Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Politics for Pros- moderated -- Ignore unavailable to you. Want to Upgrade?

To: LindyBill who wrote (82880)	11/2/2004 6:23:17 PM
From: LindyBill	Read Replies (2) \| Respond to of 793835

HORSERACE - More Thoughts to Consider 1. If these exit numbers NRO is reporting, you would have to change the difference by Bush by 1.47% just in Bush's favor provided that they are working off a 59-41 M/W split, provided that men and women break the way they did in 2000 and provided that the turnout is 48% M/ 52% W. 2. What other kinds of sampling skews do these exit polls contain? Are they a realistic sampling of Republicans/Democrats, whites/blacks/Hispanics, etc? We do not know either way. 3. Consider the source of these exit polls. This is a completely untested system. They junked VNS in 2002, and decided to go with this consortium deal. But who knows whether this will be an improvement? As a matter of fact, exit polling has not been decent since 1998. 4. For those of you prone to panic, think about why I decided to excise ARG from my polling averages. I did so because I have no idea who they are, whom they are calling, how they are deriving their numbers, etc. The same holds true with these numbers that many of you are undoubtedly seeing on the Internet. Who is telling these bloggers these numbers? What sort of sampling derived these numbers...are they "scrubbed" or not? Where are those sources getting them from? The parties do not have direct access to this data...it is ultimately coming from the press to the parties, filtered down through party circles, and then to the bloggers. (I am surprised we have not seen an exit poll that shows Kerry: 49%, Bush: Purple and Yellow Elephant) Basically, all we have is a bunch of numbers without supporting internals reprinted roughly fifth hand...numbers that will never formally be published by the exit polling consortium (read: no accountability for false leaks). In other words, to be worried about these numbers is, in turn, to be totally unworried about good method when it comes to data analysis. 5. Again, have a little faith in BC04. If Republicans have not yet shown up at the polls, the GOP is on it. They are not going to let their voters sit on their hands. 6. The mid-day numbers seem to be trending back to Bush...are Bush voters really out voting from 2 PM to 5 PM? And yet the numbers are trending back.