Date:         Sat, 4 Jan 1997 01:48:32 GMT
Reply-To:     gullionc@AOL.COM
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         gullionc@AOL.COM
Organization: AOL http://www.aol.com
Subject:      Re: simple effects w/ repeated measures

My recollection is that you can set this up as follows:
 
data out;set in;
  dummy = 1;
run;
 
proc glm;
  class dummy;
  model a1b1 a1b2 a1b3 a2b1 a2b2 a2b3 = dummy /nouni;
  repeated a 2, b 3 [contrast] / short summary nou nom;
run;
 
This will produce valid tests of a, b and a*b, as well as undefined test
results for dummy, dummy*a, dummy*b, and dummy*a*b, because DUMMY is a
constant.
 
It's been a while since I've used this setup, but I'm pretty sure it will
work.
 
Best wishes,
 
Christina M. Gullion, PhD