|
Jack,
How about this - the first step just sets up some sample data that matches
yours:
** Set up sample data **;
data test;
do i = 1 to 57;
network = '01';
output;
end;
do i = 1 to 14;
network = '02';
output;
end;
do i = 1 to 29;
network = '03';
output;
end;
run;
proc sort;
by network;
run;
** Calculate groups of 20 or less **;
data groups;
set test;
by network;
retain group network_count 0;
if first.network or network_count eq 20 then
do;
group ++ 1;
network_count = 0;
end;
network_count ++ 1;
output;
run;
proc freq;
tables network*group / list;
run;
----------------------------------------------------------------------
Pete Lund
WA State Caseload Forecast Council
515 15th Ave SE
Olympia, WA 98504-0962
(360) 902-0086 voice
(360) 902-0084 fax
(360) 971-0962 pager
peter.lund@cfc.wa.gov
----------------------------------------------------------------------
-----Original Message-----
From: Jack Shoemaker [mailto:JShoemaker@ACCORDANT.NET]
Sent: Monday, September 11, 2000 12:01 PM
To: SAS-L@LISTSERV.UGA.EDU
Subject: Counting and subsetting question
Dear SAS-L,
I have a set of 100 people each with an attribute called network. The
distribution of these people is
Network Count
======= =====
01 57
02 14
03 29
I'd like to break these people into groups of no more than twenty, but
have each group start on a new network boundary. For example,
Network Group Count
======= ===== =====
01 1 20
01 2 20
01 3 17
02 4 14
03 5 20
03 6 9
I though this was easy until I tried to show someone else how to do this.
Can one of you experts take me to school? TIA - Jack
--
Jack N Shoemaker / JShoemaker@Accordant.net
Visit our patient communities at http://www.accordant.com or our corporate
site http://www.accordant.net
| 
