Date:         Tue, 15 Jul 1997 17:13:44 -0500
Reply-To:     Terri Akey <akeythe@MAIL.AUBURN.EDU>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         Terri Akey <akeythe@MAIL.AUBURN.EDU>
Subject:      Contrasts using the GLM procedure
Comments: To: SPSSX-L@VM.MARIST.EDU.
Content-Type: text/plain; charset=ISO-8859-1

I have a question about using the /lmatrix contrasts in the GLM program to
do follow-up tests.  Is there a way to specify a separate error term so as
not to assume homogeneity of variance for the different contrasts?
 
Thanks,
 
Theresa Akey
4054 Haley Center
Department of Educational Foundations,
Leadership, and Technology
Auburn University       Phone: (334) 844-3076
Auburn AL 36849 Fax: (334) 844-3072
E-mail: akeythe@mail.auburn.edu