<>
Johanna wondered
I am working with a dataset where two identifier numbers uniquely identify
an individual. How do I create one identifier from the household and person
identifiers in STATA?
Martin suggested egen,group() but this would seem to be not very
handy, as the resulting ID will be an arbitrary integer. When I had
to do something like this with directorID (a number) and case code, I did
g dircase = string(directorID)+string(casenr)
That has the advantage of retaining both variables' contents in
readable form, and making it string avoids any overflow problems.
Naturally you could put in a separator such as +"_"+ in the
expression if you wish.
That separator may be necessary to insure uniqueness -- unless you
are certain that the length of str(directorID) is constant (which you
can insure with an appropriate format,
