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 #107: c-macdonald, if you are truly the statistician you say you are... [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
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Sun Nov-28-04 07:24 PM
Response to Reply #59
107. c-macdonald, if you are truly the statistician you say you are...
Edited on Sun Nov-28-04 07:43 PM by TruthIsAll
you would know the answer to #2 by looking at the data.

Either you are a statistician who did NOT look at the data, or you are exaggerating your credentials.

BTW, I'm NOT a statistician. But I have three degrees in mathematics, for whatever that's worth.

To wit:
The Margin of Error for a polling sample size n in simply 1/sqrt(n).

For a sample size of 900, the MOE = 1/30 = 3.33%
For a sample size of 1600, the MOE = 1/40 = 2.50%
For a sample size of 2500, the MOE = 1/50 = 2.00%

The standard deviation statistic used to calculate the 95% confidence interval = MOE/1.96

So that is how I derived the MOE and Std dev, based on the exit poll sample size for each state.

You refer to the "unrandomness" of exit poll data. That begs the issue. The question should be "HOW ACCURATE HAVE EXIT POLLS BEEN HISTORICALLY"? Who cares if they are random or not if they are virtually always accurate to with a fraction of ONE percent? This is in fact the case. Let me add (ad nauseam) that exit polls have been extremely accurate wherever they have been used in the world. In fact, the last three German elections have been accurate to within ONE-TENTH OF ONE PERCENT.

As far as OUR exit polls are concerned, they have ALSO been extremely accurate historically. After all, that's how Mitofsky earns a living, isn't it? And he's the INVENTOR of exit polling, isn't he? And doing them for over 25 years, right?

Why was there no question regarding the accuracy of exit polls prior to 2000? Hmmmm....

Remember FL 2000? Initially, they called it for Gore. But then a computer "glich" in Volusia county spread 16,022 of Gore's votes to third party candidates (I believe the Socialist got 10,000) and Voila! - they called FL for Bush.

It turned out, though, that the exit polls were right since 175,000 votes were spoiled (double and triple-punched cards), the great majority of them in minority precincts who voted overwhelmingly for Gore. The minorities left the polling booths thinking their votes were counted. Of course, little did they know...

Yes, the Exit Polls were right after all. Gore won FL by at least 50,000 votes.

Printer Friendly | Permalink |  | 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