Melady Preece writes:
>Question 2: Can an effect size of .29 (or .33) be considered
>clinically significant?
With all due respect, clinical significance can only be assessed by
clinicians (i.e., subject matter experts). The doctors and nurses I work
with want me to tell them what a clinically significant difference is, but
as I am still trying to figure the difference between good cholesterol and
bad cholesterol, I usually have to plead ignorance.
As a general rule, though, you need to assess clinical significance in the
original units of measurement. A standardized and unitless measure like
effect size has no direct clinical interpretation. Suppose you were shopping
at a store and the sign reads "half a standard deviation discount". Is that
enough of a discount to make the purchase worthwhile? You would have no idea
unless you could convert back to dollars, pounds, euros, yen, etc.
Steve Simon, [EMAIL PROTECTED], Standard Disclaimer.
STATS: STeve's Attempt to Teach Statistics. http://www.cmh.edu/stats
Watch for a change in servers. On or around June 2001, this page will
move to http://www.childrens-mercy.org/stats
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
