Oh, so this is more about understanding what and why Proc Genmod returns such results and not actual data issues. Not my area. I suggest you post this as a new question in one of the Analytics forums. You will likely get the best advice if you can provide/mock-up some sample data and code with creates such "issues". Just as a side note: You can also calculate your log() directly as part of the SQL as done below. data have;   infile datalines truncover;   input (sex age area n pop period) (:8.);   datalines; 1 1 3 20 20 1995 1 1 2 5 275 1995 1 1 1 10 300 1995 1 2 3 8 310 1995 1 2 2 26 280 1995 ; run; proc sql;   create table want as   select area, period, sum(n) as nsum, sum(pop) as popsum, log(sum(pop)) as logpop   from have   group by area, period   ; quit; ... View more