Hi On Wed, 5 Nov 2003, Wuensch, Karl L wrote:
> I don't think you can without writing the syntax, at least not if you want > the sort of statistics you need properly to interpret a canonical regression > -- that is, you need to use the MANOVA program -- and even there you have to > be VERY careful reading the output -- the output for the redundancy > statistics may be worded in such a way that leads you to confuse the > proportion of variance in the X variables explained by the Y canonical > variate with the proportion of variance in the Y variables explained by the > X canonical variate. I abandoned SPSS for canonical analysis years ago, > switching to SAS. > > If there is an easy way to do it in SPSS, I expect Jim Clark will let us > know. SPSS for windows will accept syntax, of course. > > Karl W. > > -----Original Message----- > From: Hetzel, Rod [mailto:[EMAIL PROTECTED] > Sent: Wednesday, November 05, 2003 6:57 AM > To: Teaching in the Psychological Sciences > Subject: one more spss question > > > Hi folks. > > Thanks to all of you who answered my previous spss windows question. I have > one more question. Does anyone know how to conduct a canonical correlation > analysis using spss windows? CCA is essentially a multiple regression with > more than one dependent variable. I know how to write the syntax for a CCA, > but I'm trying to teach someone how to do it using the windows program. > Thanks! I don't know much about canonical correlation, period, let alone an easy way to do it in spss. SPSS comes with a macro "Canonical correlation.sps' that you should find in the SPSS program directory (at least it was in mine). Looking at it suggests that syntax will be necessary. Best wishes Jim ============================================================================ James M. Clark (204) 786-9757 Department of Psychology (204) 774-4134 Fax University of Winnipeg 4L05D Winnipeg, Manitoba R3B 2E9 [EMAIL PROTECTED] CANADA http://www.uwinnipeg.ca/~clark ============================================================================ --- You are currently subscribed to tips as: [EMAIL PROTECTED] To unsubscribe send a blank email to [EMAIL PROTECTED]
