Date:         Sat, 28 Jun 2008 23:25:28 -0400
Reply-To:     Muthia Kachirayan <muthia.kachirayan@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Muthia Kachirayan <muthia.kachirayan@GMAIL.COM>
Subject:      Re: Table combining tables? PROC FREQ or maybe array? or checkall?
In-Reply-To:  <200806281956.m5SAlkbM007491@malibu.cc.uga.edu>
Content-Type: text/plain; charset=ISO-8859-1

On Sat, Jun 28, 2008 at 3:56 PM, Howard Schreier <hs AT dc-sug DOT org> <
schreier.junk.mail@gmail.com> wrote:

> On Sat, 28 Jun 2008 15:32:01 -0400, Peter Flom
> <peterflomconsulting@MINDSPRING.COM> wrote:
>
> >Hi again
> >
> >Another weird problem, but I'm sure the wizards of SAS-L can help..... it
> might be a case where Ron Fehd's checkall macro can help, but I'm not
> completely sure.....Ron's macro is at
> >http://www.sascommunity.org/wiki/Processing_Check-All-That-Apply
> >
> >Here's what I've got:
> >
> >1) A "check all that applies" question..... coded as a bunch of variables,
> each 0, 1, or 98 (for no, yes, or missing)
> >2) Another variable  with three levels
> >
> >Here's what I want:
> >A table of the second variable, for the cases where the first variables =
> 1.
> >
> ><<<
> >proc freq data = today;
> > table prop_part/out = Reason1;
> > where v3_1 =1;
> >run;
> >
> >proc freq data = today;
> > table prop_part/out = Reason2;
> > where v3_2 =1;
> >run;
> >>>>
> >
> >etc.
> >
> >Which gives output that looks like:
> >        prop_
> >Obs     part     COUNT    PERCENT
> >1                 13       .
> >2     0: Non       1      0.2674
> >3     1: Som     178     47.5936
> >4     2: All     195     52.1390
> >
> >for each reason.
> >
> >But then how do I get the results into a table that looks like this
> >
> >              None     Some   All    .
> >Reason1         1       178   195   13
> >Reason2
> >Reason3
> >
> >etc.?
> >
> >
> >
> >TIA as always
> >
> >Peter
> >
> >
> >
> >
> >
> >Peter L. Flom, PhD
> >Statistical Consultant
> >www DOT peterflom DOT com
>
> Please post data set TODAY.
>

Peter:

Your sample data is missing for the prop_part taking 0 and MISSING values. I
have added the first record with prop_part = 0: Som; second record with
'missing' in the sample data set used  under.

A single data step solution using array and  giving an output  data set,
NEED.

data today;
  input V3_1-V3_15  prop_part & $;
  cards;
  1  0  0  0  0  0  0  0  0  0   0   1   0   0   0   0: Som
  1  0  0  0  0  0  0  0  0  0   0   1   0   0   0   missing
  1  0  0  0  0  0  0  0  0  0   0   1   0   0   0   1: Som
  1  0  0  0  0  0  0  0  0  0   0   1   0   0   0   2: All
  1  0  0  0  0  0  1  0  0  0   0   1   0   0   0   2: All
  0  0  0  0  0  1  1  0  0  0   0   1   0   0   0   2: All
  0  0  0  0  0  0  0  0  0  0   0   0   0   0   0   2: All
  1  0  0  0  0  0  0  0  0  0   0   1   0   0   0   1: Som
  1  0  0  0  0  0  0  0  0  0   0   1   0   0   0   1: Som
  1  0  0  0  0  0  0  0  0  0   0   1   0   0   0   1: Som
  1  0  0  0  0  0  0  0  0  0   0   1   0   0   0   1: Som
  0  1  0  0  0  0  0  0  0  0   0   1   0   0   0   2: All
;
run;

data need;
array count[15,0:3] _temporary_;
array v v3_1-v3_15;
length a $1;
set today end = last;
do over v;
   a = prop_part; * Or  a = substr(prop_part,1,1);
   if notdigit(a) then i = 3; * Missing Value;
   else  i = input(substr(prop_part, 1, 1), 1.);
   count[_i_, i] ++  (v = 1);
end;
if last then
   do reason = 1 to dim(v);
      none = count[reason, 0];
      some = count[reason, 1];
      all = count[reason, 2];
      unknown = count[reason,3];
      output;
   end;
drop v: a i prop:;
run;


Muthia Kachirayan