Date:         Fri, 3 Jul 1998 01:07:38 -0500
Reply-To:     "Nichols, David" <nichols@SPSS.COM>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         "Nichols, David" <nichols@SPSS.COM>
Subject:      Re: Singular Special Transformation matrix
Comments: To: Andrew Martin <a.martin@NEPEAN.UWS.EDU.AU>

If you'll show us the command syntax, I or someone else will be glad to
point out why it's singular. Specifying one that's not is the only way
around this.
 
David Nichols
Principal Support Statistician and
Manager of Statistical Support
SPSS Inc.
 
>----------
>From:  Andrew Martin[SMTP:a.martin@NEPEAN.UWS.EDU.AU]
>Sent:  Thursday, July 02, 1998 12:05 AM
>To:    SPSSX-L@UGA.CC.UGA.EDU
>Subject:       Singular Special Transformation matrix
>
>Hi Folks
>
>I have 6 repeated measures and one between subjects factor (sex) and am
>wanting to perform nonorthogonal contrasts comparing various pairs of the
>repeated measures for each level of sex (using MWITHIN). To do this I have
>used the Transform subcommand and specified the matrix using the Special
>keyword. For all but one of the specified pairs it works fine - but when I
>introduce one pair into the matrix it gives me the following message:
>
>Error #12265
>Singular matrix for MANOVA - The Special Transformation matrix is singular.
>The analysis is terminated.
>
>I'd appreciate any ideas as to what is causing the problem and how I can
>perform these repeated measures contrasts so that I get around it.
>
>Regards
>
>Andrew
>*******************************************
>Andrew Martin
>Faculty of Education
>University of Western Sydney, Nepean
>PO Box 10 Kingswood NSW 2747 Australia
>Tel: 0247 360 104
>Fax: 0247 360 400
>E-mail: a.martin@uws.edu.au
>