TO ME.

TO RAISE THE MARGIN OF ERROR FROM HISTORICAL PERFORMANCE IS REALLY COUNTERPRODUCTIVE. I'M SURE BYU'S POLLS, ALONG WITH MITOFSKY'S, ARE MUCH MORE ACCURATE THAN YOU GIVE THEM CREDIT FOR.

TO BELIEVE YOU AND OTHER NAYSAYERS, THE ART AND SCIENCE OF POLLING IS MOVING BACKWARD, NOT FORWARD. I REFUSE TO BELIEVE THAT. I THINK POLLSTERS GET BETTER, NOT WORSE, AND CONTINUOUSLY PERFECT THEIR CRAFT.

OR DOES THE OLD SAYING "PRACTICE MAKES PERFECT" APPLY TO EVERYONE OTHER THAN POLLSTERS?

Printer Friendly | Permalink | | Top

jkd

(151 posts) Send PM | Profile | Ignore

Fri Jan-28-05 02:28 AM
Response to Reply #109

112. Are exit polls completely random? That is the question.

Edited on Fri Jan-28-05 02:38 AM by jkd

I have never mentioned the national poll in any post, but if you keep insisting that I respond, I will. The +/-1% was a pre-determined MOE based on the poll design. The figure remained constant as the data arrived during the day. It was intended to be the final margin of error after data had been weighted, compared to key precincts, and adjusted to the actual vote. They were not based on 1/sqrt (n) nor were any of Motofsky’s state polls.

The design of exit polls is not completely random, but only random in stages. Professor Christensen explains that well in post # 99. Mitofsky gives 3.0% MOE for Florida. You have 1.87%. How can you be using the same formula? Exit poll results are based on poll design.

Professor Christensen was not understating when he gave me the MOE for his poll. He’s very proud of the poll’s tradition.

Are exit polls completely random?………………That is the question.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Fri Jan-28-05 08:12 AM
Response to Reply #112

115. Nice attempt to spin, but the MOE was stated as 1.0%. Period.

Edited on Fri Jan-28-05 08:12 AM by TruthIsAll

Mitofsky can't have it both ways.

If the MOE was 1.0% for the National, then it was also under 2.0% (1.87%) for Florida, using the same simple formula based on sample size - and which applies to every poll I have ever seen. Pick up any text on statistical sampling theory. Furthermore, this was an EXIT poll. And exit polls are more accurate than pre-election polls, although the naysays claim otherwise.

And Mitofsky is being disingenuous as he tries to back away from his own numbers and camouflage the true results when he says that the deviations were due to "reluctant Republican responders" at the exit polls. That is pure BS which has been thoroughly debunked mathematically. Go to the Mystery Pollster site if you want the details.

Printer Friendly | Permalink | | Top

jkd

(151 posts) Send PM | Profile | Ignore

Fri Jan-28-05 10:15 AM
Response to Reply #115

117. That formula is not appropriate

Exit polls do not relate to that simple formula because they are not purely random samples. Mitofsky certainly is not free from reticule because the poll probably has some bias that he is reluctant to admit.Do you really believe that the polls were accurate and that he is taking all this heat to cover-up fraud?

Printer Friendly | Permalink | | Top

skids

(1000+ posts) Send PM | Profile | Ignore

Sat Jan-29-05 12:59 AM
Response to Reply #117

120. Wait, so your saying Mitofski uses a woman's purse? n/t

Printer Friendly | Permalink | | Top

jkd

(151 posts) Send PM | Profile | Ignore

Sat Jan-29-05 01:22 AM
Response to Reply #120

122. No

I don’t want to start any rumors. I always just hate it when my wife wants me to hold her purse while she’s doing some activity. “Can’t I just sit it on the ground nearby. I promise I’ll keep an eye on it.”

I just meant that Mitofsky is under scrutiny (in the cross hairs). That might be an improper use of the word. If it is, I’m sorry. I didn’t mean to confuse anybody.

Printer Friendly | Permalink | | Top

skids

(1000+ posts) Send PM | Profile | Ignore

Sat Jan-29-05 01:43 AM
Response to Reply #122

125. OK.

Reticle is the less ambiguous spelling, and yes that's a bit of an awkward way to use it, since ridicule also fits there.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Fri Jan-28-05 01:19 AM
Response to Reply #107

110. OH, YOU MUST MEAN THIS.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x300746

Printer Friendly | Permalink | | Top

jkd

(151 posts) Send PM | Profile | Ignore

Fri Jan-28-05 02:36 AM
Response to Reply #110

113. Yes

I have seen these figures subject to change in many posts.

Printer Friendly | Permalink | | Top

TruthIsAll

(1000+ posts) Send PM | Profile | Ignore

Fri Jan-28-05 08:00 AM
Response to Reply #113

114. I changed them once since I used the Nov. 3 numbers and

updated them on Jan. 15. Also, I made a transcription error, which I immediately corrected. I attempt to be as accurate as possible, contrary to the naysays. If I make a mistake in a calculation, I fix it. There have been just TWO changes made since the original.

The first was to get the final vote count. We all want accurate data, don't we?

The second was due to an error (I mixed 2-party vs. total votes in calculating the deviations). We want to fix errors as soon as we see them, don't we?

There was NO effective change in the results other than Bush going from 16 states to 15 beyond the MOE. Of course, you don't like the state MOE's, but they are individually calculated and correct. Forty-three states still deviated to Bush, before and after the update.

There was NO effective change in the results other than Bush going from 16 states to 15 beyond the MOE. Of course, you don't like the state MOE's, but they are individually calculated and correct.

