Date:         Fri, 9 Oct 1998 10:13:22 -0400
Reply-To:     "Fehd, Ronald J." <rjf2@CDC.GOV>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         "Fehd, Ronald J." <rjf2@CDC.GOV>
Subject:      Re: Ordering word tokens
Content-Type: text/plain

> > From:       Robert Lokhamp [SMTP:rlokhamp@WORLDNET.ATT.NET]
> > In the course of a linguistic study, I've created a data set WORDS with
> > 10 character variables W1-W10. It looks something like that:
> > W1      W2      W3      W4      W5        W6    W7    W8    W9   W10
> > HUMPTY  DUMPTY  SAT     ON      A         WALL
> > HUMPTY  DUMPTY  HAD     A       GREAT     FALL
> I am trying to
> > find an effective way of replacing the data set with one where the
> > tokens would be in alphabetic order like
> > W1      W2      W3      W4      W5        W6       W7    W8    W9   W10
> > A       DUMPTY  HUMPTY  ON      SAT       WALL
> > A       DUMPTY  FALL    GREAT   HAD       HUMPTY
>
I think you want to sort the words in each observation.
Check out the ordinal function;
the documentation should read:
The ORDINAL function returns the
"Nth-sized item in the ascending list"
so you might use
 
array word {*} $ W1-W10;
array wordsort {*} $ X1-W10
 
do I = 1 to dim(word);
 wordsort{I} = ordinal(I,of word{*}); end;
 
Ron Fehd  the macro maven  CDC Atlanta GA