WHO CAN SOLVE THIS FINAL EXIT POLL PUZZLE: 130,261,391,522,652 BUSH VOTES?

[TruthIsAll]

Of course, the big question is how did Bush magically win the
613 new respondents from the Preliminary Exit Poll of 13047 to
the 13660 Final.

But here is another anomaly, just as strange, in the
tranformation from a Kerry win for 13047 respondents to the
Bush win for 13660.

In calculating the changes in Bush respondents from 13047 to
13660 (based on the percentages) for the demographic
characteristics shown below, there are just FIVE incremental
changes which cover all but three aand which are duplicated
over all categories. 

These are the mysterious five: 130,261,391,522,652 

The three exceptions: 
When Decided (Week, Other) and Voted in 2000 (Other).

For Kerry (shown below) the incremental changes are unique and
not duplicated, as one would expect.

Am I missing something here?
This numerical duplication cannot just be due to chance.
Can anyone determine the formula which explains it?

Bush exit poll respondents based on exit poll percentages
BP1 = Exit Poll 7:38pm 11027   
BP2 = Exit Poll 12:22am 13047
BF = Final 2:05pm 13660

The changes are EXACTLY duplicated to at least 2 decimal
places based on the original calc, but are rounded to the
nearest integer to obtain the number of votes. Since Exit Poll
percentages are rounded to the nearest 1% to begin with, the
differences may be off by one. 

For instance, applying the percentages, the Bush female vote
increased by BF - BP2 = 391.41, which rounds down to 391. 

BF and BP are displayed as as rounded-off integers 6784 and
7176 (difference of 392), even though they compute to 6784.44
and 7175.85 (difference of 391.41). I just want to cover all
bases in case someone catches it. 


	BP1	Chg	BP2	Chg	BF	Chg%	Change from 13047 to 13660 				
Poll	11027	2020	13047	613	13660	4.70%	130	261	391	522	652

GENDER
Male 	5624	1161	6784	391	7176	5.77%			391		
Female 	4962	909	5871	391	6263	6.67%			391		

EDUCATION
No H.S.	5183	949	6132	261	6393	4.26%		261			
H.S. 	5293	970	6263	522	6784	8.33%				522	
Col 	5624	1030	6654	391	7045	5.88%			391		
ColGrad	5569	955	6524	261	6784	4.00%		261			
PostG 	4411	808	5219	522	5741	10.00%				522	

RACE/GENDER
WM	6396	1172	7567	522	8089	6.90%				522	
WF	5734	1050	6784	391	7176	5.77%			391		
NWM	3198	455	3653	261	3914	7.14%		261			
NWF	2426	444	2870	261	3131	9.09%		261			

AGE
18-29 	4742	869	5610	261	5871	4.65%		261			
30-44 	5514	1010	6524	391	6915	6.00%			391		
45-59 	5183	949	6132	522	6654	8.51%				522	
60+	5624	1030	6654	391	7045	5.88%			391		

INCOME
0-$15K 	3639	667	4306	391	4697	9.09%			391		
$15-30 	4301	788	5088	391	5480	7.69%			391		
$30-50 	5183	949	6132	261	6393	4.26%		261			
$50-75 	5844	1071	6915	391	7306	5.66%			391		
75-100 5514	1010	6524	652	7176	10.00%					652
100-150 5844	1071	6915	522	7437	7.55%				522	
150-200 5844	1071	6915	652	7567	9.43%					652
200+ 	6396	1172	7567	652	8220	8.62%					652

IDEOLOGY
Lib 	1323	242	1566	130	1696	8.33%	130				
Mod 	4631	718	5349	522	5871	9.76%				522	
Cons 	9042	1656	10699	261	10959	2.44%		261			
Prot	6175	1131	7306	391	7698	5.36%			391		
Cath 	5403	990	6393	391	6784	6.12%			391		
Jewish	2536	465	3001	261	3262	8.70%		261			
Other 	2205	404	2609	391	3001	15.00%			391		
None 	3198	586	3784	261	4045	6.90%		261			

MILITARY
Yes	6065	1111	7176	261	7437	3.64%		261			
No	5072	929	6002	391	6393	6.52%			391		

