RE: st: manipulating matrix elements

[Steve Vaisey <vaisey@unc.edu>]

I'm on digest mode, so forgive the off-cycle nature of my replies.

I just checked the archives and saw Nick's question about the additive 
element.  The reference for this is:

Martin, John Levi. 1999. "Entropic Measures of Belief System 
Constraint." Social Science Research 28:111-134.

A simpler and perhaps more useful exposition is given in:

-----. 2002. "Power, Authority, and the Constraint of Belief Systems." 
American Journal of Sociology 197:861-904.

Adding the ln(M) [where M = number of cells] is meant to standardize the 
entropy (actually, the negative of the entropy) so that it varies from 0 
to 1 (i.e., from more to less entropic).  This standardization assumes, 
however, that no (or few) cells have particularly low counts.  As such, 
it's really only useful in large-n applications.  But since that's what 
I have, that's OK.

I still like the .ado version, since it allows me to do other things I 
might want to do.  Would you mind, Nick, adding the additive term into 
the program you wrote?  Thanks.

Steve




--
====================================================
Stephen Vaisey
National Science Foundation Graduate Research Fellow
Caroline H. and Thomas S. Royster, Jr. Fellow
Department of Sociology
University of North Carolina at Chapel Hill
200 Hamilton Hall (CB #3210)
Chapel Hill, NC 27599
====================================================

*
*   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/