QUANTS: To match WPE = - 6.77%, Optimizer needs wtd alpha >= 1.15

[TruthIsAll]

This will not make much sense to most of you.
It is for those familiar with the USCV analysis to ponder.

The model could not derive a feasible solution for
Mitofsky-supplied alpha = 1.12. It could only do so when alpha
was raised to 1.15. 

Th model derived alpha = 1.42 in High Bush precintcs. 
It was only 1.0 in High Kerry and evenly distbuted between
1.10 to 1.17 in the three middle partisanship groups.

What does this say about: 1) the model, 2) rBr?

EXIT POLL RESPONSE OPTIMIZATION MODEL							
6/10/05 1:34 PM							
							
OBJECTIVE:							
Determine values of constrained variables required to derive a
target Kerry/Bush percentage split using aggregate exit poll
response data.							
							
Precinct Variable Input Range (Min, Max) Constraints: 							
1-Response rates: equate to weighted average							
2-Kerry 2-party win percentages							
3-Alpha (K/B): equate to weighted average							
4-WPE: input  (optional: Min= Max= E-M)							
							
OPTIMIZER TARGET VOTE (2-party)							
Kerry	48.77%						
Bush	51.23%						
							
WEIGHTED AVERAGE RESPONSE							
Overall Rate	53.0%						
Alpha (K / B)	1.15						
							
RESPONSE INPUT CONSTRAINTS							
							
1250	Strong Bush		Strong Kerry				
Prcts	40	415	540	165	90		
							
KERRY WIN%							
Min	0%	25%	40%	60%	70%		
Max	25%	40%	60%	70%	100%		
							
EXIT POLL RESPONSE							
Min	5%	5%	5%	5%	5%		
Max	80%	80%	80%	80%	80%		
							
ALPHA (K/B)							
Min	0.10	0.10	0.10	0.10	0.10		
Max	10.00	10.00	10.00	10.00	10.00		
							
WPE							
E-M	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		
Min	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		
Max	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		
							
OPTIMIZER OUTPUT SUMMARY							
							
			EXIT POLL / VOTE DEVIATION 				
	Percentage			Amounts (mm)			
	Poll	Vote	Diff		Poll	Vote	Diff 
Kerry	52.16%	48.77%	-3.39%		63.11	59.01	-4.10
Bush	47.84%	51.23%	3.39%		57.89	61.99	4.10
							
Categ.	HighB	Bush	Even	Kerry	HighK		Total/Avg
Number	40	415	540	165	90		
Weight	3.2%	33.2%	43.2%	13.2%	7.2%		
							
EXIT POLL							
Kerry	28.8%	42.6%	54.7%	63.6%	70.6%		52.16%
Bush	71.2%	57.4%	45.3%	36.4%	29.4%		47.84%
							
RESPONSE							
Total	48.0%	49.9%	53.6%	57.4%	57.8%		53.00%
WtdDev	-5.0%	-3.1%	0.6%	4.4%	4.8%		0%

ALPHA							
K/B	1.42	1.15	1.17	1.10	1.00		1.15
Dev	23.4%	0.4%	1.6%	-4.6%	-13.4%		0.00%

2-PARTY VOTE 							
Kerry	23.8%	39.6%	50.4%	60.7%	70.7%		48.77%
Bush	76.2%	60.4%	49.6%	39.3%	29.3%		51.23%
							
VOTES (mm)							
Kerry 	0.92	15.89	26.35	9.69	6.16		59.01
Bush	2.95	24.28	25.93	6.28	2.55		61.99
TOTAL	3.87	40.17	52.27	15.97	8.71		121.00
Diff	-2.03	-8.38	0.42	3.40	3.61		-2.98

EXIT POLL (mm)							
Kerry 	1.12	17.12	28.57	10.16	6.15		63.11
Bush	2.76	23.05	23.70	5.81	2.56		57.89
TOTAL	3.87	40.17	52.27	15.97	8.71		121.00
Diff	-1.64	-5.93	4.86	4.35	3.58		5.22
							
WPE	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		-6.77%
E-M	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		-6.77%
Diff	0.0%	0.0%	0.0%	0.0%	0.0%		0.0%
							
							
							
							
							
							
OPTIMIZATION MODEL							
							
Categ.	HighB	Bush	Even	Kerry	HighK		Total/Avg
Prcts	40	415	540	165	90		1250
Kerry	23.8%	39.6%	50.4%	60.7%	70.7%		48.77%
							
Kerry	10	164	272	100	64		610
Pct	23.8%	39.6%	50.4%	60.7%	70.7%		48.77%
Bush	30	251	268	65	26		640
Pct	76.2%	60.4%	49.6%	39.3%	29.3%		51.23%
							
ALPHA							
K/B	1.42	1.15	1.17	1.10	1.00		1.15
AvgDev	23%	0%	2%	-5%	-13%		0%
							
RESPONDERS							
Total	19	207	289	95	52		663
Pct	48.0%	49.9%	53.6%	57.4%	57.8%		53.00%
							
Kerry	6	88	158	60	37		349
Pct	28.8%	42.6%	54.7%	63.6%	70.6%		52.16%
Bush	14	119	131	34	15		314
Pct	71.2%	57.4%	45.3%	36.4%	29.4%		47.84%
							
REFUSERS							
Total	21	208	251	70	38		587
Pct	52.0%	50.1%	46.4%	42.6%	42.2%		47.00%
							
Kerry	4	76	114	40	27		261
Pct	19.2%	36.5%	45.5%	56.7%	70.9%		44.98%
Bush	17	132	137	30	11		327
Pct	80.8%	63.5%	54.5%	43.3%	29.1%		55.02%
							
							
WPE							
Kv-Bv	-52.3%	-20.9%	0.8%	21.3%	41.4%		-2.46%
Kp-Bp	-42.3%	-14.8%	9.3%	27.2%	41.1%		4.31%
							
WPE	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		-6.77%
E-M WPE	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		-6.77%
							

[kster]

1. Kick.nt

[OnTheOtherHand]

2. I think E/M were sloppy-wrong here

I'm not sure that anyone noticed this issue until a few weeks ago (Bruce O'Dell noticed that the alphas in his regressions seemed too high), but it probably takes a bigger alpha to yield a 6.5-point mean WPE.

