Dear list
Does anybody know whether STATA can perform multiple classification
analysis?
Further, could you recommend good reading materials for that?
Thank you in advance.
Kihong
I am mot clear what you mean by "multiple classification analysis."
In my field, that term refers to a regression model in which one
has a set of categorical predictors such as region and religion,
neither of which has an particular ordering. To use the variables,
one has to create a set of dummies for each. One can also
"dummy up" an order variable, sometimes after categorizing it.
For example, one could create categories out of a variable
measuring years of schooling. An MCA model then is a regression
model in which one uses set of dummies to represent categorical
variables with or without continuous covariates.
In this case, the equivalent of MCA can be obtained by judicious
use of the XI command.
The results of an MCA contain a set of coefficients relative to a
reference category, as does any dummy variable analysis. In
sociology, the term MCA usually means that one takes those
coefficients and converts them to deviations from the grand mean.
SPSS has a routine that does this automatically. To the best of
my knowledge STATA does not.
The literature on this is spotty. The original technique was developed
in a paper by Emanual Melichar, published in, if I recall, the Proceedings
of the Survey Research section of the ASA, sometime in the late 60's.
The old maroon version of the SPSS manual, for version 4 if I recall,
has a nice discussion. See me offline if you need more info.
=====================================
Richard T. Campbell
Professor of Sociology
Phone: 312/413-3759
Fax: 312/996-5104
More info: http://www.uic.edu/~dcamp
*
* 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/
