[R] post hoc comparison in repeated measure

From: array chip <arrayprofile_at_yahoo.com>
Date: Wed 10 May 2006 - 03:26:38 EST


Hi, I have a simple dataset with repeated measures. one factor is treatment with 3 levels (treatment1, treatment2 and control), the other factor is time (15 time points). Each treatment group has 10 subjects with each followed up at each time points, the response variable is numeric, serum protein amount. So the between subject factor is treatment, and the within subject factor is time. I ran a 2-way ANOVA with repeated measures considering time as the within subject factor:
aov(response~treat*time+Error(subject/time),dat)

The results told me that the treatment is marginally significant (p=0.04). I would like to know where that significance came from, so I did ALL pairwise t tests
(treat1 vs. control, treat2 vs. control, treat1 vs.
treat2) at each of the time point. There are 2 ways I can do these t tests, using the MSE from the ANOVA
(the MSE used for the treatement effect in the ANOVA,
i.e. the treatment by time interaction) as the t test error, or simply ran ordinary t tests using only the data of the treatment levels in comparison. What I found is that using the first approach, I couldn't find any pairwise comparison statistically significant which I thought I should find at least one significant, because the ANOVA treatment effect is mariginally significant (p = 0.04). Using the second approach, I did find some pariwise comparisons significant. Can anyone explain to me why?

BTW, is there a R function that can do post hoc comparison on repeated measure ANOVA (from avo() with Error term)?



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Wed May 10 03:36:50 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Thu 11 May 2006 - 10:10:08 EST.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.