WHEN DECIDED
Today 	4521	698	5219	652	5871	12.50%					652
3Days 	4631	718	5349	130	5480	2.44%	130				
Week 	5624	1030	6654	0	6654	0.00%					
Month 	4190	768	4958	783	5741	15.79%					
Before 	5514	1010	6524	391	6915	6.00%			391		

REGION
East	4411	938	5349	261	5610	4.88%		261			
Midw	5403	990	6393	261	6654	4.08%		261			
South	5955	1091	7045	522	7567	7.41%				522	
West	4962	909	5871	522	6393	8.89%				522	

PARTY-ID
Dem 	992	182	1174	261	1435	22.22%		261			
Rep 	10145	1858	12003	130	12134	1.09%	130				
Ind	4962	909	5871	391	6263	6.67%			391		

VOTED IN 2000
No	4301	1049	5349	522	5871	9.76%				522	
Gore	882	162	1044	261	1305	25.00%		261			
Bush	9924	1818	11742	130	11873	1.11%	130				
Other	1434	263	1696	1044	2740	61.54%	


The Kerry numbers:

	KP1	Chg	KP2	Chg	KF	Chg%
	11027	2020	13047	613	13660	4.70%

GENDER
Male 	5183	949	6132	-122	6010	-1.98%
Female 	5955	1091	7045	-79	6967	-1.12%

EDUCATION
No H.S.	5734	1050	6784	46	6830	0.67%
H.S. 	5624	1030	6654	-234	6420	-3.51%
Col 	5183	949	6132	152	6284	2.47%
ColG	5348	914	6263	21	6284	0.34%
PostG 	6396	1172	7567	-54	7513	-0.72%

RACE/GENDER
WM	4521	828	5349	-295	5054	-5.52%
WF	5183	949	6132	-122	6010	-1.98%
NWM	7719	1284	9002	150	9152	1.66%
NWF	8491	1555	10046	199	10245	1.98%


AGE
18-29 	6175	1131	7306	70	7376	0.96%
30-44 	5403	990	6393	-109	6284	-1.71%
45-59 	5624	1030	6654	-97	6557	-1.46%
60+	5293	970	6263	21	6284	0.34%


INCOME
0-15K 	7278	1333	8611	-5	8606	-0.06%
15-30 	6506	1192	7698	88	7786	1.15%
30-50 	5734	1050	6784	46	6830	0.67%
50-75 	4962	909	5871	3	5874	0.05%
75-100 	5403	990	6393	-246	6147	-3.85%
100-150 4962	909	5871	-134	5737	-2.28%
150-200 5183	949	6132	-395	5737	-6.44%
200+ 	4521	828	5349	-568	4781	-10.62%

IDEOLOGY
Lib 	9593	1627	11220	391	11611	3.48%
Mod 	6285	1151	7437	-60	7376	-0.81%
Cons 	1764	323	2088	-39	2049	-1.85%


RELIGION
Prot	4742	869	5610	-146	5464	-2.61%
Cath 	5514	1010	6524	-103	6420	-1.58%
Jewish	8491	1555	10046	62	10108	0.62%
Other 	8270	1515	9785	323	10108	3.30%
None 	7719	1414	9133	19	9152	0.21%


MILITARY
YES	4742	869	5610	-10	5601	-0.17%
NO	5844	1071	6915	-85	6830	-1.23%

WHEN DECIDED
Today 	5955	960	6915	188	7103	2.72%
3Days 	5955	960	6915	598	7513	8.65%
Week 	5293	970	6263	294	6557	4.70%
Month 	6726	1102	7828	-452	7376	-5.77%
Before 	5514	1010	6524	-240	6284	-3.68%

REGION
East	6396	1172	7567	82	7650	1.09%
Midw	5514	1010	6524	33	6557	0.51%
South	4962	779	5741	-3	5737	-0.06%
West	5844	1071	6915	-85	6830	-1.23%

PARTY ID
Dem 	9924	1818	11742	415	12157	3.54%
Rep 	772	141	913	-94	820	-10.26%
Ind	5734	1050	6784	-91	6693	-1.34%

HOW VOTED IN 2000
No	6506	931	7437	-60	7376	-0.81%
Gore	10035	1838	11873	421	12294	3.55%
Bush	992	182	1174	55	1229	4.70%
Other	7168	1313	8481	1218	9699	14.36%
				

[stellanoir]

1. 'cause he rigged it silly. . .

