I am trying to genereate counts from a 3-way cross-classification of
variables into CSV, and eventually into EXCEL. I have many such 3 way
tables to make, with two variables (B (2 levels) and C (4 levels))
being
constant across a large set of variables (A) which represent the third
dimension. Of course, for any particular value of A, any particular
cross-classification with B & C may have a count of 0.
STATA clearly generates nice 3D tables on screen using the TABLE
command, and I have worked with the REPLACE option, but I am
struggling
to find a way to produce
a dataset out of STATA that I can readily and quickly. For example,
the TABLE, replace option does not generate row when the
cross-classification yields 0 observations, making it very difficult
to
cut and paste sections of data out of table, replace and put them
into
the spreadhseet. TABLE, replace, also has a disconcerting habit of
switching the order of variables across various A values, further
making
this messy and error-prone work.
I'd welcome any ideas. This must be very straight foward.
Thanks,
ROb
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
