On Thu, Apr 17, 2008 at 3:49 PM, Jane <program.sas@gmail.com> wrote:

> Can I do this without using array?

you can transpose the data to a long and tall foramt, then a simple SQL or data step can get what you need. for example: the transpose the data looks like: ID _name_ vaue aa var1 2 aa var2 8 ....

bb var1 2 bb var2 9 ....

proc sql; select distinct ID from dataset order by ID where value=1; quit;

> > > On Thu, Apr 17, 2008 at 3:44 PM, Yu Zhang <zhangyu05@gmail.com> wrote: > > > an untested code, HTH. > > > > Yu > > > > > > data wanted; > > set have; > > array temp {*} var1-var30; > > call sortn(of var1- var30); > > > > do _n_=1 to hbound(temp); > > if temp(_n_)=1 then do;kp=1;leave;end; > > end; > > > > if kp ; > > drop kp; > > run; > > > > On Thu, Apr 17, 2008 at 3:31 PM, Jane <program.sas@gmail.com> wrote: > > > > > Hi all > > > > > > I have a stupid question. If I have a lot of variables, like: > > > > > > var1 var2.....var30. > > > > > > If any of them equal to 1, I want to pull out that row. Just like: > > > > > > id var1 var2 var3 var4 var5 var6 var7... > > > aa 2 8 9 4 8 9 0 > > > bb 7 1 2 5 . 0 12 > > > cc 0 2 1 .................... > > > > > > Then I can get the dataset like: > > > id > > > ---- > > > bb > > > cc > > > > > > How can I do this in some simple way? I do not need to use > > > "if..then.." for > > > 30 variables. > > > > > > Thank you very much! > > > Jane > > > > > > > >