TIA glad to see you're still at it. Keep it coming though!

[TruthIsAll]

7. THE GRAPH

[TruthIsAll]

9. HERE IS THE MATH SEQUENCE FORMULA. WHAT DOES IT MEAN?

THE FORMULA, USING THE EXACT CALCULATED NUMBERS BEFORE ROUNDOFF:

X(I) = X(I-1) + 130.47


X(1) = 0
X(2) = X(1) + 130.47 = 130.47 (130 rounded)
X(3) = X(2) + 130.47 = 260.94 (261 rounded)
X(4) = X(3) + 130.47 = 391.41 (391 rounded)
X(5) = X(4) + 130.47 = 521.88 (522 rounded)
X(6) = X(5) + 130.47 = 652.35 (652 rounded)

[TruthIsAll]

12. WOW! 13047 RESPONDENTS AND 130.47 INCREMENTAL (1%) VOTES!

There were 13047 respondents in the Preliminary Exit Poll!

130.47 is EXACTLY 1.0% of 13047!

They applied multiples from 1% to 5% across the board

1% of 13047 = 130.47
2% of 13047 = 260.94
3% of 13047 = 391.47
4% of 13047 = 521.88
5% of 13047 = 652.35


COINCIDENCE?
HARDLY.

[dzika]

13. Wow.

I was not prepared to have my mind blown today.

I can't even get my head around how there could be a percentage increase with exact integers between the last reported unweighted poll and the final weighted poll.

Taking off the tin foil hat, drinking the kool-aid, believing Mitofsky, chanting to myself: "There was no fraud, there were reluctant responders, the exit polls were flawed..."

...and then my head explodes.


I'm having a hard time going back through the looking glass on this one.

[glitch]

21. What dzika said.

Wow. Really, really wow.

[bleever]

26. And *'s gain of 782 equals 6%,

though strictly speaking it's 782.94, which we would round up.

However, if we take the rounded down vote tallies for 3%, and double it, we do indeed get 782.

And how is it that there were exactly 391 male and female votes added to * because of the addition of 613 to the final total? Exactly identical gains from both sexes?

[LatePeriduct]

27. Or was it....

Purged voters rounded down from the original sample size?

Because according to the theory out there now, only a certain select group are allowed to represent the population.

[TruthIsAll]

29. ERROR, MEA CULPA: THERE IS NO SIMPLE 1% FACTOR; USE THIS INSTEAD

There is no simple 1% adjustment factor.

I'm sorry for the confusion.
Errors are made in the spreadsheet modeling process.
The key is to catch and fix them before posting.
I missed this one.

I have made errors before and will again.
But I always fix them.

THESE ARE THE CORRECT NUMBERS FROM ANOTHER THREAD
	Prel.	Final		122.26	million votes										
	13047	13660	Chg	KP	KF	KChg	BP	BF	BChg	KVP	BVP	KVF	BVF	KChg	BChg
GENDER	(no change from 11027 to 13047)														
Male
	46%	46%	0%	47%	44%	-3%	52%	55%	3%	26.43	29.24	24.75	30.93	-1.69	1.69
Female
	54%	54%	0%	54%	51%	-3%	45%	48%	3%	35.65	29.71	33.67	31.69	-1.98	1.98
	100%	100%				-3.0%			3.0%	62.08	58.95	58.42	62.62	-3.67	3.67
										50.78%	48.22%	47.78%	51.22%	-3.00%	3.00%
															
															
EDUCATION	(no change from 11027 to 13047)														
No
H.S.	4%	4%	0%	52%	50%	-2%	47%	49%	2%	2.54	2.30	2.45	2.40	-0.10	0.10
H.S.
	22%	22%	0%	51%	47%	-4%	48%	52%	4%	13.72	12.91	12.64	13.99	-1.08	1.08
College
	31%	32%	1%	47%	46%	-1%	51%	54%	3%	17.81	19.33	18.00	21.13	-0.38	1.14
ColGrad	26%	26%	0%	48%	46%	-2%	50%	52%	2%	15.26	15.89	14.62	16.53	-0.64	0.64
PostG
	17%	16%	-1%	58%	55%	-3%	40%	44%	4%	12.05	8.31	10.76	8.61	-0.62	0.83
	100%	100%				-2.40%			3.00%	61.39	58.75	58.46	62.65	-2.81	3.78
										50.21%	48.05%	47.82%	51.24%	-2.30%	3.09%
															
