S needs to be trimmed.

Q1_4text = tranwrd(Q1_4text,trim(s),';');

I'm not sure this is the best approach, can you explain the overall goal. This looks like a check all that apply with OTHER category that you are trying to capture.

On Wed, Feb 20, 2008 at 10:02 AM, Xuhong Zhu <zhuxuhong2000@gmail.com> wrote: > Hello Data _null_, > Thanks for your help. However,I found some strange results after I > tried your solution on my dataset > . I output 10 observations here. Hope you could help me take a look! > Always, thanks very much! > > data Websur_2; > length Q1_4text $50; > set Websur_1 (keep= Q1_4); > > Q1_4text=Q1_4; > length s $30; > do s = 'ARNP','DO','MD','MS', 'PhD', 'RN', 'Other (Please Specify)'; > Q1_4text = tranwrd(Q1_4text,s,';'); > end; > Q1_4text=left(compress(Q1_4text,';')); > drop s; > run; > > proc print data=websur_2 (obs=10); > run; > > /******* partial output ******/ > Obs Q1_4text > > 1 MD > 2 > 3 > 4 > 5 > 6 MD > 7 > 8 > 9 MDOther (Please Specify)MBA > 10 > > Obs Q1_4 > > 1 MD;MS > 2 MD > 3 MD > 4 MD > 5 DO > 6 MD;MS > 7 MD > 8 MD > 9 MD;Other (Please Specify);MBA > 10 MD > > > > > On Feb 19, 2008 4:51 PM, data _null_, <datanull@gmail.com> wrote: > > Assuming the removal of A; B; or C; leaves the desired text. > > > > data work.test; > > input line $50.; > > newLine=line; > > do s = 'A;','B;','C;'; > > newline = tranwrd(newline,s,' '); > > end; > > newline=left(compress(newline,';')); > > drop s; > > cards4; > > A;B;C;south asian; > > C;Indian; > > A;C;B; > > ;;;; > > run; > > proc print; > > run; > > > > On Feb 19, 2008 4:41 PM, Xuhong <zhuxuhong2000@gmail.com> wrote: > > > Hello folks, > > > > > > I need your help with my coding scheme. > > > > > > I have a variable falling into the following pattern: > > > > > > A;B;C;south asian; > > > C;Indian; > > > A;C;B; > > > ... > > > > > > What I want is to find a method to create a new variable which will > > > contains > > > the text fields which are not the 'A' and 'B' and 'C'. > > > > > > Thanks a lot! > > > > > >