At 12:20 AM 2/1/03 -0500, Karl L. Wuensch wrote:
I am familiar with the Psych. Methods article Dave mentioned (in fact, I referred my colleague to it, it is a gem), and Dennis' points pretty well matched mine with respect to why not to dichotomize. My colleague's query was, however, "if you are going to dichotomize a continuous subject variable and compare the resulting two groups on a second continuous variable, even though that is not generally a good idea, is it more useful to use a median split (upper half vs lower half) or to compare the tails (such as upper third versus lower third)?"i think my reply would be in this case ... that ... since it is NOT a good idea to consciously throw away data by dichotomizing a variable on which you have continuous scores for Ss ... then, NEITHER of these approaches would be recommended and, you won't get me to argue for one or the other
if you argue for one or the other ... that seems like you are encouraging deliberate bad practice
IMHO
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
