Or you could try a SQL way which don't need to transpose dataset. If there were many variables, you can use a macro variable to wrap them all. data have;
infile cards missover;
input id var1 $ var2i $ var2ii $ var2iii $ var2iv $;
cards;
11 a a c
12 b
13 b b c
14 a a b c d
15 a c
16 b c d
17 b b d
18 b
19 a a c d
20 a d
;
run;
proc sql;
create table want as
select var1,sum(var2i='a')+sum(var2ii='a')+sum(var2iii='a')+sum(var2iv='a') as a,
sum(var2i='b')+sum(var2ii='b')+sum(var2iii='b')+sum(var2iv='b') as b,
sum(var2i='c')+sum(var2ii='c')+sum(var2iii='c')+sum(var2iv='c') as c,
sum(var2i='d')+sum(var2ii='d')+sum(var2iii='d')+sum(var2iv='d') as d
from have
group by var1;
quit;
Ksharp
... View more
