Hi, I want to ask which way is more effective to further analyse (multiple comparisons) a mixed model repeated measures anova with 2 fixed factor and 1 random? anova(lme(expr~treatment*age,random=~1|trial, data) Is searching for an effect of one factor in each of the subsamples defined by the second factor the best way anova(lme(expr~age,subset=treatment=="sal",random=~1|trial, data). or use multcomp? Thank you Panagiotis -- View this message in context: http://r.789695.n4.nabble.com/best-way-to-further-analyse-a-mixed-model-tp3874592p3874592.html Sent from the R help mailing list archive at Nabble.com.