	13047	13660	Chg	KP	KF	KChg	BP	BF	BChg	KVP	BVP	KVF	BVF	KChg	BChg
RACE AND GENDER	(no change from 11027 to 13047)														
WM	36%	36%	0%	41%	37%	-4%	58%	62%	4%	18.05	25.53	16.29	27.29	-1.76	1.76
WF	41%	41%	0%	47%	44%	-3%	52%	55%	3%	23.56	26.07	22.06	27.57	-1.50	1.50
NWM	10%	10%	0%	69%	67%	-2%	28%	30%	2%	8.44	3.42	8.19	3.67	-0.24	0.24
NWF	13%	13%	0%	77%	75%	-2%	22%	24%	2%	12.24	3.50	11.92	3.81	-0.32	0.32
	100%	100%				-2.75%			2.75%	62.28	58.51	58.45	62.34	-3.83	3.83
										50.94%	47.86%	47.81%	50.99%	-3.13%	3.13%
	(no change from 11027 to 13047)														
AGE	13047	13660	Chg	KP	KF	KChg	BP	BF	BChg	KVP	BVP	KVF	BVF	KChg	BChg
18-29
	17%	17%	0%	56%	54%	-2%	43%	45%	2%	11.64	8.94	11.22	9.35	-0.42	0.42
30-44
	27%	29%	2%	49%	46%	-3%	50%	53%	3%	16.17	16.51	16.31	18.79	-0.99	0.99
45-59
	30%	30%	0%	51%	48%	-3%	47%	51%	4%	18.71	17.24	17.61	18.71	-1.10	1.47
60+	26%	24%	-2%	48%	46%	-2%	51%	54%	3%	15.26	16.21	13.50	15.84	-0.64	0.95
	100%	100%				-2.5%			3.0%	61.78	58.89	58.64	62.69	-3.14	3.83
										50.53%	48.17%	47.96%	51.28%	-2.57%	3.13%
INCOME	(no change from 11027 to 13047)														
0-$15K
	9%	8%	-1%	66%	63%	-3%	33%	36%	3%	7.26	3.63	6.16	3.52	-0.33	0.33
$15-30
	15%	15%	0%	59%	57%	-2%	39%	42%	3%	10.82	7.15	10.45	7.70	-0.37	0.55
$30-50
	22%	22%	0%	52%	50%	-2%	47%	49%	2%	13.99	12.64	13.45	13.18	-0.54	0.54
$50-75
	23%	23%	0%	45%	43%	-2%	53%	56%	3%	12.65	14.90	12.09	15.75	-0.56	0.84
$75-100
	13%	14%	1%	49%	45%	-4%	50%	55%	5%	7.79	7.95	7.70	9.41	-0.64	0.79
$100-150
	11%	11%	0%	45%	42%	-3%	53%	57%	4%	6.05	7.13	5.65	7.67	-0.40	0.54
$150-200
	4%	4%	0%	47%	42%	-5%	53%	58%	5%	2.30	2.59	2.05	2.84	-0.24	0.24
$200+
	3%	3%	0%	41%	35%	-6%	58%	63%	5%	1.50	2.13	1.28	2.31	-0.22	0.18
	100%	100%				-3.38%			3.75%	62.36	58.12	58.84	62.38	-3.30	4.02
										51.01%	47.54%	48.13%	51.02%	-2.70%	3.29%
IDEOLOGY	(no change from 11027 to 13047)														
	13047	13660	Chg	KP	KF	KChg	BP	BF	BChg	KVP	BVP	KVF	BVF	KChg	BChg
