Svend,
Thanks again for your suggestions.
You're right that I could create a set of -drop- statements to remove all
null records (where all reshaped variables are missing) so that -reshape-
would create the unbalanced data set I need. Despite my general aversion
to using SPSS, I took a look at their version of -reshape-, called
-restructure-. There is a handy option to drop all null records, which
easily creates unbalanced data sets. It would be nice if Stata's
-reshape- command had such an option. Given the likelihood of adding new
variables as the data analysis progresses, I will likely use SPSS to
reshape my data.
Mike Frone
****************************************************************
Michael R. Frone, Ph.D.
Senior Research Scientist
Research Institute on Addictions
State University of New York at Buffalo
1021 Main Street
Buffalo, New York 14203