topic Re: Save proc freq values into new dataset in SAS Programming

Save proc freq values into new dataset

Justin9 — Sun, 27 Jun 2021 20:42:34 GMT

Please can someone tell me how to save the unique values that are shown from a proc freq of a variable in the 'Results' tab and save them as a new dataset with just that one variable and all of the values? The proc freq of my variable shows that there are over 100 different values.

Re: Save proc freq values into new dataset

PaigeMiller — Sun, 27 Jun 2021 20:49:34 GMT

You use the OUT= option of the TABLES statement in PROC FREQ.

Re: Save proc freq values into new dataset

Justin9 — Sun, 27 Jun 2021 21:06:51 GMT

Thanks! Is there a way to exclude the 'count' and 'percent' columns in the output (from the out= option, not the proc freq), so that I am left with just the values of the variable? Is there also a way to exclude the missing row in the output?

Re: Save proc freq values into new dataset

PaigeMiller — Sun, 27 Jun 2021 21:41:56 GMT

... so that I am left with just the values of the variable

You can use the NLEVELS option in the PROC FREQ statement if you just want the levels and not the count or percent. And ODS output to get it into a data set.

Re: Save proc freq values into new dataset

tarheel13 — Sun, 27 Jun 2021 23:21:25 GMT

If you just want the unique values, you could do select distinct var from dataset in proc sql.

Re: Save proc freq values into new dataset

Reeza — Mon, 28 Jun 2021 01:40:05 GMT

@Justin9 wrote:
Thanks! Is there a way to exclude the 'count' and 'percent' columns in the output (from the out= option, not the proc freq), so that I am left with just the values of the variable?

Yes, you can use data set options in the output such as DROP/KEEP.

tables sex / out=want (drop = percent);

@Justin9 wrote:
Is there also a way to exclude the missing row in the output?

Yes, you can use data set options in the output such as WHERE.

tables age / out=want (where = not missing(age));

Putting it all together:

data class;
set sashelp.class;
if age=14 then call missing(age);
run;


proc freq data=class noprint;
tables age / out=want (where=(not missing(age)) 
                       drop = percent  )  ;
run;

proc print data=want;
run;