HEY MP, WHY DON'T YOU WRITE ABOUT THIS MODEL?

[TruthIsAll]

that is
if
you
can
understand
it...

oh, that was a low blow wasn't it?
Let's see if Lampley tells the ref to penalize me.

I understand.

You are too constrained by the maximization of profit to look at an constrained optimization analysis which puts the Lie to rBr.


http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x375366

[LightningFlash]

1. TIA, I question why Febble requires variance in all precincts...

Why is it required to prove variance in all precincts, when the actual Cencus data simply refutes that?

Hmmmm someone seems to be "reluctant" to take the challenge, that is very fascinating for me....Does the variance also mean it's required to be random? I don't see the "east RBR" theory being random, that's odd....
;)

[Internut]

2. Please show how

the Census Survey data refutes the existence of variance in all precincts.

[LightningFlash]

3. The optimization analysis refutes this.

Just by simply crossing the hard data from http://exitpollz.org, with that shown cencus data and their groupings. http://www.census.gov/population/www/socdemo/voting/cps2004.html

Unless, of course, you make the far-reaching assumption that 55% of refusers nationwide stood up for Bush and that only in the east, where the pattern was not random, was it distributed.

;)

[Internut]

4. You keep speaking in generalities.

Can you show WHICH census (note the spelling) data, combined with WHICH exit poll data, using which formula, shows what.

[LightningFlash]

5. You keep speaking in circles.

I don't need to point out exactly where, simply see the optimization analysis. It refutes everything and can be found on this page.

If you don't have the ability to look that up, then surely you aren't even worth the time. It's very easy to plug in the numbers, and receive your result.

No matter how you vary the data, even if you consecutively double the margin of sampling error, 55% of refusers are required. That might as well be over 60% of vote refusters, in the ENTIRE 2004 election, because Mitofsky also re-adjusts the data to show an increase in Kerry refusers.

http://www.census.gov/population/socdemo/voting/cps2004/tab06-1.xls
http://www.census.gov/population/socdemo/voting/cps2004/tab14.xls
http://www.census.gov/population/socdemo/voting/cps2004/tab13.xls
http://www.census.gov/population/socdemo/voting/cps2004/tab15d.xls

[Internut]

6. Just pointing to the Census survey

and saying "see for yourself" is not sufficient.

You have to show how that data supports your arguments. So far you have not even attempted to do so.

[LightningFlash]

8. Well, I figured it would be simple for you to find it.

My mistake, I guess.

U.S. Total Pct Pct MoE (000) 1/sqrt(n) Diff
.Total 125,736 58.3 100% 0.30% 377 0.28% 0.02%

.Male 5
8,455 56.3 46.49% 0.40% 234 0.41% -0.01%
.Female
67,281 60.1 53.51% 0.40% 269 0.39% 0.01%

.White alone
106,588 60.3 84.77% 0.30% 320 0.31% -0.01%

..White non-Hispanic alone
99,567 65.8 79.19% 0.30% 299 0.32% -0.02%

.Black alone
14,016 56.3 11.15% 1.10% 154 0.84% 0.26%
.Asian alone
2,768 29.8 2.20% 1.70% 47 1.90% -0.20%
.Hispanic (of any race)
7,587 28.0 6.03% 1.20% 91 1.15% 0.05%

.White alone or in combination
107,930 60.3 85.84% 0.30% 324 0.30% 0.00%
..White non-Hispanic alone or in combination
100,726 65.7 80.11% 0.30% 302 0.32% -0.02%
.Black alone or in combination
14,324 56.1 11.39% 1.10% 158 0.84% 0.26%
.Asian alone or in combination
2,980 30.7 2.37% 1.70% 51 1.83% -0.13%

Table 4a. Reported Voting and Registration of the Total
Voting-Age Population, by Sex, Race and Hispanic Origin, for
States: November 2004

Not that difficult to do this with any of the graphs.

[Internut]

9. You really have to be more specific -

what are you trying to show with this (and you gotta learn to format better)

Here is my response to you on the other thread:


Demographics:

Census Survey Data          | Final Exit Poll 
(from the "Voted" column    | (13,660 respondents) 
----------------------------|-------------------------------------
Male    58.45/125.73  = 46% |     46%  
Female	67.281/125.73 = 54% |     54% 
                            |
White   106.59/125.73 = 79% |     77% (36+41)
Non-White               21% |     22% (10+12)

That's a pretty good match. What exactly are you seeing in the census data that would show that the final exit poll's numbers are "cooked"?

