This is a potentially very simple (and quite possibly silly) question, but its not one I've yet been able to fathom. When I extract the xwaveid dataset (in Stata 13) from SN6614, the pidp variable summary doesn't match that shown in the online documentation <www.understandingsociety.ac.uk/documentation/mainstage/dataset-documentation/wave/xwave/datafile/xwaveid/variable/pidp>
This seems to be because there are 27925 pidp cases where the pidp number is 0. According to the online summary, there shouldn't be any cases where pidp==0.
Is there something very obvious I am missing here?
Updated by Redmine Admin about 10 years ago
- Status changed from New to In Progress
- % Done changed from 0 to 90
By way of clarification, we operate we three types of variables for the documentation, a) continuous variables, b) categorical variables, and c) identifiers. The first type is documented with basic summery statistics, the second with frequencies, and third type only with an overall count of values greater than zero ("[value GT 0]".
Looks like your example is reported as a type a instead of a type c. Thanks for making us aware of this inconsistency. The way we intend to report pidp can be seen here;