Date:         Wed, 15 Sep 2010 10:30:16 -0700
Reply-To:     "Terjeson, Mark" <Mterjeson@RUSSELL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Terjeson, Mark" <Mterjeson@RUSSELL.COM>
Subject:      Re: A data step question: comparison of values of multiple
              variables
Comments: To: Daixin Yin <dyin@CCR.CA.GOV>
In-Reply-To:  A<201009151704.o8FGSYvA002492@willow.cc.uga.edu>
Content-Type: text/plain; charset="us-ascii"

Hi Daixin,

Here is one approach:

data sample;
    infile cards missover;
    input PATIENT_ID _01 _02 _03
          _04 _05 _06 _07 _08 _09;
cards;
001 8310
002 3510 8310
003 7500
004 1010
005 6910 8310
006 . 8310
007 0600
008 2100 2100
009 8000 8000 8000
010 . 3500
;
run;

proc transpose data=sample out=temp1;
    by PATIENT_ID;
    var _:;
run;

proc sql;
    create table diffs as
    select PATIENT_ID
    from temp1
    where col1 ne .
    group by PATIENT_ID
    having count(distinct col1) gt 1
    ;
quit;




Hope this is helpful.


Mark Terjeson
Investment Business Intelligence
Investment Management & Research
Russell Investments
253-439-2367


Russell
Global Leaders in Multi-Manager Investing




-----Original Message-----
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of
Daixin Yin
Sent: Wednesday, September 15, 2010 10:04 AM
To: SAS-L@LISTSERV.UGA.EDU
Subject: A data step question: comparison of values of multiple
variables

Hi there,
I have a quesion about comparing values of multiple variables within an
observation.

Here's how my dataset looks like:
 PATIENT_ID  _01   _02   _03   _04  _00  _05  _07  _60  _06  _08  _09
-----------------------------------------------------------------------
        001  8310
        002  3510  8310
        003  7500
        004  1010
        005  6910  8310
        006        8310
        007  0600
        008  2100  2100
        009  8000  8000  8000
        010        3500

Each observation has at least one non-missing value in variable _00 to
_60. I want to identify observations that had unequal values in variable
_00 to _60, such as PATIENT_ID 002 and 005 in the above sample dataset.

I believe that there must be more than one easy and efficient way to do.
Thanks for your help!!