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 #57: It has everything to do with [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 locked.
Home » Discuss » Topic Forums » Election Reform Donate to DU
Febble Donating Member (1000+ posts) Send PM | Profile | Ignore Sat Jun-16-07 05:24 AM
Response to Reply #55
57. It has everything to do with
the OP.

There are two broad theories that could account for the anomalies that the OP draws attention to: that there was something wrong with the poll and that there was something wrong with the count. I have explained why poll artefacts could account for it. However, if the reason was something wrong with the count, then the implication is that there was vote corruption in certain places - big cities, for example. That Bush's vote was "padded" in some places but not others.

But if this was the case, variance in padding rate should be reflected both in variance in poll-count discrepancies and variance in Bush's performance relative to 2000. And, unfortunately for that theory, these two measures are not correlated with each other. This means that theory two (above) is a lot less likely than theory one.

It looks to me as though National precinct sample of big cities happened to have non-representative turnout rate. This is not surprising in such a small sample. And the demographic composition, will, as I said, be warped by the extent of the reweighting process itself.


To address your question:

You are correct that precinct data isn't an EP question. But that doesn't mean it's immune from sampling error - quite the reverse. It is extremely vulnerable to sampling error because the precinct sample sizes are so small. Remember that precincts are sampled as well, and the sample sizes are small.

The EP weightings would not be based on turnout figures. In fact the national sample would be a very poor source of turnout change estimates by Size of Place, because of the small sample of precincts in each category.

Indeed I'm curious about the question, and actually spent a bit of time corresponding with anaxarchos over it, and poring over the weightings in the National dataset (which were available for free download for over a year). He remained convinced that the reweightings could only be explained by fraud. I was unconvinced by his argument then, and remain unconvinced now, for the reasons I have given.

As for your comment about rehashing old arguments about the vagaries of exit polls: it's an old argument because it's an important one. The exit polls were substantially discrepant from the count. This means either that the exit polls were badly skewed, or the count was. If the exit polls were badly skewed, inferences made from them about the count will also be misleading. Sure, I keep making this point, but it's because it's an important one. You cannot make a silk purse out of a sows ear. If the exit poll data had to be substantially reweighted because of bias in the poll then the demographic cross-tabs will be unreliable. There is strong evidence for bias in the poll. Additionally, you cannot reliably infer turnout change figures from a small number of precincts, because of sampling error in the precinct sample.
Printer Friendly | Permalink | Reply | 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