It's sort of a nasty calculation to estimate, because WPE is a mess. But I just roughed out that for a precinct that is really 51.5% Bush to have a -6.5 WPE, it takes about a 7-point difference in completion rates; if it were really 80% Bush or 20% Bush, maybe an 11-point difference.

The actual numbers are going to be all over the map, but if you stare hard at page 12 of the O'Dell paper, you'll see that it implies an average alpha of somewhere around 1.158 (because the average ln alpha will be around 0.1467). So that could be interpreted as a bit under an 8-point gap, say, about 56.9% to 49.1% (if we are trying to hit about 53% overall), which seems to fit with what I roughed out.

At this point, I don't think that this matters much because (1) I don't have any strong prior assumptions about how bad the response bias could be, and (2) the vote shift models I've seen create a slope that doesn't show up in the E/M scatterplot. Some folks find it crazy to think a poll could be that far off, but that scatterplot is a mess, so I figure the poll was a mess, too. But we need more out of E/M, and we need to keep thinking about what kinds of wholesale fraud would be consistent with the data we have.

[Melissa G]

3. Kick for TIA! OTOH, Agree wholeheartedly with your last line...

But we need more out of E/M, and we need to keep thinking about what kinds of wholesale fraud would be consistent with the data we have.

(so much for my lunch break..)

[LightningFlash]

4. Consistency...

In what has fully been consistent and not is electronic manipulation and its been proven.

[TruthIsAll]

5. OTOH, STOP IT, ALL READY. DO YOU ALSO BELIEVE 1+1=3?

The polling was NOT a mess, it was as correct as you can be.
The vote counting WAS a mess, it was FRAUD.

Enough of this very LAME conjecturing.

You are a smart guy.
You are a professor.
Trust science.
Trust your own common sensce.

I would say trust the Optimizer, but that is like saying trust me whan I tell you that 1+1 =2. NO TRUST IS REQUIRED IN THAT REGARD.

rBr is TOTALLY DEAD.
Stop promoting faith-based pipe dreams.

The burden of proof is on YOU to prove that it was NOT stolen.

[Melissa G]

6. NO TIA, rBr is a ZOMBIE!!!! It's dripping in dirty rags...it's coming to

Get us!!! Oh no, wait... that's just my kids watching the Saturday afternoon horror movie... rBr is DEAD. Definitely Dead...Back to housecleaning...

[OnTheOtherHand]

7. let me tell what people I know think

TIA,

First of all, in the world outside, few if any people care what you or I think, so there is no point in arguing about whether the burden of proof is on me, on you, on E/M, or whatever.

It seems to me that the optimizer can (basically) tell you what certain numbers must be given other numbers, but it can't tell you whether the "certain numbers" are possible or not.

Now, it would be a nice world if I could tell you, "OK, you got me, I'm convinced" and every single public opinion analyst would go along with my change of mind. Or, for that matter, if you could tell me, "Yeah, you're right, rBr is possible," and every single DU member would agr-- OK, that's ridiculous, but at least refrain from trashing you for it. Not the case.

A lot of folks who work in polling or with polls believe deeply that polls can, do, and will go very wrong. The history of polling is the history of people learning from the ways in which polls have gone very wrong. The idea of response bias may seem contrived to you, but people who work with polls believe in it.

I guess what I have to say to you is that in my opinion, there isn't enough evidence in and about the exit polls to convince a typical "expert" that massive, decisive fraud is proven or even especially likely. Working with your optimizer would not persuade them.

I can stop saying this, but my silence will not change the fact. Whatever else one says about the latest USCV working paper, it has not taken the AAPOR world by storm. Some folks think the world would be different if a few of us would just stop posting, but it wouldn't.

One thing I would like you to know is that arguing, over and over, that I think Bush probably won the popular vote makes me miserable. Do you know the website sorryeverybody.com? That's how I feel. So, while you and I may find bits of fun in this, basically it is an utterly lousy experience for everyone.

The fact that we are even having a serious discussion about who won the 2004 election says to me that the country has a huge problem. If I thought I could fix the problem by pretending to be convinced of things I'm not, I would seriously consider it. For that matter, I probably kid myself that it matters if I say what I think. But I don't know what else to do.

See ya,
OTOH

[LightningFlash]

8. Watch the actual math, the numbers, instead of the rhetoric.

And you have yet to prove anything else, with it already being demonstrated by the so called "Bev Harris" groups that Florida was stolen with democrats also involved..

More testimony piling up,
http://www.waynemadsenreport.com/scoops/Lemme.htm

[kster]

9. The story that

just keeps giving and giving.

[TruthIsAll]

11. Yes, we have a huge problem. It's the stolen elections.

Three rimes in a row.

And sophisticated people like yourself who, for whatever reason, stick their head in the sand and claim the polling was bad.

Come on.
Three times in a row.

[kster]

10. Kick. to get rid of the silliness..nt