On 28 Sep 2007, [EMAIL PROTECTED] wrote:
> but I tell you there's just so much you can do with a poorly
> designed survey.
Making a good survey is incredibly hard. And it doesn;t help that most
'surveys' start out with a result already in mind. But even surveys
which really want to find out something (Election opinion polls, for
example) don't do too well.
> Hey, negative results contribute to human knowledge too, right?
True, but no one wants to publish negative results.
A man who was living in a country ruled by a cruel dictator was
unfairly arrested, blindfolded, and transported to a
prison. Once there, he discovered that his cellmate was a
mathematician. After a few days, the man managed to escape from
the prison. Unfortunately, the prison was located in the middle
of a burning desert, hundreds of miles away from the nearest
inhabited regions. After stumbling around in the desert for
days, the man was recaptured and moved back to his jail
cell. When his cellmate saw him return, he told him, "I could
have told you that escaping by foot was impossible. I tried the
same thing two months ago." At this, the now sunburned, parched,
and hungry man became very upset and exclaimed, "Why didn't you
tell me that sooner!" The mathematician simply shrugged his
shoulders and asked, "Who publishes negative results?"
--
Alok
A Difficulty for Every Solution.
-- Motto of the Federal Civil Service
