On Wed, 24 Mar 2004 05:26:36 GMT, doubter
wrote:

On Tue, 23 Mar 2004 22:17:43 -0600, frank wrote:

On Tue, 23 Mar 2004 21:00:54 GMT, doubter
wrote:
What is important is that the sample be representative of the makeup of
the universe being sampled. If someone declines to participate, they
would be replaced by a member of the subset that the original person
represented.


Sounds great but no survey can really do this unless they survey all.
How do you decide that it represents the universe being sampled unless
you survey the universe first? Inotherwords, what would be your basis
to determine the survey is accurate?

The universe is defined by the customer who hires the pollster. In the


Inotherwords, by "assumption". Therefore the results are only as good
as the assumption !!!