First off, can we get an operational definition of 'self esteem'? I
think in your case it might come down to, attitudes (or something) as
expressed on a test. Or a test score. I agree that there is something
there in the case of specific students and moments, but I haven't' seen
anyone express it in a way I can put numbers, averages, spreads, and
regressions on.
When you say "The test does not have great reliability." do I take it
you mean the test-retest correlation is weak? LIke, one day to the
next? And would this difference have much to do with repeating the
test? The Heisenberg Uncertainty Principle works in social science, as
well as Physics.
Assuming that I am on reasonably dry ground at this point, I'd say, the
analysis should be for the _change_ in rating for each person, over the
month. If this value was usually over 0 (indicating a general increase
in the measure), then we could search for possible associations/causes.
I would look for age, family income, gender, years since puberty (0 if
not yet there), and yes, the initial measure indicating self esteem.
You're the sociologist, you pick possible factors! If you don't have
that data, well, speculation is the only answer :). Just label it as such.
It would take a lot of data to overcome the perceived weakness in the
measuring instrument, to detect things like asymptotic approaches to a
maximum, or mildly non-normal data behavior. I don't think these are
likely to be fruitful approaches to understanding. Don't forget: just
because something is easy to calculate a correlation against the
response, doesn't mean it is a significant factor causing the response.
But you knew that.
Jay
Dale Berger wrote:
> Dear Colleagues,
>
> A student is evaluating a summer program for junior high students. One of
> the goals was to raise 'self esteem.' Measures were taken before the
> program, at the end, and a month later. He expected that the program would
> be most effective for those who entered with especially low self esteem. He
> divided the students into quartiles based on the pretest and compared these
> subgroups on change. He found that his hypothesis was supported - there was
> greatest positive change for those who entered in the lowest quartile.
>
> However, further examination showed a clear 'regression toward the mean'
> effect, including a small negative change for the group that entered in the
> highest quartile. (The test does not have great reliability.)
>
> Question: How should he analyze these data?
>
> I would appreciate some discussion of this situation. Thank you.
>
> Dale Berger
>
>
>
> =================================================================
> Instructions for joining and leaving this list and remarks about
> the problem of INAPPROPRIATE MESSAGES are available at
> http://jse.stat.ncsu.edu/
> =================================================================
--
Jay Warner
Principal Scientist
Warner Consulting, Inc.
4444 North Green Bay Road
Racine, WI 53404-1216
USA
Ph:
(262) 634-9100
FAX:
(262) 681-1133
email:
[EMAIL PROTECTED]
web:
http://www.a2q.com
The A2Q Method (tm) -- What do you want to improve today?
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
