Statistical analysis suggests possible fraud in polling data

[SomeTh...@aol.com]

http://blog.revolution-computing.com/2009/10/statistical-analysis-suggests-possible-fraud-in-polling-data.html

"Statistical analysis suggests possible fraud in polling data"

Excerpts

"Does this constitute evidence that some of the Strategic Vision polls were "tinkered with" by hand? Nate Silver thinks it might:

Certain statistical properties of the results reported by Strategic Vision, LLCsuggest, perhaps strongly, the possibility of fraud, although they certainly do not prove it and further investigation will be required.

"

This deals with one firm, Strategic Vision.

"the firm is unlike most polling organizations in that it refuses toreveal even the most basic details of its methodology (sample size ordates of surveys, for example). For this it has recently been censured by the American Association for Public Opinion Research. "

I'll let the experts comment further...

Jim Soper

www.CountedAsCast.com

[Paul Lehto]

The prestige and perceived success of a pollster consists in large
part in predicting the outcome of "real" elections. If the elections
are being rigged, the most successful pollsters will learn to weight
their polls accordingly (to the extent this is possible). This is why
I went with Rasmussen/FOX polls prior to the 11-08 election, and
indeed they turned out to be closest to being right "on the money."

paul



On 10/31/09, Lisa Pease <lpe...@gte.net> wrote:
> There was a pollster back in, I think it ws 1968 but it might have been
> 1972, who was caught red-handed essentially making up his polling numbers. I
> don't recall if he received any punishment - the exposure was enough to put
> him out of business, I think. But who's to say that isn't happening all over
> the place? They find out what numbers they need, and then report numbers
> that match their needs.

>
>
>
> -----Original Message-----
> From: CalifElecti...@yahoogroups.com
> [mailto:CalifElecti...@yahoogroups.com] On Behalf Of
> someth...@aol.com
> Sent: Wednesday, October 28, 2009 2:37 PM
> To: CalifElecti...@yahoogroups.com; voting...@yahoogroups.com;
> Election...@googlegroups.com
> Subject: [CalifElectionProtection] Statistical analysis suggests possible
> fraud in polling data
>
>
>
>
>
>

> http://blog.
> <http://blog.revolution-computing.com/2009/10/statistical-analysis-suggests-
> possible-fraud-in-polling-data.html>
> revolution-computing.com/2009/10/statistical-analysis-suggests-possible-frau
> d-in-polling-data.html

>
> "Statistical analysis suggests possible fraud in polling data"
>
> Excerpts
>
> "Does this constitute evidence that some of the Strategic Vision polls were
> "tinkered with" by hand? Nate Silver thinks it might:
>
> Certain statistical properties of the results reported by Strategic Vision,
> LLCsuggest, perhaps strongly, the possibility of fraud, although they
> certainly do not prove it and further investigation will be required.
>
>
> "
>
> This deals with one firm, Strategic Vision.
>
> "the firm is unlike most polling organizations in that it refuses toreveal
> even the most basic details of its methodology (sample size ordates of
> surveys, for example). For this it has recently been censured

> <http://www.aapor.org/AAPOR_Raises_Objections_to_Actions_by_Strategic_Vision
> _LLC.htm> by the American Association for Public Opinion Research. "

>
> I'll let the experts comment further...
>
> Jim Soper
>
> www.CountedAsCast.com
>
>
>
>
>
>
>
>
>

--
Paul R Lehto, J.D.
P.O. Box #1
Ishpeming, MI 49849
lehto...@gmail.com
906-204-4026

[honest electionscow]

This is why we are now doing our own exit polling, AND all methods are transparent, raw polling data is made available, methods of " adjusting" are explained (to correct for party, gender, and age bias, for instance) so that someone with high school math can repeat and understand the calculations, and the citizens conducting the polling can be contacted and interviewed. Fool proof? No. Capable of catching fraud? A high probability, given a large enough sample. EDA and Election Integrity conducted a large national exit poll in the 08 election. Our success is not determined by matching the "real" election results, but in doing the most scientifically honest poll possible. Suspicions of fraud in the official election results are raised when all but one race, or a few races, in a given election , closely match polling results, and all other reasons for this phenomenon are exhausted.  Another report is forthcoming.  Some of us would like to do exit polling every national election. Marj Creech
 

--

To post, send email to Election...@googlegroups.com. Please review the  "Posting Guidelines" page.

Please forward EI messages widely and invite members to join the group at http://groups.google.com/group/ElectionIntegrity/members_invite.

If you're not a member and would like to join, go to http://groups.google.com/group/ElectionIntegrity and click on the "join" link at right. For delivery and suspension options, use the "Edit my membership" link.

--
www.recallvotingmachines.com