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

Reply #90: trash talk aside, you're wrong again [View All]

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
Home » Discuss » Topic Forums » Election Reform Donate to DU
OnTheOtherHand Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Aug-27-09 09:04 PM
Response to Reply #89
90. trash talk aside, you're wrong again
As I explained previously, analyzing pre-election polls from all states leads to the same conclusion. I'll briefly walk you through the reasoning, again. If the extent of fraud varies from state to state, then in general states with more fraud should have (1) larger deviations between the official returns and pre-election polls, and (2) larger red shifts (exit poll deviations). Therefore there should be a significant correlation between surprise and red shift. Of course, in any one state, things could play out differently due to sampling error or other circumstances, but across all the states, it's reasonable to expect a relationship. But it doesn't exist, even when I used your poll numbers. The plot is here; I assume you've seen it before.

2) Compare the Unadjusted exit poll methods:
WPE Margin
IMS 10.9
DSS 13.5
VNS 11.3

Ok, still big But the 12:40am Composite had Kerry ahead by 54.1-45.4%, an 8.7% margin.
Why didnt you pick that one?

In 2009, you still don't know the answer to that question? Really?

You seem to think that you can take the three different ways of reckoning Within Precinct Error and call them "Unadjusted exit poll methods." That is a very strange mistake for you of all people to make, because WPE depends on the official precinct-level returns. None of these numbers is an "exit poll method()" -- you've derived them from after-the-fact estimates of one source of error in the exit poll, and that on the assumption that the official results are correct. The Call 3 Best Geo estimate was the pollsters' best contemporaneous estimate of the vote shares and margin, based on exit poll data in comparison with historical results.

Your question ("Why didn't you pick that one?") could have been an excellent one, if it had been asked in good faith four years earlier. The reason not to use the composite estimate is in the definition of the estimate: "The Composite Estimate is a weighted average of the Prior Estimate and the Best Survey Estimate." What is the "Prior Estimate"? "Prior Estimates are based upon pre-election surveys conducted in each state." (You know this, right? I mean, are you kidding me?!)

So, apparently you're trying to argue that I'm exaggerating the gap between the exit poll estimate and the pre-election polls because if one uses a composite estimate that averages in the pre-election polls, the gap narrows! Doh.

So give Kerry another 1-2% and hes at 54-55%.

The pre-elects, properly adjusted, matched the exit polls.

No, because the Best Geo estimate put Kerry at 56.9%. This information has been freely available since January 2005.

No, too, because you claim to have "properly adjusted" "the pre-elects" -- but, setting aside the properness of your adjustment, you've cherry-picked a single pre-election poll, which just so happens to be the one that gives Kerry the biggest margin. (Gee, imagine that?) The CNN/USA Today/Gallup Poll where Bush led? Down the memory hole. And then you added a few more points. And you still couldn't match the exit poll projection. Impressive, but not in a good way.

(Oh, and, no yet again, because your contemporaneous and very generous "base case" projection for Kerry was 53.8%, so to goose it to "54-55%" now is yet another reach.)
Printer Friendly | Permalink | Reply | Top

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

Powered by DCForum+ Version 1.1 Copyright 1997-2002
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