[LightningFlash]

10. Because we are expected to believe instead...

That women, in the final weighting, refused and these Kerry voters who had refused must therefore tip the scale.

Census Gender statistics matched Exit Poll weights to within
0.5%:
Female Male Votes (000)
Exit 54% 46% 122,267
Census 53.51% 46.49% 125,736
Vote 67,281 58,455

The tables below give various scenarios and probabilities
using Census gender voting statistics and the 13047 exit poll
voting percentages.

Kerry ties Bush when he gets 52.75% of the female vote and
45.75% of the male vote. This is the breakeven (B/E)
scenario.

The probability that Bush would at least tie in the popular
vote is 0.71% (1 in 140).

The probability that Bush would go from 48.25% in the poll to
his recorded vote of 50.25% is
1 in 2,083,900.

The probability that Bush would go from 47.75% in the poll to
his recorded vote of 50.25% is
1 in 485,887,839

Kerry Vote scenarios
13047 B/E 13660
Male 47.5% 47.0% 46.5% 46.0% 45.75% 45.5% 45.0% 44.5% 44.0%
Female 54.5% 54.0% 53.5% 53.0% 52.75% 52.5% 52.0% 51.5% 51.0%

Votes:
Male 27766 27474 27182 26889 26743 26597 26305 26012 25720
Female 36668 36332 35995 35659 35491 35323 34986 34650 34313

Total 64434 63806 63177 62548 62234 61920 61291 60662 60034
Kerry% 51.25% 50.75% 50.25% 49.75% 49.50% 49.25% 48.75% 48.25% 47.75%




Bush Vote scenarios
13047 B/E 13660
Male 51.5% 52.0% 52.5% 53.0% 53.25% 53.5% 54.0% 54.5% 55.0%
Female 44.5% 45.0% 45.5% 46.0% 46.25% 46.5% 47.0% 47.5% 48.0%

Votes:
Male 30104 30397 30689 30981 31127 31273 31566 31858 32150
Female 29940 30276 30613 30949 31117 31286 31622 31958 32295

Total 60044 60673 61302 61930 62245 62559 63188 63816 64445
Bush% 47.75% 48.25% 48.75% 49.25% 49.50% 49.75% 50.25% 50.75% 51.25%

Prob 0.00% 0.00% 0.00% 0.16% 0.71% 2.50% 16.35% 50.00% 83.65%
1 in 486mm 2.1mm 22,577 609 140 40 6 2 1


Tell me are we expected to believe that VOTERS overall, through-out the election refused and lied? That includes Kerry(female) and Bush(female) I somehow doubt this very much....

[Internut]

11. First of all, yes, around 50% of the voters

overall refused to answer exit pollers' questions. That's just a fact. You are free not to believe it.

Second - you are parroting the above numbers without even an iota of understanding. I showed you that the census data closely matches the final exit poll data as well. Does that prove the final exit poll's correctness?

[LightningFlash]

12. So are you seriously advocating that....

55% of america REFUSED to respond to exit polls for Bush?

That's in the ultimatum, what it would take. Nearly all of them women. At the same time, it takes the oversampling of the same women on the opposite side.

You don't seem to grasp the failure in the final exit-poll weightings very easy, do you?

It's not just gender which virtually switched places. In order for this to take place, so many of the poll results had to have been *DROPPED*

Are you still claiming this is reality?

" VOTE BY RELIGION BUSH
KERRY NADER
TOTAL 2004 2000 2004 2004

Protestant (53%) 56% +0 43% 1%

Catholic (27%) 49% +2 50% 1%

Jewish (3%) 23% +4 77% *

Other (7%) 20% -8 75% 4%

None (11%) 29% n/a 70% 1%

VOTE BY RACE BUSH
KERRY NADER
TOTAL 2004 2000 2004 2004

White (77%) 55% +1 44% 1%

African-American (11%) 10% +1 90% 0%

Latino (9%) 41% +6 56% 2%

Asian (2%) 38% -3 61% *

Other (2%) 38% n/a 58% 1%

VOTE BY RACE AND GENDER BUSH
KERRY NADER
TOTAL 2004 2000 2004 2004

White Men (36%) 58% n/a 41% 1%

White Women (41%) 52% n/a 47% 1%

Non-White Men (10%) 28% n/a 69% 1%

Non-White Women (13%) 22% n/a 77% 1%"

