Below code should give your desired output. data want;   set have(rename=(other=other_old));   by loc;   if first.loc then seq=.;   seq+1;   retain old_ID new_ID old_PL new_PL Other;   if seq=1 then do; old_id=id; old_pl=pl; other=other_old; end;   if seq=2 then do; new_id=id; new_pl=pl; output; end;   drop id pl other_old seq; run; ... View more

If you are in clinical field, I would strongly recommend - Common Statistical Methods for Clinical Research with SAS Examples by Glenn A. Walker. This book almost covers all SAS Stat procedures applicable in clinical trials. ... View more

data one;           length Role A B C D E $20;           infile datalines dlm=',';           input Role A B C D E $;           datalines; IT,Bone,Diabetes,haemoglobin, ,Fibrin Doctor,Bone, ,Haemoglobin,Blood,  Doctor, ,Diabetes,haemoglobin,Blood,Fibrin Clinic,Bone, , ,Blood,Fibrin Clinic,Bone, ,Haemoglobin,Blood,       ; run; proc sql;           create table two as select count(a) as Bone, count(b) as Diabetes,count(c) as Haemoglobin,           count(d) as Blood,count(e) as Fibrin,count(*) as total,role from one group by role; quit; data two;           set two;           call symput(role,strip(put(total,best.)));           drop total; run; proc transpose data=two out=three name=Answer prefix=_;           id role; run; data final;           set three;           length It Doctor Clinic $20;           Clinic = strip(put(_clinic,best.))||'('||strip(put((_clinic/&clinic)*100,best.))||'%)';           doctor = strip(put(_doctor,best.))||'('||strip(put((_doctor/&doctor)*100,best.))||'%)';           It = strip(put(_it,best.))||'('||strip(put((_it/&it)*100,best.))||'%)';           drop _:; run; proc print noobs; run; ... View more

*Create sample Member dataset; data Member;           length Memno 8 Firstname Lastname address1 address2 city county zipcode $100;           infile datalines dlm=',';           input Memno Firstname Lastname address1 address2 city county zipcode; datalines; 12345,Kimberly,cowan,32 Moore ave, ,Nassau,NY,11556 ; run; *Create sample Demohistory dataset; data Demohistory;           length Memno 8 Firstname Lastname address1 address2 city county zipcode $100;           infile datalines dlm=',';           input Memno Firstname Lastname address1 address2 city county zipcode; datalines; 12345,Kimberly,cowan,130 Jerusalem st,apt #415,Nassau,NY,11556 12345,Kimberly,cowan,32 Moore ave, ,Nassau,NY,11556 12345,Kimberly,cowan,130 Jerusalem st,apt #415,Nassau,NY,11556 12345,Kimberly,cowan,130 Jerusalem st,apt #415,Nassau,NY,11556 12345,Kimberly,cowan,32 Moore ave, ,Nassau,NY,11556 ; run; *Sort the data before merging; proc sort data=member; by memno firstname lastname address1 address2 city county zipcode; run; proc sort data=Demohistory; by memno firstname lastname address1 address2 city county zipcode; run; data output;           merge member(in=a) demohistory(in=b);           by memno firstname lastname address1 address2 city county zipcode;           if b and not a; *Output records present in Demohistory which are not present in Member dataset; run; proc print; run; ... View more

Thank you. Will check with Tech support. ... View more

Hi Cynthia, Thanks a lot for the reply. Forgot to mention that I have SAS 9.1.3 in my machine, I am sorry about that. Is there a way to achieve this in SAS 9.1.3? Thanks! ... View more

Hello, You need to change the order of variables on the COLUMN statement. PROC REPORT processes variables in the order in which they are named on the COLUMN statement. Variables referenced in a COMPUTE block cannot refer to variables listed to their right in the COLUMN statement. col age sex new name; ... View more

Hi, In Excel you may add an apostrophe (single quote) at the begining of text - that should fix it. It forces Excel to treat the cell as text instead of a number. For e.g. '1-5 '6-10 '11-14 ... View more