----- Forwarded message from Herman Rubin -----
The model must come from the investigator, not from the data.
Data only helps to choose between models, estimate parameters,
and related questions.
----- End of forwarded message from Herman Rubin -----
Actually, I think the apparent conflict between data exploration
techniques and formal inference techniques comes from too static a
view of scientific research. Models rarely spring fully formed into
the mind of an investigator who then goes out and collects the first
data ever on the subject under study. More commonly, observational
data and subject matter folklore (presumably based on experience with
data) combine with theoretical considerations to suggest possible
models. As time goes on, models become more fully formed and data
gathering grows more focussed on specific questions. Eventually a
testable hypothesis may be formulated and data gathered carefully to
test it. Even then exploratory techniques are useful for checking
assumptions and finding leads to further research.
Unfortunately, this process takes years of working in the field, and
can not easily be compressed into the time frame of a typical doctoral
dissertation. If the candidate's committee insists that the
dissertation end in testing a formal hypothesis, the the last 10% or
so of the process, then the first 90% is likely to be pretty hare
brained. I can remember from graduate school days students in the
social sciences trying to find a hypothesis to test via a process
somewhat like throwing darts at a wall with no target on it.
The other side of the coin comes after graduation as people at the
earliest exploratory stages of research chomp at the bit to find some
formal test to calculate. If you are not sure what hypothese you want
to test there is a good chance testing a hypothesis would be
premature.
_
| | Robert W. Hayden
| | Work: Department of Mathematics
/ | Plymouth State College MSC#29
| | Plymouth, New Hampshire 03264 USA
| * | fax (603) 535-2943
/ | Home: 82 River Street (use this in the summer)
| ) Ashland, NH 03217
L_____/ (603) 968-9914 (use this year-round)
Map of New [EMAIL PROTECTED] (works year-round)
Hampshire http://mathpc04.plymouth.edu (works year-round)
The State of New Hampshire takes no responsibility for what this map
looks like if you are not using a fixed-width font such as Courier.
"Opportunity is missed by most people because it is dressed in
overalls and looks like work." --Thomas Edison
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
