Date:         Wed, 15 Nov 2000 12:07:50 +0100
Reply-To:     Per Jensen <perjn@TDK.DK>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Per Jensen <perjn@TDK.DK>
Organization: Tele Danmark Internet Cyberspace Launchpad
Subject:      Re: counting "non missing" values

Hi

You can use SQL with the COUNT function to do the job. See below.

data test;
input @1 var1 2. @3 var2 $1. @5 var3 2.;
datalines;
5 X  .
. Y  .
4   2.6
;
run;

proc sql;
create table numcount as
select count(distinct var1) as no_var1,
       count(distinct var2) as no_var2,
       count(distinct var3) as no_var3
from test;
quit;

Good luck

Per

welterh@my-deja.com skrev i meddelelsen <8utkdd$ofp$1@nnrp1.deja.com>...
>Hi!
>
>I would like to count the number of "non-missing" observations, but
>don't know how to do it. Is there an easy solution?
>
>Here my input dataset:
>
>OBS  VAR1 VAR2 VAR3
>1      5   X     .
>2      .   Y     .
>3      4   .     2.6
>
>And heres the answert I would like to have:
>VAR1 2
>VAR2 2
>VAR3 1
>
>Any suggestions???
>
>Cheers
>Heiko
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.