Jeff...you may want to look at:

Cohen, J.  (1983). The cost of dichotomization.  Applied psychological
measurement, 7, 249-253.

Also, and unfortunately I don't have the cite handy, about 2 years ago
McCallum et al did an article on dichotomization in Psychological Methods.

hope this helps...dale

Dale N. Glaser, Ph.D.
Executive Manager, Strategic Research
TrendSource
4891 Pacific Highway; Suite 200
San Diego, CA 92110
Phone: (619) 718-7467 (ext. 128)
Fax: (619) 718-7490
Email: [EMAIL PROTECTED]


-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Behalf Of Dang, Jeff
Sent: Tuesday, March 02, 2004 1:33 PM
To: '[EMAIL PROTECTED]'
Subject: [edstat] Continuous versus Dichotomous


Edstat,

I have personally found that a lot of health researchers like to aggregate
normally distributed, continuous outcomes into dichotmous outcomes.  In some
cases, this is done because the researcher is more familiar with dicohotmous
outcomes (disease/no disease) and seeks to interpret their results in terms
of odds ratios within a logistic regression.

In some cases, this can be problematic because you lose information.  For
instance those near the cut-off point are forced into one group or another.
Thus, you exaggerate the differences for some individuals.

If anybody on the listserv can refer me to articles related to this problem
I would be most appreciative.  Thank you.

Jeff Dang
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to