I have an anova model that looks like this: aov(log(Y)~X+Error(Participant/X)) Is there a way to do a planned contrast on a repeated measures model... something with an error term? I get an error when I do this: contrasts(X)=cbind(-3,-1,1,3) summary.lm(aov(log(Y)~X+Error(Participant/X))) Any help would be appreciated! -Ryan
Bugzilla from rmh3093 at gmail.com wrote:> > I have an anova model that looks like this: > > aov(log(Y)~X+Error(Participant/X)) > > Is there a way to do a planned contrast on a repeated measures > model... something with an error term? I get an error when I do this: > contrasts(X)=cbind(-3,-1,1,3) > summary.lm(aov(log(Y)~X+Error(Participant/X))) > >Check the example in se.contrast, and consider switching to nlme/lme for this type of work. If you need a better answer, post a complete example. Dieter -- View this message in context: nabble.com/Planned-contrast-on-repeated-measures-tp22869889p22881896.html Sent from the R help mailing list archive at Nabble.com.