Date:         Wed, 15 May 2002 11:20:19 -0700
Reply-To:     prasad.s.ravi@HOUSEHOLD.COM
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Prasad Ravi <prasad.s.ravi@HOUSEHOLD.COM>
Subject:      Re: Count across the row
Comments: To: qjiang <qjiang@GMCF.ORG>
Content-type: text/plain; charset=us-ascii

I am not sure if this counts as an easy way but you could do this in a data
step.


DATA ONE;
INPUT ID      v1 $      v2 $      V3 $     V4 $      V5 $;
DATALINES;
1       E       E       E       .       3
2       E       .       .       2       E
;
RUN;
PROC SORT DATA=ONE;
 BY ID;
RUN;
DATA TWO;
   SET ONE;
   BY ID;
   TOTAL_COUNT=0;
   COUNT_E=0;
   ARRAY VARS (*) V1-V5;
   DO I=1 TO DIM(VARS);
      TOTAL_COUNT+1;
       IF VARS(I)='E' THEN COUNT_E+1;
       PCT_E=(COUNT_E/TOTAL_COUNT)*100;
   END;
   DROP I;
   IF LAST.ID THEN OUTPUT;
RUN;
PROC PRINT;
RUN;


      OUTPUT

TOTAL_
              Obs    ID    v1    v2    V3    V4    V5     COUNT    COUNT_E
PCT_E

               1          1    E       E      E                3
5                   3                  60
               2          2    E                          2      E
5                   2                  40


Prasad Ravi









                        qjiang                 To:    SAS-L@LISTSERV.UGA.EDU
                        <qjiang@GMCF.ORG>      cc:
                        Sent by: "SAS(r)       Subject:    Count across the row
                        Discussion"
                        <SAS-L@LISTSERV.U
                        GA.EDU>


                        05/15/2002 10:55
                        AM
                        Please respond to
                        qjiang





Hi all,

I have a PROC COMPARE output data set with about 400 variables compared.
The data set is like:

ID      v1      v2      V3      V4      V5
1       E       E       E       .       3
2       E       .       .       2       E
.....

Is there an easy way to count the percent of equals ?

My expected output
ID      %eq
1       60%
2       40%

Thanks in advance.

Qin