Liberal
	22%	21%	-1%	86%	85%	-1%	12%	13%	1%	23.13	3.23	21.82	3.34	-0.27	0.27
Moderate
	45%	45%	0%	57%	54%	-3%	41%	45%	4%	31.36	22.56	29.71	24.76	-1.65	2.20
Conservative
	33%	34%	1%	16%	15%	-1%	82%	84%	2%	6.46	33.08	6.24	34.92	-0.40	0.81
	100%	100%				-1.67%			2.33%	60.95	58.87	57.77	63.01	-2.32	3.28
										49.85%	48.15%	47.25%	51.54%	-1.90%	2.68%
RELIGION	(no change from 11027 to 13047)														
Prot	53%	53%	0%	43%	40%	-3%	56%	59%	3%	27.86	36.29	25.92	38.23	-1.94	1.94
Catholic
	27%	27%	0%	50%	47%	-3%	49%	52%	3%	16.51	16.17	15.51	17.17	-0.99	0.99
Jewish	3%	3%	0%	77%	74%	-3%	23%	25%	2%	2.82	0.84	2.71	0.92	-0.11	0.07
Other
	7%	7%	0%	75%	74%	-1%	20%	23%	3%	6.42	1.71	6.33	1.97	-0.09	0.26
None
	10%	10%	0%	70%	67%	-3%	29%	31%	2%	8.56	3.55	8.19	3.79	-0.37	0.24
	100%	100%				-2.60%			2.60%	62.17	58.56	58.67	62.07	-3.50	3.51
										50.85%	47.90%	47.99%	50.77%	-2.86%	2.87%
MILITARY	(no change from 11027 to 13047)														
	13047	13660	Chg	KP	KF	KChg	BP	BF	BChg	KVP	BVP	KVF	BVF	KChg	BChg
Yes
	18%	18%	0%	43%	41%	-2%	55%	57%	2%	9.46	12.10	9.02	12.54	-0.44	0.44
No
	82%	82%	0%	53%	50%	-3%	46%	49%	3%	53.13	46.12	50.13	49.12	-3.01	3.01
	100%	100%				-2.5%			2.5%	62.60	58.22	59.15	61.67	-3.48	3.48
										51.20%	47.62%	48.38%	50.44%	-2.84%	2.84%
WHEN DECIDED	(no change from 11027 to 13047)														
	13047	13660	Chg	KP	KF	KChg	BP	BF	BChg	KVP	BVP	KVF	BVF	KChg	BChg
Today
	6%	5%	-1%	53%	52%	-1%	40%	45%	5%	3.89	2.93	3.18	2.75	-0.07	0.37
Last3Days
	3%	4%	1%	53%	55%	2%	41%	42%	1%	1.94	1.50	2.69	2.05	0.07	0.04
Last Week
	2%	2%	0%	48%	48%	0%	51%	51%	0%	1.17	1.25	1.17	1.25	0.00	0.00
Last Month
	10%	10%	0%	60%	54%	-6%	38%	44%	6%	7.34	4.65	6.60	5.38	-0.73	0.73
Before
	79%	79%	0%	50%	46%	-4%	50%	53%	3%	48.29	48.29	44.43	51.19	-3.86	2.90
	100%	100%				-1.80%			3.00%	62.63	58.62	58.07	62.62	-4.60	4.03
										51.23%	47.95%	47.50%	51.22%	-3.76%	3.30%
															
