If you use CATPCA with multiple nominal scaling level for all variables,
you have MCA, but MCA gives more MCA-specific output (discrimination
measures, contributions). Multiple nominal scaling level fits a variable
according to the centroid model: A category is in the centroid of the
components scores of the cases that have scored the category. Then a
variable has multiple sets of quantifications, one set for each
dimension. The variable is represented by a set of points. All the other
scaling levels fit a variable according to the vector model (PCA):
category quantifications are restricted to be on a line, resulting in
one set of quantifications for all components and multiple component
loadings. Thus the variable can be represented by a vector (the loading
vector).
With MCA you have the centroid model: all variables are analyzed at the
multiple nominal scaling level. With CATPCA, you can have the centroid
model, or the vector model, or a mix of the centroid and vector model.
Anita van der Kooij
Data Theory Group
Leiden University
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of
Zetu, Dan
Sent: 15 April 2008 15:03
To:
[hidden email]
Subject: CATPCA vs MCA
Dear list,
What is the difference between CATPCA and MCA? When one should use one
vs the other and what is the difference in output?
Thanks a lot,
Dan
-------------------------------
Dan Zetu
Analytical Consultant
R. L. Polk & Co.
248-728-7278
[hidden email]
=======
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
**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
**********************************************************************
=====================
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
