Date: Mon, 19 Apr 2004 18:07:10 +0200
Reply-To: MOSER Winfried <firstname.lastname@example.org>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: MOSER Winfried <email@example.com>
Content-Type: text/plain; charset="US-ASCII"
I need to "scan" a database with many, mostly nominal Variables, for
Is there a possibility to produce a matrix, which shows me the
phi-coefficients (or other measures for nominal variables) of all
possible combinations of variables, (something like the output-matrix of
the procedure for bivariate correlations).
thanks in advance