	13047	13660	Chg	KP	KF	KChg	BP	BF	BChg	KVP	BVP	KVF	BVF	KChg	BChg
REGION	(no change from 11027 to 13047)														
East	22%	22%	0%	58%	56%	-2%	41%	43%	2%	15.60	11.03	15.06	11.57	-0.54	0.54
Midwest	26%	26%	0%	50%	48%	-2%	49%	51%	2%	15.89	15.58	15.26	16.21	-0.64	0.64
South	31%	32%	1%	44%	42%	-2%	54%	58%	4%	16.68	20.47	16.43	22.69	-0.76	1.52
West	21%	20%	-1%	53%	50%	-3%	45%	49%	4%	13.61	11.55	12.23	11.98	-0.77	1.03
	100%	100%				-1.98%			3.00%	61.78	58.62	58.98	62.45	-2.70	3.72
										50.53%	47.95%	48.24%	51.08%	-2.21%	3.04%
															
PARTY ID	(Rep 36% in 11027)														
Democrat
	38%	37%	-1%	90%	89%	-1%	9%	11%	2%	41.81	4.18	40.26	4.98	-0.46	0.93
Republican
	35%	37%	2%	7%	6%	-1%	92%	93%	1%	3.00	39.37	2.71	42.07	-0.43	0.43
Indep	27%	26%	-1%	52%	49%	-3%	45%	48%	3%	17.17	14.85	15.58	15.26	-0.99	0.99
	100%	100%				-1.67%			2.00%	61.97	58.40	58.55	62.30	-1.88	2.35
										50.69%	47.77%	47.89%	50.96%	-1.54%	1.92%
	13047	13660	Chg	KP	KF	KChg	BP	BF	BChg	KVP	BVP	KVF	BVF	KChg	BChg
VOTED IN 2000	(no change from 11027 to 13047)														
No	17%	17%	0%	57%	54%	-3%	41%	45%	4%	11.85	8.52	11.22	9.35	-0.62	0.83
Gore	38%	37%	-1%	91%	90%	-1%	8%	10%	2%	42.28	3.72	40.71	4.52	-0.46	0.93
Bush	41%	43%	2%	9%	9%	0%	90%	91%	1%	4.51	45.11	4.73	47.84	0.00	0.50
Other	4%	3%	-1%	65%	71%	6%	13%	21%	8%	3.18	0.64	2.60	0.77	0.29	0.39
	100%	100%				0.50%			3.75%	61.81	57.99	59.27	62.49	-0.79	2.65
		
 
 

[LatePeriduct]

31. Numbers there

Are still all totally impossible, can't have happened.

[TruthIsAll]

32. MAX% OF 2004 VOTERS WHO COULD HAVE VOTED BUSH IN 2000=50.45/122.6=41.26%

.

[rzemanfl]

2. If you went to H&R Block with FIFTEEN deductible expenses that

were each exactly $391.00, and you didn't have a good explanation of why that was, they probably would refuse to do your return. I don't know anything about the math, but recall that shortly after the election there were posts about something that could be used to determine when numbers were not random

[RubyCat]

4. I found this page about Benford's Law

Here's a good article from the New York Times, August 4th 1998, that describes one of the methods used to flag suspicious data. Here is a sample quote from the article: "Dr. Hill is one of a growing number of statisticians, accountants and mathematicians who are convinced that an astonishing mathematical theorem known as Benford's Law is a powerful and relatively simple tool for pointing suspicion at frauds, embezzlers, tax evaders, sloppy accountants and even computer bugs."


http://www.rexswain.com/benford.html

[burned]

3. yes!

This is the first TIA post I 'got' on contact.
I think it should be recommended so the rest of the numerically challenged will get a chance to see it..

[rzemanfl]

5. Here is part of the answer:

It's quite simple really, 261 non-white Jewish-Republican high
school dropout military veterans who voted for Gore in 2000 switched
their votes to Bush, reflecting a national pattern amongst non-white Jewish-Republican high school dropout military veterans who voted for Gore in 2000.

[TruthIsAll]

8. LOLOLOLOLOLOL

LOL

[BeFree]

6. Bullets...

....from a smoking gun.

Somebody raise another red flag, eh? There be some about that haven't read the first one yet.

[skids]

10. More on the sample size.

http://nashuaadvocate.blogspot.com/2005/01/news-election-2004-nep-adds-hundreds.html#comments

In addition, it should be noted that the number of entries in the raw data release for the national poll is not 13660, it is 13719.

Another note that should be made is that it was stipulated by NEP that some of the entries were counted more than once, because their techs were lazy. I don't remember the exact group that this was, but I think it may have been telephone surveys. They counted them more than once because of a shortsighted design: they had a flip-switch in the data which told which version of the survey a person took. However, some people (again I think it may have been some of the telephone surveys) took a survey which included questions from more than one of those survey classes, so instead of fixing it right, they made duplicate entries in the data for these people -- one for each survey type. However I do not remember which poll this effected -- could have been the national, or could have been the states.

As an aside, the _All file which contains the merged results from the states and the national poll contains 77006 entries. In both files, all entries have a nonzero weight, so none are zeroed out, though some are weighted as low as .035 (they are worth 3.5% of a person) and as high as 11.48 (they are worth 11 and a half people)

Now as to whether all this matters -- is there a statement from NEP that when the additional surveys were added that they did not re-weight the older surveys at the same point? Because if not, then these bets are off. That the addition of the new surveys cooincided with a reweighting of the entire dataset provides them with plausible deniability on this count.

[TruthIsAll]

11. How does that explain the formula sequence across all demographics?

There were 13047 respondents in the Preliminary Exit Poll!

130.47 is EXACTLY 1.0% of 13047!

