You are viewing an obsolete version of the DU website which is no longer supported by the Administrators. Visit The New DU.
Democratic Underground Latest Greatest Lobby Journals Search Options Help Login
Google

Reply #84: Why would undecided voters not follow the "rules" this time? [View All]

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is archived.
Home » Discuss » Topic Forums » Election Reform Donate to DU
autorank Donating Member (1000+ posts) Send PM | Profile | Ignore Sun Dec-04-05 11:14 PM
Response to Reply #79
84. Why would undecided voters not follow the "rules" this time?
TruthIsAll:

FEBBLE:
I did, in fact, look carefully at your model - what I hadn't
looked in detail was your 1st November post. I knew you had a
high probability for Kerry winning the EV. I hadn't
appreciated that you had extended it to the popular vote. 

TIA
Pleasant surprise, yes?
Febble, my quantitative engine just won't quit, will it?
Leaves no stones crying out.

FEBBLE
Your model does assume that the polls were random samples
(Wang's did not, although my understanding is that he assumed
that overall there was no net bias). This is the problem I
have with it. I did understand your assumption about
undecideds - that was not the problem. Although I think that
you should also have put a probability value on your
assumptions being correct, as Wang did. But your model allows
us to put in our own assumptions, which is fair enough. But it
still assumes that the only error in the polls was sampling
error. It does not allow the possiblity for bias in the polls.
The point of my "mantra" is simply that your
probability estimates depend on the assumption of random
sampling, and we simply cannot make that assumption when it
comes to polls.

TIA
Just when you think I'm all done, you find another 
analysis of mine that you must try to rebut.
Notice I said "try".

You imply that the polls were not random samples.
Yet the national and state professional pollsters claimed they
were.
Who are you to question their methodologies?
They all quoted MoEs calculated using the standard formula: 
MoE = .98/sqrt(n).

Only Mitofsky's Exit Poll MoEs were adjusted - "cluster
effect".
But even he claims a 1.0% MoE for the National Exit Poll.
Of 13047 "randomly sampled" respondents.

Ok, that's your job.
To question time-tested sampling techniques.
To naysay the results.

You dare not argue the math.
But ALWAYS find fault with the "assumptions".

Now it's the breakout of the undecided vote (UV).
I understand.
We all must do what we 'gotta do.

In the state and national projection model, I used sensitivity
analysis (SA) to test a variety of UV allocation assumptions.
I always do this. It's a powerful tool. SA implicitly
recognizes uncertainty in key model drivers. In this case, the
main "driver" is the UV; that's why I ran 5000
simulations for each of five Kerry UV allocation scenarios.
Each scenario assumed that Kerry would win the majority of
UVs: 
60%, 67%, 75%, 80%, 87%...

In fact, all National Exit poll timelines (as well as the Ohio
Exit poll) had Kerry winning UVs in the week prior to the
election by over 60%. 

FEBBLE:
That is one respect in which we differ. The other is that I do
not accept that what has happened before will necessarily
happen again - or that trends will continue as they are going.
It's why I took issue with your 4th degree polynomial,
although I completely agree that the approval numbers look
terrible for Bush. But my experience with trends is that they
often change direction. 

TIA:
Who is talking about "trends" here?
Are you building a strawman for the holidays?
That UVs break for the challenger is a historic fact.
Don't let OTOH try to convince you otherwise.
He's just blowin' smoke on this one.

As for your comment that "my experience with trends is
that they often change direction". Just what are you
trying to say here? What is the point? Let's dispense with any
new talking-point generalities. I don't know which trends you
are referring to, but the UV breakout is not a trend. Of
course trends eventually change direction; but it's a strawman
argument, because I never used or considered trends in my UV
assumptions. 

As Friday used to say: Just the facts, maam'. 

FEBBLE:
So I did not put a lot of certainty in the incumbent rule, nor
in the undecideds-break-for-the-challenger rule. Although I
desperately hoped they would hold. 

TIA
Careful, now. That's a contradiction. Why did you hope the
"rules" would hold unless you knew that they held in
the past? That's why they're called "rules". 

What was it in this election which caused you to lack
confidence in the "rules"? Did human nature really
change in this election? Or was it that  you knew that Bush
was running for re-selection again? And that the
"rules" don't apply to him. Is that why? Anyway,
that begs the issue. Do you have evidence as to why the
"rules"  did not hold in this election? If so,
enlighten us. 

No, Febble. The incumbency and approval rules held. Everyone
agrees that Kerry won the late undecideds and that Bush job
approval was 48.5% on election day. Everyone, that is, except
you.

FEBBLE
As for the Gore-Bush thing - as I said in the post, and I've
said many times before, given what we KNOW about the way
people report past vote, I do not see this as a clincher. Yes,
those proportions are impossible IF people reported their 2000
vote correctly. But we know they do not, and that they often
misreport having voted for the incumbent. 

TIA
Which voters misrepresented: Gore or Bush?
Don't answer. I already know what you are going to say.

FEBBLE
This has been shown in longitudinal studies where the
pollsters KNOW how their respondents said they in a given
year, yet the SAME voters report having voted for someone
different when asked several years later, in both the UK and
the US. So we know we cannot rely on "reported vote"
data.

TIA
Latitude. Longitude.
Febble, you're way off the map.

So I guess we can't rely on any surveys of past behavior.
So much for the scientific method.
So much for experimental design.
So much for pollsters.
So much for Mitofsky.

FEBBLE
Anyway, happy holidays to you and to all! See you after I've
submitted my dissertation.

TIA
Happy holidays.
Good luck in your research.
But be careful not to rely on any survey data in your
dissertation.
Can't trust 'em.


Printer Friendly | Permalink |  | Top
 

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

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


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

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

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

© 2001 - 2011 Democratic Underground, LLC