Right, Nick! I only get the digest, and on top of that, I made a complete mess of my mailer today... Sorry for the delayed reply.
I do not know SPAD, so can't really give you an answer. Furthermore, I'm
really no expert in correspondence analysis: -mca- and -coranal- are
implemented from the step-by-step instructions given in Blasius &
Greenacre (1994)(*). You should perhaps compare this paper with the SPAD
manuals.
The first check you should make, though, is to compare the results you
obtain with and without the -notrans- option of -mca-. -mca- is simple
correspondence analysis applied to the so-called 'Burt matrix' (collection of all 2x2 combinations of variables). This results in high 'intertia' due to the presence of a diagonal of ''ones''. By default, -mca- applies a
correction for this. The -notrans- option disables this correction. Check if it is the correction that makes the difference with SPAD.
I can only add that I am quite confident that the simple correspondence
analysis computations are correct (same as -coranal- which produces same
results as in SPSS). But my help stops here... I do not have enough
expertise in the field to point you to finer explanations.
Hope this helps,
Philippe.
(*) Blasius J. and M. Greenacre (1994), 'Computation of Correspondence
Analysis' in Greenacre M. and J. Blasius (Eds.), Correspondence Analysis
in the Social Sciences - Recent Developments and Applications, Academic
Press, London.
<<<<<<<<<<<<<<<<<<<<<<<
RE: st: RE: MCA and cluster on STATA7?
From: "Nick Cox" <[email protected]>
Subject: RE: st: RE: MCA and cluster on STATA7?
Date: Thu, 23 Jan 2003 17:43:45 -0000
Next Article (by Date): st: Bootstrapped survival estimates Ty Ridenour
Previous Article (by Date): st: RE: command impute "Schonlau, Matthias"
Top of Thread: st: RE: MCA and cluster on STATA7? "Nick Cox"
Articles sorted by: [Date] [Author] [Subject]
Fr�d�rique Richy
>
> For the ACM, you advice me to contact Philippe Van Kerm but
> how can I join
> him. Does he have an e.mail? or does he regularly consult Statalist?
>
Philippe gets the digest. He therefore
gets messages at most daily.
Nick
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
**********************************************************************
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.
This footnote also confirms that this email message has been swept by
for the presence of computer viruses.
**********************************************************************
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
