Re: [R] Multiple comparison and two-way ANOVA design

From: Prof Brian Ripley <>
Date: Mon 16 Jan 2006 - 05:56:11 EST

Did you read the TukeyHSD help page? You chose to have one factor, so you cannot blame R for your own choice.

Do be very careful that you understand what these so-called post hoc tests do (and also what the main effect of a factor in the presence of an interaction means, for in R it is not the same as in SPSS).

[It really isn't fair to send UTF-8 mail, and the formatting has gone wrong en route to me (if it was ever right). Plain text and wrapped lines please.]

On Sun, 15 Jan 2006, Andrej Kastrin wrote:

> Dear useRs,
> I'm working on multiple comparison design on two factor (2 3 levels) ANOVA. Each of the tests I have tried (Tukey, multcomp package) seem to do only with one factor at a time.
> fm1 <- aov(breaks ~ wool * tension, data = warpbreaks)tHSD <- TukeyHSD(fm1, "tension", ordered = FALSE)
> $tension diff lwr upr p adjM-L -10.000000 -19.35342 -0.6465793 0.0336262H-L -14.722222 -24.07564 -5.3688015 0.0011218H-M -4.722222 -14.07564 4.6311985 0.4474210
> I'm interested in posthoc comparisons between various levels of factor 1 and various levels of factor 2, both at the same time.I know that is possible in SPSS, but... is there any R solution?
> Cheers, Andrej
> mailing list do read the posting guide!

Brian D. Ripley,        
Professor of Applied Statistics,
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

______________________________________________ mailing list PLEASE do read the posting guide!

Received on Mon Jan 16 06:04:00 2006

This archive was generated by hypermail 2.1.8 : Mon 16 Jan 2006 - 14:14:24 EST