To use Gender as a predictor in classification trees, use
format 1. This is the same format you would use if Gender
were a row or column variable in a multi-way table.
Anthony Babinec
[hidden email]
"Be the change you want to see in the world."
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of
Mark Webb
Sent: Sunday, October 28, 2007 3:09 AM
To:
[hidden email]
Subject: Representing Nominal data in CHAID
I have nominal data in various formats.
Could list members please comment on the suitability of the various formats
if I want to use the data as independent variables in Chaid/Classification
Trees.
Using Gender as an example [1=M 2=F]
Format 1
One variable/column with either a 1 or a 2.
Format 2
Two columns - One for males with a 1 as the indicator
One for females with a 1 as the indicator
Format 3
Two columns - One for males with a 1 as the indicator
One for females with a 2 as the indicator
Am I correct in saying that in the case of format 1 data the results would
indicate the extent to which Gender is a descriptor & in the cases of format
2 & 3 whether male vs female is a descriptor.
Are format 2 & 3 equivalent ?
Any comments on preferred format ?
Regards
Mark
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
