Date:         Tue, 20 Nov 2001 15:27:24 +0000
Reply-To:     "R. Allan Reese" <R.A.Reese@GRI.HULL.AC.UK>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "R. Allan Reese" <R.A.Reese@GRI.HULL.AC.UK>
Subject:      Query: Repeated measure, polynomial contrast
Content-Type: TEXT/PLAIN; charset=US-ASCII

I have a table of data with one value for each month at a number oflocations. Clearly the monthly values at any location are correlated, so
it seems to me appropriate to fit a quadratic curve across the months and
test for differences in the coefficients between locations.  What I cannot
fathom how to do is to stop SPSS fitting and testing polynomials up to
order 11, and therefore reporting no error term.  Can anyone advise please
whether a specific set of values after the POLYNOMIAL keyword can do this,
or if I'm barking ... up the wrong tree / mad.

GLM
  jan feb mar apr may jun jul aug sep oct nov dec BY station
  /WSFACTOR = response 12 Polynomial
  /METHOD = SSTYPE(3)
  /EMMEANS = TABLES(station)
  /CRITERIA = ALPHA(.05)
  /WSDESIGN = response
  /DESIGN = station .

R. Allan Reese                       Email: r.a.reese@gri.hull.ac.uk