Hi there,
I was wondering if there is a way to add a column with the std dev next to the sum of certain variables in PROC TABULATE. I only seem to get the STD to work when i am calculating the mean value of variables but I use weights to calculate the sum of variables and I want to include the standard deviation of using those weights (basically the same standard deviation number you get when u use surveymeans to calculate the number of occurrences a variable appears).
I have tried putting the WEIGHT statement in various places, I've tried using the vardef=weight but nothing seems to work.
Here is my code
PROC TABULATE
DATA=kid.prs (where=(PRS=1));
VAR total;
CLASS FEMALE / ORDER=UNFORMATTED MISSING;
CLASS race / ORDER=UNFORMATTED MISSING;
CLASS hosp_division / ORDER=UNFORMATTED MISSING;
CLASS zipinc / ORDER=UNFORMATTED MISSING;
CLASS dispuniform / ORDER=UNFORMATTED MISSING;
CLASS hosp_locteach / ORDER=UNFORMATTED MISSING;
CLASS gtube / ORDER=UNFORMATTED MISSING;
CLASS MDO / ORDER=UNFORMATTED MISSING;
CLASS TLA / ORDER=UNFORMATTED MISSING;
CLASS Trach / ORDER=UNFORMATTED MISSING;
CLASS syndromic / ORDER=UNFORMATTED MISSING;
keyword all sum / style=[fontwidth=wide];
keylabel all="Total";
TABLE /* Row Dimension */
FEMALE RACE hosp_division zipinc dispuniform hosp_locteach,
/* Column Dimension */
Total*(gtube)*(Sum='N'*f=comma12.0 ColPctSum='%'*f=pctf.) / nocellmerge;
;
WEIGHT DISCWT;
Format race racef.;
Format female femalef.;
Format hosp_division divisionf.;
Format zipinc zipincf.;
Format dispuniform dispf.;
Format hosp_locteach hosp_locteachf.;
RUN;