You imply criticism in your statement, noting that I made a few updates to the analysis. But that does not demean my analysis. In fact, you are praising it; the update/fix indicates that I pay attention to detail and strive for accuracy.

That's more than one can say about the naysayers, who only seek to evade and mislead while pretending to be democrats seeking the truth.

I assume you also correct your reports, personal or business, as new data comes in. And that you correct your errors as soon as you find them.

Printer Friendly | Permalink | | Top

berniew1

(1000+ posts) Send PM | Profile | Ignore

Fri Jan-28-05 08:36 AM
Response to Original message

116. Vote machine fraud & malfeasance to reduce minority votes counted explain

the gap between exit polls and "official counts"

Widespread systematic vote machine fraud, and dirty tricks and suppression of minority registration and voting in at least 20 states in 2004 Election:
summary of some of the documentation

Although it was thought that voting inTouchscreen machine counties produces no paper trail, this was not the case this year since the widespread problems found in the 2000 election resulted in 60 non-partisan organizations interested in fair elections forming a consortium in 2004 to provide an election incident reporting system(EIRS) for reporting election irregularities. Thousands of such irregularities were reported in several swing states, especially in touchscreen counties, thus providing a paper trail for election irregularities down to the precinct and even machine level in many cases. From this and other investigations, including the Ohio recount effort, vote machine fraud and systematic dirty tricks and suppression to reduce minority votes has been documented in most states investigated. It is also clear that this was a systematic effort, not isolated local events.

http://www.flcv.com/ussumall.html

Printer Friendly | Permalink | | Top

ROH

(521 posts) Send PM | Profile | Ignore

Sat Jan-29-05 12:42 AM
Response to Reply #116

118. Targeted Emphasis?

Edited on Sat Jan-29-05 01:15 AM by ROH

This DU article:
http://www.democraticunderground.com/articles/05/01/26_wrong.html
mentions that Democrats achieved good results in some Senate, House and Governor elections.

---------------------------

This article:
http://www.wjla.com/news/stories/1104/185641.html
comments on the House elections:

"In the House, Bush's home state was a political killing field for Democrats, four of whom lost their seats after being forced into new, GOP-heavy districts designed by Majority Leader Tom DeLay.

Other than Texas, Democrats took away four seats in Republican hands but lost two of their own. That meant an overall loss rather than a gain that might have provided a psychological boost even if it left them short of the 12 needed to win control."

---------------------------

This webpage details the Senate results:
http://wid.ap.org/campaign2004/senatenew.html

Some notable results:
SD: Thune (R) 50.6%, Daschle (D) 49.4%
FL: Martinez (R) 49.4%, Castor (D) 48.3%
NC: Burr (R) 51.6%, Bowles (D) 47.0%
CO: Salazar (D) 50.7%, Coors (R) 47.2%
WI: Feingold (D) 55.4%, Michels (R) 44.1%

---------------------------

exitpollz.org has interesting comparisons of the Presidential and Senatorial elections in Colorado and Wisconsin:
http://www.exitpollz.org/pages/colorado.html
http://www.exitpollz.org/pages/wisconsin.html
and the Florida and Miami contests:
http://www.exitpollz.org/pages/florida.html

---------------------------

Finally, this article:
http://globalresearch.ca/articles/KEE501A.html
provides a considerable amount of information regarding Ohio, including the comparisons of the Kerry-Edwards votes against the votes recorded for C. Ellen Connally, a comparatively little-known African-American municipal judge from Cleveland who was running as a Democrat for the position of Chief Justice against a well-funded Republican incumbent.

---------------------------

If major election fraud took place, maybe the emphasis of the fraud was placed on the Presidential and some key Senatorial contests (e.g. in SD, FL, NC), while most Senatorial contests (consider CO and WI) and most House, Governor and other elections may have been relatively unaffected?

Printer Friendly | Permalink | | Top

skids

(1000+ posts) Send PM | Profile | Ignore

Sat Jan-29-05 12:55 AM
Response to Reply #118

119. That's the general sense I get, yes.

There were more president-specific and federal-senate-specific reports in EIRS than there were lower ticket switching events.

Printer Friendly | Permalink | | Top

ROH

(521 posts) Send PM | Profile | Ignore

Sat Jan-29-05 01:11 AM
Response to Reply #119

121. In that case...

there should be the tell-tale comparison signs, like the Kerry / Connally comparison and the Colorado Presidential / Senatorial point: http://www.exitpollz.org/pages/colorado.html

This question is posed...
"As we see that the exit poll for the senatorial election in Colorado was spot on! It should be noted that it was the same respondents who were interviewed first on how they had voted in the presidential and then in the senatorial election. So what is the explanation for the much less accurate exit poll for the President?"

What is the answer?

Printer Friendly | Permalink | | Top

skids

(1000+ posts) Send PM | Profile | Ignore

Sat Jan-29-05 01:24 AM
Response to Reply #121

123. I don't have an answer myself.

But if you want answers, you need data. If you'd like to see data collected for the entire country, and analysed by qualified statiticians, then support uscountvotes.org with whatever resources you are willing to contribute.

Printer Friendly | Permalink | | Top

ROH

(521 posts) Send PM | Profile | Ignore

Sat Jan-29-05 01:35 AM
Response to Reply #123

124. Data is essential...

I would like to see exit polls from more of the Senate contests. The CO and WI data certainly provides food for thought.

Printer Friendly | Permalink | | Top

ROH

(521 posts) Send PM | Profile | Ignore

Sat Jan-29-05 10:30 AM
Response to Reply #121

126. Any answers? (n/t)

Printer Friendly | Permalink | | Top

DU AdBot (1000+ posts)

Tue Apr 30th 2024, 08:39 AM
Response to Original message

Advertisements [?]

Top

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.