COINCIDENCE?
HARDLY.

Layziness?
Coincidence?

Or Design?

[LatePeriduct]

17. It doesn't.

I'm no expert statistician or anything, but I know someone who is pretty well versed to it.

According to evidence of programmers blogging before the election, they talked about something called the "Dunbar" theory.

In the theory they chronicle how social networking relationships, the larger they get and the more answers received can not be represenative of the population.

They advocated that instead, what is neccesary is data reduction, and that will be represenative of the population. Removing they state what are existing biases.

But their way of working the numbers is the exact reverse of how a democracy works.

They essentially are stating the mean is in-precise and that a minority over-rules all.

Aka: Caged voters purged off the rolls. It looks like they re-weighted the sample sizes after the voters were purged instead of before, so no one would get caught.

This is a really bad dilemma. The BOE in every single main state has most likely caged voters and participated in fraud.

[Joe Chi Minh]

19. Surely, if the

formula - or do they call it an "algorithm"? - can be established, it that will mean game, set and match. Check mate.

[LatePeriduct]

20. You need more then that.

I know who wrote the column on that blog and this is a posted excerpt from it.

"In my post about the Dunbar Number I offered some evidence on the levels of satisfaction of various group sizes based on some empirical data from online games. There I was able to show that even though the Dunbar Number might predict a mean group size of 150 for humans, that in fact for non-survival oriented groups the mean wassignificantly less, probably between 60 to 90.

I also offered a second hypothesis, that there is a dip insatisfaction level of groups at around the size of 15. Unfortunately, I could only offer anecdotal evidence that this threshold existed. My personal belief was that this dip was caused by not enough "attention" being given to everyone and that group gatherings of this size risk becoming too noisy, too boring, too long, or some combination thereof.

Yet groups of this size are not large enough to allow for different perspectives (i.e. insufficient requisite variety) or for other group processes to come into play.Dunbar Triage: Too Many Connections. As someone who now has over 171 professional "connections" in myLinkedIn Profile, 198 "friends" on Orkut, many more non-intersecting friends and acquaintances on Tribe.Net, LiveJournal, and othersocial networking services, as well as a plethora of correspondentsthat I only interact with via email, I am trying reconcile amismatch between my connections and my own Dunbar Number.How do I maintain meaningful relationships with over 300 people?

Venture Capitalist Jeff Nolan relates similar concerns:"It strikes me that the social networking theory holds that the more volume you have, the bigger your network will become by introducing degrees of separation roughly along the lines of Metcalfe's Law. I disagree, human networks do not grow in value by multiplying, but rather by reduction."

Taken out of http://rigged-aggregators.blogspot.com

In order to conclusively prove it, you will need the ballot counting algorithm that TrueVote MD of Maryland and other now has in their posession.

You will then need the tables and timestamps from the audit archives for each state's board of elections.

It would show exactly who was filtered out, where, and when because it saves it in a cache.

Then you would need the highest authority to fight the people in congress to get it litigated, because all the republic freepers are sure to label it disinformation until it finally goes away and another made up Jeff Gannon scandal comes out.

[skids]

24. I'm not trying to explain it.

Just give some perspective and background information. NEP readily admits they toyed with the weights -- they planned to do so from the start, as it was part of the design of the survey to provide a final total that was adjusted to the final vote. Your sharp eyed analysis would suggest that on the Democratic side, they toyed with a lot of subgroup weights in very slight amounts, whereas on the Republican side, they "bumped" the weights on a very few subgroups by 1% increments. The 1% thing is fishy, no doubt, and the fact that the subgroup weighting is done differently on each side is also fishy, but it isn't like NEP couldn't figure out a way to explain it if they were put on the spot about it.

What makes the numbers especially fishy is that if you sum up the shift in each category, it should be the same, with a small deviation for those who left questions blank (the deviations could have an excuse to get very large if a question appeared on one survey but not on another, but a large deviation would suggest a perhaps intentionally extreme weighting of respondents that left questions blank.)

