I can recall this being discussed but not
the outcome. I think also that it is
occasionally a user request at users'
meetings, but never elicits much enthusiasm,
a good reason for that being that extra
machinery does not appear essential for the
purposes people want.
What seems clear is that Stata won't apply
value labels to a string variable, so this
boils down to setting up a correspondence
(not necessarily one-to-one) between two
string variables.
Unless the number of "labels" was very
small, the cleanest way of defining such
labels might be via a definition data
file which was then -merge-d with the
main dataset.
Nick
[email protected]
George M Hoffman MD
> i recall a discussion a few years ago about a command to
> encode a variable
> whether numeric or string, but cannot findit.
> any direction?
*
* 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/