http://www.exitpollz.org/cnn2004epolls/Pres_epolls/US/P/00/epolls.0.html

Can there be any denial whatsoever, that the final weighting was not impossible? It sure was, if you account for the fact it decreased completely on one side only, by virtually the same number as it increased on the other side. Using 660 additions & re-weighting. Your argument isn't balancing with the numbers.

[Internut]

13. You keep switching subjects -

I only asked you one thing. You said that the census survey data somehow helped prove the non-validity of the final exit polls. I asked you to show it. Still waiting for it.

[LightningFlash]

14. I see you are reaching now.

Perhaps with better understanding of the catergory weightings, you too will understand just how the final poll results are cooked and therefore not possible.

http://www.exitpollz.org/CNN_national1253.html

VOTE BY GENDER

BUSH

KERRY

NADER

TOTAL
2004
2000
2004
2004



Male (46%)
55%
+2
44%
0%



Female (54%)
48%
+5
51%
0%

18. VOTE BY RELIGION

BUSH




KERRY

NADER

TOTAL
2004
2000
2004
2004



Protestant (54%)
59%
+3
40%
0%



Catholic (27%)
52%
+5
47%
0%



Jewish (3%)
24%
+5
75%
*



Other (7%)
24%
-4
74%
2%



None (10%)
31%
+1
68%
1%



57. PRESIDENTIAL VOTE IN 2000

BUSH




KERRY

NADER

TOTAL
2004
2000
2004
2004



Did Not Vote (17%)
45%
n/a
54%
1%



Gore (37%)
10%
n/a
90%
0%



Bush (43%)
91%
n/a
9%
0%



Other (4%)
19%
n/a
69%
5% "

I'm sure if around 3 million 2000 voters didn't die who had voted for Bush, the poll weightings might be able to be explained somehow. Even with that, I highly doubt it.

[TruthIsAll]

15. Internut, quit hijacking my thread. Its for MP.

Now, as far as the census is concerned, two points

1. There are 3.5 million votes missing.
2. It confirms that 53.5% of the voters were women.
3. So the 13047 poll was correct to within .5% maximum.
Since 2) is proof of accuracy, then we can assume 1) is also.

Now, Kerry had 54% of the 13047 female vote.
How did it go down to 51% for the 13660?

How come the final exit poll says that 43% of Bush 2000 voters came to the polls this time?
It's IMPOSSIBLE.

The 13047 timeline had it at 41% - and Kerry won by 51-48%.
The 41% was at the top end of the 1% MoE.

How did it jump to 43%, well outside the 1.0% MoE, which is irrelevant anyway, since 43% is IMPOSSIBLE?

Let's calculate the maximum it could be.
Bush had 50.45 mm votes in 2000.
But 1.77 mm died
That leaves him with 48.7mm,
And 48.7/122.2 = 39.8% (his maximum)

So the 13660 poll overstated his vote by 4 million. That assumes 100% of Bush 2000 voters still alive came to the polls in 2004.

Oh, BTW...
48.7/125.7 = 38.7% if you believe the census.

Do you care to still argue that the final exit poll was correct?

Had enough?

[Internut]

16. All I am saying (and you seem to be missing) is that

the Census survey numbers do not contradict the final exit poll numbers. If they do, please show exactly where they contradict. Note that the exit polls say nothing about the total absolute number of voters.

[TruthIsAll]

17. I just proved that the 13660 is bogus and the 13047 is correct.

Do you want to try to contradict my How Voted analysis?

The census confirms that the 13047 male/female weights were correct.
So was the 13660- but ONLY for the 54/46 weights, NOT for the Kerry 54% share. It had the Kerry female vote at 51%, which is bogus.
They could not change the weights so they changed the vote%.

But they changed the How voted in 2000 weights.
So the 13660 was incorrect, both in the How Voted and in the Kerry vote - both off by a minimum 3%.

So that's it.
You just lost the argument.
Not that you ever had one.
Get over it.

