There are usually two ways to get it.
One way: using data step
[pre]
data dataset1;
input id subid ;
datalines;
10 1
10 3
10 5
20 2
20 4
;
data dataset2;
input id subid emp ;
datalines;
10 1 10
10 2 20
10 3 30
10 4 40
10 5 50
20 1 10
20 2 20
20 3 30
20 4 40
20 5 50
;
data new;
merge dataset2(in=two) dataset1(in=one) ;
by id subid;
if one and two;
run;
proc print data=new noobs;
run;
[/pre]
Another way: using proc sql
[pre]
proc sql;
create table another as
select one.id,one.subid,two.emp
from dataset1 as one,dataset2 as two
where one.id=two.id and one.subid=two.subid
;
quit;
proc print data=another noobs;
run;
[/pre]
The code above are working.