On 13 Jul 2000 08:56:24 -0700, [EMAIL PROTECTED] (Bob Hayden)
wrote:
> While I agree that online surveys are of dubious value, some of the
> opposition sounds too moralistic or contentious for my taste. ...
< snip, rest >
I was thinking "too contentious", too, until I noticed that the
Subject really does say "survey" - which is the category that is so
tough, until there is another 10 or 30 years research in how to
un-skew responses. I suspect, if you know enough about a particular
question to do un-skew it, you probably don't need to ask it.
The one-shot questionnaire is not nearly as promising as recruiting
subjects for a panel. Then the surveyor can look for changes.
Now, you could "collect" as if for a cross-sectional survey, in order
to study, "Behavior of cult members in response to ...."
For a silly extreme, check out a (1998) best-books survey,
http://books.rpmdp.com/besboo/ever/molire.htm --
According to this survey: Seven of the top ten novels (in English)
were written by Ayn Rand and by L. Ron Hubbard.
(They also topped the surveyer's related list of Best Non-fiction.)
The rest of the list, I judge, is heavily influenced by the prejudices
of the handful of people who stuffed the ballot box for 1-10, since I
never saw such a list with so much Robert Heinlein (6).
You'd think folks might learn, but in 1999, Time-Life-Warner-ABC-
(whatever) held a big voluntary survey, and learned that
international voters favored Kemal Ataturk as Man of the Century in
about 8 different categories, from Best Statesman (because he founded
modern Turkey) to -- oh, maybe -- best athlete, and best singer (he
founded modern Turkey).
I think Time decided to report separately by continent, for one
solution. So, you too might leave yourself free to decide how to
count and report the votes, AFTER you see the problems arising.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
