I'd start with a simple test to look at the contingency (2-way crosstabs) tables of each classifying variable with the outcome variable and look at the off-diagonal cell counts. Then I would look at the simple contingency tables of each predictor variable with the other to see whether they are correlated.
My approach is to start simply and then proceed to various statistical techniques if the simple inspection indicates it.
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of Stats Q
Sent: Wednesday, February 07, 2007 4:18 AM
To:
[hidden email]
Subject: Classification
Dear list,
I have three dichotmous variables that can classify people into disorder or
no disorder, but I want to determine the consistency/incocnsistency in how
these three variables have classified people e.g., did variable 1 classify
cases 1 and 2 as disorder whilst variables 2 and 3 didn't. Is there
something I can do in SPSS to check this?
Thanks in advance.
_________________________________________________________________
MSN Hotmail is evolving - check out the new Windows Live Mail
http://ideas.live.com
Locked
