[R-sig-eco] Post-hoc test after rm-ANOVA

[Ivailo]

On Fri, Sep 14, 2012 at 1:55 PM, Nils Gülzow <nils.guelzow at gmx.de> wrote:
> Dear community,
>
> I have a general statistic question. I have done a rm-Anova with
> ezAnova package. My rm-Anova  was a mixed within-and-between,
> design, containing two factors (each has two levels) for between and
> the sampling days for within. The ezAnova worked fine and now I would
> like to perform a Post-hoc test (like TukeyHSD) to get the significant
> differences between the factors over time.
>
> Is it simply using the TukeyHSD to test the differences between the
> factors or how can I get the significant difference between the
> factors over time?
>
> Thanks in advance,
> Nils
> Nils Gülzow

Dear Nils,

if you'd like to stick the the ezAnova package, then the ezPlot()
function produces error bars that might help you to visually evaluate
post-hoc comparisons. You can always run TukeyHSD() on an aov() object
or use glht() from the multcomp package if you have a lm() object.

Cheers,
Ivailo
-- 
UBUNTU: a person is a person through other persons.