A couple of things to keep in mind about exit polls, or at least the 
ones that were leaked. The first is that they are fairly shallow 
samples. With a national sample size of 12,000 you're talking about an 
average of 240 individuals per state which is about 1/4 the normal 
sample size used for political polls. That blows up the margin of error 
by a factor of 2 to plus/minus 7% - 10% instead of the usual 3 1/2% - 
5%.  And given that the early results were based on a fraction of the 
planned 12,000 interviews, the margin of error may widen by another 
factor of 1.5x to 2x.

Second, early results are going to be heavily biased towards the 
individuals who vote early. That may or may not be a problem depending 
on whether those people are representative of the overall population. In 
this case they were heavily weighted towards Democratic women, who as a 
group, tended to vote more heavily towards Kerry.

So if you combine the two, you have a result that is biased towards 
Kerry and has a huge margin of error, i.e. you got nothing. I haven't 
looked at it personally, but its my understanding that the final exit 
poll results were reasonably consistent with the actual results. Which 
ironically is what they were designed to do rather than as some kind of 
real time forecasting tool :-)

--rick



Ryan Ware wrote:

>On Wed, 10 Nov 2004 15:20:43 -0600 (CST), Fred H Olson
><fholson at cohousing.org> wrote:
>  
>
>>The controversy continues about the recent election
>>(See, for example, "Evidence Mounts That The Vote May Have Been Hacked"
>>                 http://www.commondreams.org/views04/1106-30.htm )
>>and there is an open source angle / possible remedy:
>>
>>The Open Voting Consortium (OVC) is a non-profit organization dedicated
>>to the development, maintenance, and delivery of open voting systems for
>>use in public elections.  http://openvotingconsortium.org/
>>
>>PS sorry about posting twice about Fedora Core 3 the other day, I
>>erroneously concluded that old address for the list that I sent to first
>>was not working.
>>
>>Fred
>>
>>--
>>Fred H. Olson  Minneapolis,MN 55411  USA        (near north Mpls)
>>Communications for Justice - My new listserv org.       UU, Linux
>>My Link Page: http://fholson.cohousing.org       Ham radio:WB0YQM
>>fholson at cohousing.org   612-588-9532   (7am-10pm Central time)
>>    
>>
>
>>From the article
>""This was no mere mistake. Exit polls cannot be as wrong across the
>board as they were on election night.  "
>
>Why so much stock in exit polls?  People can lie.  Pollsters can lie. 
>Personally I believe no exit polling should be allowed at all.  Or if
>it is, it may not be broadcast until after the polls have closed
>across the entire country.
>
>"I suspect foul play."  So do I, but from different people than the
>articles author.
>
>_______________________________________________
>TCLUG Mailing List - Minneapolis/St. Paul, Minnesota
>Help beta test TCLUG's potential new home: http://plone.mn-linux.org
>Got pictures for TCLUG? Beta test http://plone.mn-linux.org/gallery
>tclug-list at mn-linux.org
>https://mailman.real-time.com/mailman/listinfo/tclug-list
>  
>


_______________________________________________
TCLUG Mailing List - Minneapolis/St. Paul, Minnesota
Help beta test TCLUG's potential new home: http://plone.mn-linux.org
Got pictures for TCLUG? Beta test http://plone.mn-linux.org/gallery
tclug-list at mn-linux.org
https://mailman.real-time.com/mailman/listinfo/tclug-list