Message-ID: <1305550470019-3526143.post@n4.nabble.com>
Date: 2011-05-16T12:54:30Z
From: wwreith
Subject: MANOVA Post Hoc Testing Tukeys
I used manova() with one predictor variable and four factor levels call them
A, B, C, and D. There are 12 response variables. I now want to perform
pairwise comparisons for A-B, A-C, etc. for all 12 response variables.
If I were doing an ANOVA test I would run TukeyHSD() and be done. However
manova in R does not have Tukey's test or an equivalent that I know of.
Online I read that at this point I should run 12 anovas and then use a
Tukey's test after each ANOVA.
Here is my question: Is this the best way to go about Post Hoc Testing in R
after a MANOVA test? Also if this is the best path shouldn't I have some
kind of familywise error correction like Bonferroni of .05/13=.0038? The
post I saw said nothing about a Bonferroni correction.
Thanks for the help.
--
View this message in context: http://r.789695.n4.nabble.com/MANOVA-Post-Hoc-Testing-Tukeys-tp3526143p3526143.html
Sent from the R help mailing list archive at Nabble.com.