Date:         Tue, 15 Mar 2005 12:22:12 -0500
Reply-To:     "Howard Schreier <hs AT dc-sug DOT org>" <nospam@HOWLES.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Howard Schreier <hs AT dc-sug DOT org>" <nospam@HOWLES.COM>
Subject:      Re: Fwd: Separating character variables for transpose

Try this:

   proc transpose data=sashelp.class out=n prefix=n;
   by name;
   run;

   proc transpose data=sashelp.class out=c prefix=c;
   by name;
   var _character_;
   run;

   data _all_;
   set n c(where = (_name_ ne 'Name') );
   run;

On Tue, 15 Mar 2005 16:23:59 -0000, SUBSCRIBE SAS-L Anonymous
<sas@BOOSECAT.COM> wrote:

Sorry if this is a repost...I sent to an address entry "SASL" however it
doesn't look like it sent it to the full SAS-L@LISTSERV.UGA.EDU
address......................


Hi all,
I'm kind of continuing from something I posted yesterday. In short, I'm
wanting to transpose the data in a dataset. However I have both character
and numeric fields that I need to transpose so I need to separate these out
without hardcoding the variables.

I've determined what variables are char/num several different ways but
urrently I'm doing:
proc contents noprint data = ds1 out = varlist (keep = name type);
run;

data char num;
set varlist;
if type = 1 then output num;
if type = 2 then output char;
run;

So I now know what variables are character and which are numeric. How can
I merge this back into my original "DS1" dataset so I know what to break
out? I've tried several different things including trying to create a
macro variable to hold the name of the char variables (IE, charlist =
charlist || ' ' || name).

If anyone has a minute and is able to solve this, I will be very grateful!
I have similar problems to this come up often. As always, thanks for you
time and help!

Andy