[Internut]

18. The census data says absolutely nothing

about who voted for whom, whether in 2000 or in 2004. How can it contradict any kind of voting preference statistics when it does not contain that data?

You keep claiming that census data somehow contradicts the final exit polls. Please explain how, pointing to the same category numbers in the two surveys that do not match.

[TruthIsAll]

19. I'm going to say it one more time. Listen close.

The Census numbers agree with the 13047 and 13660.
The 54%/46% is confirmed.

So both are correct - until we look more closely.

Look at the How Voted in 2000 weights.
It is an IMPOSSIBLE 43% in the 13660.

It is also impossible 41% in the 13047 - but within the MOE.
Kerry WON the 13047.

So the census confirms that the 13047 is correct in the gender split.
Therefore, we CAN BE CONFIDENT THAT THE CENSUS 125.7 mm is ALSO PROBABLY CORRECT to within the 0.30% MoE since the male/female split is correct to within 0.5% (exit poll values are rounded to the nearest percent).

So the census confirms
1) 3.5 million votes were lost.
2) The 13047 poll was correct in the gender split.
3) The 13660 poll was correct in the gender split.
4) But IMOSSIBLE How Voted weights PROVE THAT THE 13660 is bogus.
5) Therefore, since the 13047 is correct, we can assume that Kerry won 54% of the female vote.
6) Since the 13660 was just proven incorrect in the How Voted weightings, and the 13660 Gender weight was NOT changed, then the 13660 Kerry gender vote of 51% is INCORRECT AS WELL.

DID YOU GET ALL THAT YET?

[Internut]

20. So the census confirms that the 13047 is correct in the gender split.

I agree with that.

"Therefore, we CAN BE CONFIDENT THAT THE CENSUS 125.7 mm is ALSO PROBABLY CORRECT to within the 0.30% MoE since the male/female split is correct to within 0.5% (exit poll values are rounded to the nearest percent)."

One does not follow from the other. Just because people didn't fib about whether they were male or female (it is pretty hard to do that) does not mean that they did not fib about whether they voted or not. Census Bureau itself says that its results routinely inflate those numbers. Of course, you ignore the Census Bureau when it says that, because it does not fit your "proof".

[TruthIsAll]

21. And you have completely ignored my whole post.

From now on, I will remember to put you on mental "ignore" without actually doing it, just so that I can see what mischief you are up to.

You have avoided my challenge regarding How Voted in 2000.

All that work to no avail.
You are expert at diversion.
So long Internut.

[Internut]

23. I did not ignore your whole post -

I actually quoted from it and pointed out where it was wrong. Apparently you don't want to argue those points. I understand.

[autorank]

25. Self delete-wrong place

[TruthIsAll]

7. This model has NOTHING to do with Census data

Please don't confuse the two.

The data used in the model is Bush/Kerry vote split, aggregate precinct partisanship groupings, weighted average alpha (K/R =1.12) and mean response (53%) and partisan WPE (within precinct error).

It takes off from there...

[Name removed]

22. Deleted message

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

[Helga Scow Stern]

24. He posts about DU on his site....surely he must visit it.....

or might his head explode?

[Name removed]

27. Deleted message

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

[autorank]

26. Well, no MP yet (or could I be wrong?). FMP=Friends of Mystery Pollster

No MP but FMP's (who could be MP)...stranger that science!

[Name removed]

29. Deleted message

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

[Peace Patriot]

28. TIA, see Internut's posts at...

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=375727#375759

Enough said.

[autorank]

30. This isn't Febble's erudition; it's a time drain, pure & simple.

The rhetoric is:

A: It's a nice day.
Z: What do you mean by nice?
A: Well the sun is shining, the sky is blue.
Z: Why is that nice?
A: Well, everybody thinks so, don't they?
Z: How do you know everybody things so?
A: Look, I was just saying...
Z: How do you know any of what you're talking about is true.
A: Look, I was just saying...
Z: You're avoiding my points
A: I really don't want to carry on this discussion, it's useless.
Z: Oh, avoiding my points, etc. etc.

Creating a conundrum by simply questioning everything...

-----------
Do something useful:

NEW LEADERS FOR A NEW DEMOCRATIC PARTY!

Contact the DNC and Give 'em Hell About NOT Acting on Election Fraud

[Name removed]

31. Deleted message

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