Fwd: Re: [CQ-Contest] Polling

[Jim Rhodes]

I wonder if the survey groups that are polled tend to be from the higher 
density population areas. This alone would bias the survey against 
contesters. I would be very interested in how they choose the survey 
population. Living most of my life in the midwest I have gotten used to the 
fact that our opinions don't count as much as those from either coast
.
At 10:33 AM 2/13/02, you wrote:


>Some excerpts from messages N4ZR and I exchanged last month...
>
> >At 10:00 AM 1/10/2002 -0500, N4ZR wrote:
> >
> >>It's worth remembering that a poll of a little over 1000 people is
> >>considered statistically representative of the entire adult population of
> >>the United States, within +/- 3 percentage points.  You can imagine how
> >>small a sample would give comparable results for the ham population of
> >>700,000 or so.  That makes it pretty likely that very few contesters are in
> >>the sample for any given QST reader poll.
> >
> >
> >The results of the last 3 ARRL Readex surveys have had error rates of 
> 3.6, 3.6 and 3.5 points.   And yes, out of 160K members, 6K contesters 
> won't be noticed much (4%).   Of those ARRL members on this reflector 
> (about 700?), the percentage is more like 0.4%.  So of roughly 800 
> surveys returned, only three might be from people on this reflector.
> >                                -- Tom
>
>=====
>e-mail: k1ki at arrl.org   ARRL New England Division 
>Director  http://www.arrl.org/
>Tom Frenaye, K1KI, P O Box J, West Suffield CT 06093 Phone: 860-668-5444
>
>
>--
>CQ-Contest on WWW: http://lists.contesting.com/_cq-contest/
>Administrative requests: cq-contest-REQUEST at contesting.com

Jim Rhodes K0XU
jim at rhodesend.net


--
CQ-Contest on WWW: http://lists.contesting.com/_cq-contest/
Administrative requests: cq-contest-REQUEST at contesting.com