That's a kludge. You can analyse such models in a principled way using either summary.manova or lme, both of which model the correlation in the `repeated measures' rather than wish it away.
Could you try to use a more informative subject? Although I knew what the correction was, I have never called it that. On Mon, 27 Jan 2003, Heleen Slagter wrote: > I was wondering whether there are any packages that provide for the > Greenhouse-Geisser correction, an adjustment used in univariate repeated > measures when the sphericity assumption is violated (both numerator and > denominator degrees of freedom are multiplied by GG-epsilon, and the > significance of the F ratio is evaluated with the new degrees of freedom)? I > have seen a few emails with the same question, to which nobody replied, but > am hoping that maybe in the mean time somebody has tried to work this out. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595 ______________________________________________ [EMAIL PROTECTED] mailing list http://www.stat.math.ethz.ch/mailman/listinfo/r-help