Dear Colleagues,
I have scoured the help files and been unable to find an answer to my
question. Please forgive me if I have missed something obvious.
I have run the following two models, where "category" has 3 levels and
"comp" has 8 levels:
mod1 <- lmer(x~category+comp+(1|id),data=impchiefsrm)
mod2 <- lmer(x~category+comp+category*comp+(1|id),data=impchiefsrm)
followed by:
anova(mod1,mod2)
The anova shows that the interaction term specified in the second model is
significant when added to the main effects model.
Now I'd like to run post hoc comparisons using glht to discern where the
interaction means differ. For example, for post hoc comparisons on the
means of the main effect of "category" I can run:
summary(glht(mod2,linfct=mcp(category="Tukey")))
But this only gives me the mean comparisons for the "category" main effect
means. Essentially I'd like to run the following:
summary(glht(mod2,linfct=mcp(category*comp="Tukey")))
to get the mean comparisons for the interaction means. Perhaps needless to
say, this command does not work.
Can someone tell me how to run multiple comparisons among the interaction's
means?
I suspect that specifying the correct contrasts would do it, but I can't
figure out how to setup the contrasts...
Thanks,
Larry
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