If we stick to questions where we know that almost all the respondents answered (individual N sizes for each question are available to make this assessment more accurate, and are provided below.) Just doing a few here:

Bush:

Region: 261 + 261 + 522 + 522 = 1566
Age: 391 + 522 + 391 + 261 = 1565
Gender: 391 + 391 = 782
Race/Gender: 522 + 391 + 261 + 261 = 1435

Kerry:

Region: 82 + 33 - 3 - 85 = 27
Age: 70 - 109 - 97 + 21 = -115
Gender: 0 - 122 - 79 = -201
Race/Gender: 199 + 150 - 295 - 122 = -68

...that shouldn't happen on either set of numbers, unless N varies or there are non-answers. There should be the same number of "people units" added or subtracted from each category. How can you gain 391 males and 391 females, and at the same time, when broken down by race, gain 652 females and 781 males? It isn't possible. A systematic error that excluded people or doublecounted them should have done so in all questions. They would have to cop to "accidentally" doublecounting or undercounting only on certain sets of questions, which is an incredibly sloppy mistake for so-called professionals.

But we still have the problem with N varying. How much did N for each above question really vary? Well, for the final weighting, Mitofski's own numbers are:

Gender: 13,600 = 60 less respondents
Race/Gender: 13,419 = 241 less respondents
Age: 13,580 = 80 less respondents
Region: 13,660 = this answer exists for everyone (except, apparently, the extra 59 entries in the data file which totals 13,719 lines.)

...that would mean that the "region" can be taken as gospel -- the total shift in each category should match the total shift in the region, and if it doesn't, that means that in order to get the numbers to weight the way they did, they had to extremely heavily weight people that did not respond to one question or another on the survey.

Now what if they say that they held back surveys with non-responses and didn't process them in the weighting until the end? That could be a potential excuse. A more thorough working of the numbers might be able to preempt that excuse, however.

In theory, if we find a few questions with very high N's, but where the total shift is different from the "Region" shift, we will also find that the weight assigned to the people who left the question blank is very high or very low.

Don't have time to do that now, though.

[Old and In the Way]

14. The 1% solution..............well done TIA.

[TeeYiYi]

15. When it comes to computer voter fraud . . .

. . . that's just plain lazy. They'd have been better off to make up some numbers rather than let the computer do all the work. :eyes:

TYY

[LatePeriduct]

23. Uh no.

Bush & the fellowship folks are not that smart.

Plus junior bush probably knew nothing more than yeah, someone was going to guarantee him the election.

His dad obviously knew how all the planned operations were done and oversaw fraud and helped Khashoggi and others pay the fee.

Hired a bunch of programmers who know math to basically rig things making it appear transparent, and forgot to cover up the tracks with Florida.

Tom Feeney and company have too big of a mouth. They already know about the shady secret meeting deals.

So they knew how easy it would be if provisional ballot voters were caged since they created the provision rule, and used ballot jam filtering on all the election numbers.

I've seen it explained enough that I get it. The fraud was transparent because they used datamining aggregators.

Plus how else would it be possible for the exit poll denomination to flip?

On the hard exit poll results at exitpollz.org which I found a while ago, religion, ideology and other denominations switched places between the official and the final weighted exit polls. How else could that even be accomplished.

They had to have filtered them out using the good old "Code 200" garbage and other nonsense. Anyone ever heard of nomisaka mohandu? He voted in the 2004 elections. Nobody I know of ever heard of 'em either.

[RubyCat]

16. No wonder Mitofsky kept his methodology such a secret.

The way he went from the second exit poll to the final exit poll was totally bogus.

You've done a great job reverse engineering the fraud TIA.

[Maraya1969]

18. You need to take this info somewhere

I have NO idea where but someone needs to see it!

Unfortunately this administration has gotten away so much already it will probably get away with this, unless some politician has the guts to run with it.

Barbara Boxer?

:kick:

[donkeyotay]

22. I can't solve your riddle, yet

but I'm sure as hell going to think about it. You are a patriot, TIA. You have worked so long on this, and once again, you have found statistical evidence of a corrupt election.

Really fine work. Dedication. You make slackers like myself hang our heads.

[bleever]

25. Have we addressed why, when they added

613 respondents, * gained 782 votes?

Is that just a patent absurdity, or am I missing something?

And brilliant work here, TIA. I know there's more to be teased out of this. Thanks.

[Maraya1969]

28. Kick

[intensitymedia]

30. AMAZING! Great work TIA - n/t

[autorank]

33. Welcome to DU!!! Isn't this a cool forum? Cutting edge work by TIA!!!

:hi: