At each measurement occasion, different service providers have been involved. They have used different variable names for IDs. Not a problem. But single panel waves also include lots of duplicated IDs, as identified for instance by
Code:
duplicates list idvar
Code:
duplicates report
The combination of many waves, seven or so different ID variables, and duplicate use of IDs proves to be a larger challenge than I expected.
Stata has advanced features for handling panel data, so maybe it's possible to solve the problem without doing all by hand or developing a new function? (I can't program in Stata, only in R.)
My aim is to develop panel data (long format, with -xtset-) that has no duplicate IDs within single waves and an ID variable that identifies as many individuals as possible with repeated participation across waves. The data include not only dropouts but also new participants who join in at later measurement occasions
0 Response to Panel data: Multiple IDs, partly duplicated
Post a Comment