PSPP-BUG: crosstab bug on split data?

[Leonard Evens]

I have a dataset split according to the values of one variable.

When I do a cross tabulation of two variable with the split in place,
pgpp appears to do the right thing with the valid entries.
But it does something strange with respect to missing entries.

For the first subset of the split, it reports the correct number
of missing entries in that subset.  But for the second subset, instead
of reporting the number of missing variables for that subset, it reports
the total number of missing variables in the entire dataset.

I don't think this happens in every cross tabulation. It appears to
happen when one or more of the variables involved was a
recording of one of the original variables.

Of course, I could have done something wrong, e.g., messed up the
recording, but that doesn't appear to be the case.

If anyone is familiar with this phenomenon, please let me know.  If it
is a genuine bug, I hope this constitutes a report.
-- 
Leonard Evens address@hidden
Professor Emeritus, Department of Mathematics, Northwestern University