I sent this message yesterday but from the responses I've received so far I guess I wasn't clear enough as to what I wanted at the end. I've included in this message an example of the output dataset that I'd like at the end.

I have a dataset that I need to transpose. Here is an example of what needs to be transposed.

data mailaud; input custid $1. tovid $8. audcode $10.; cards; 1 mail0299 primary 1 mail0699 secondary 1 mail0799 primary 2 mail0499 none 3 mail0799 primary 4 mail0898 secondary 4 mail297 secondary ; run;

There are specific requirements around the code however. I cannot hard code the variable names because the list of tovids keeps being updated.

I need an id level dataset where the variables are mail+#### (#### referring to the numbers at the end of mail for the tovid), aud+####, and custid.

So for example, for the tovid = mail0299, the new variable names would be mail0299, and aud0299 with values equal to 0 or 1, and the audcode respectively. I have tried doing this multiple ways and am at a loss as to how to continue. I tried doing this in a data step with macro variables being reassigned, but I get stuck when trying to use call symput since I have to update it in a separate data step. Any suggestions?

Here is what I'd like as output:

<<...>>

I'm including my last ditch code even though I know it's wrong to give you an idea of where I was headed. If it is hopelessly wrong please let me know. I'm definitely still learning the SAS programming ropes!

DATA MAILHIST; SET MAILHIST.HISTORY END = EOF; BY CUSTID; RETAIN; %MACRO DAT; %DO %UNTIL(_N_ = EOF); IF INDEX(tovid,'MAIL') THEN DO; TOV = SUBSTR(TOVID,5,3); CALL SYMPUT('X',TOV); MAIL&X = 1; LENGTH AUD&X $10; AUD&X = AUDCODE; END; %END; %MEND DAT; %DAT IF LAST.CUSTID THEN DO; ARRAY MISN[*] _NUMERIC_; DO I = 1 TO DIM(MISN); IF MISN[I] = . THEN MISN[I] = 0; OUTPUT; END; END; RUN;

Thanks!

Karen Kinsey Prevision Marketing 55 Old Bedford Road Lincoln, MA 01773 (781) 259-5142