Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Similarity measure for categorical data


From   "Clarence C. Gravlee" <cgravlee@fsu.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Similarity measure for categorical data
Date   Wed, 24 May 2006 16:39:23 -0400

Dear Statlisters,

I have a standard person-by-variable dataset and would like to create a
person-by-person similarity matrix.  The data are multiple-choice type
format (a nominal variable with 4 categories).  I'd like to compute the
similarity matrix using a simple matching measure, where the coefficient
would be the proportion of times that two people agree exactly.

As I understand it, the simple matching coefficient available in Stata
assumes that the data are binary: [MV] measure_option says "Stata treats
nonzero values as one when a binary value is expected" (p. 362).

Is there a built-in function for calculating a more general matching
coefficient for nominal data?  If not, has anyone else programmed it?

Thanks for your time.

Lance Gravlee

--
Clarence C. Gravlee, Ph.D.
Assistant Professor
Department of Anthropology
Florida State University
Tallahassee, FL 32306-7772
(850) 644-4754
(850) 645-0032 FAX
cgravlee@fsu.edu
http://lance.qualquant.net/ 

Research Associate
Center for Demography and Population Health


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index