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 #63: Theorem, not theorum [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
Febble Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Nov-23-05 03:28 AM
Response to Reply #56
63. Theorem, not theorum
Edited on Wed Nov-23-05 03:45 AM by Febble
And it is completely irrelevant to the point pollsters and others have been making about why the polls might have been in error.

The Central Limit Theorem tells you nothing about non-sampling sources of error in polls.

Probability figures with gazillion zeros may impress headline writers but they are not going to impress a statistician unless TIA shows at least some rudimentary understanding of the nature of non-sampling error in surveys, and the assumptions underlying inferential statistics.

He might also like to comment on the complete lack of correlation between swing and shift in the exit polls and see if he reconcile that with his claim that the early exit poll projections were "right".

If they were "right", how come Bush didn't gain greater advantage in precincts where they were "right" than where they were "wrong"?

The most parsimonious answer is that the error in the polls had little or nothing to do with fraud, and a lot to do with non-sampling error, either selection bias or non-response bias.

Neither of which have anything to do with the Central Limit Theorem.

And yes, polls probably aren't a lot of use for distinguishing between 49% support and 51% support. However, they are pretty good at distinguishing between 80% support and 30% support so we can be pretty confident that Bush is, indeed, toast. Unless some other catastrophic event bumps him up to 80% again (the law of 4th degree polynomials notwithstanding).


(edited for typo, and for clarity)
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