Date:         Mon, 3 Feb 1997 15:32:49 -0500
Reply-To:     "Brooke S. G. Molina" <brookem+@PITT.EDU>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         "Brooke S. G. Molina" <brookem+@PITT.EDU>
Subject:      warning message in a factor analysis
Content-Type: text/plain; charset="us-ascii"

My colleagues and I have conducted a factor analysis (principal components)
in SPSS (oblimin rotation) and have received an error message: "THE
CORRELATION MATRIX IS ILL-CONDITIONED."  Does anyone know how we can track
down the specific source of the problem and whether such difficulties with
the input matrix can be remedied?  Any help would be greatly appreciated.
 
Brooke Molina, Ph.D.
Assistant Professor of Psychiatry
Western Psychiatric Institute & Clinic
University of Pittsburgh School of Medicine
3811 O'Hara Street
Pittsburgh, PA 15213
412-383-9895
BROOKEM+@PITT.EDU