I was worried too about the moralistic and/or contentious tones of some of
these emails. While the points brought out are certainly valid, I wonder if
we end up scaring off people by being so strident in our criticisms.
Perhaps a softer approach might be to emphasize that an on-line survey (or a
convenience sample) is useful only for pilot studies and that any such
survey ought to be followed up with a more rigorous sampling scheme.
This places sharp limits on the interpretation of data from such a survey,
but at the same time is a less discouraging message.
What do other people think? I worry about someone who comes in here after
spending a lot of time collecting information and the only response they get
is that their data is worthless. Such a response might scare them away from
consulting a statistician for the next three decades.
Can we get by with calling their work a pilot study, or are we sugar coating
the bad news that they need to hear?
Steve Simon, [EMAIL PROTECTED], Standard Disclaimer.
STATS - Steve's Attempt to Teach Statistics: http://www.cmh.edu/stats
<http://www.cmh.edu/stats>
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
