Message-ID: <525AD46B.60000@psychologie.uni-freiburg.de>
Date: 2013-10-13T17:12:11Z
From: Henrik Singmann
Subject: effect sizes in lmer
In-Reply-To: <CA+3amhcO8bTMTKUgejMYOYizr=xMqewMjJEpOqgobsYGLk63Dg@mail.gmail.com>
Dear Josh,
I don't know of a measure that serves this purpose. You could implement
something like change in AIC or Omega? but if this is reasonable, others
with a stronger statistical background will have to decide.
Cheers,
Henrik
Joshua Hartshorne schrieb:
> Hi Henrik,
>
> I'm not sure if I follow your reply. Using fixef will give me 24 effects,
> including the intercept. An ANOVA would give 7. What I need to report are
> the 7 ANOVA-style effects, not the 24 me-style ones.
>
> To make this more concrete, I have 4 types of stimuli, two different types
> of tests, and 3 conditions. What the readers are going to want to know is
> whether there is an omnibus interaction. Fixef reports 6 omnibus
> interactions -- one for every level of the interaction. I will also need to
> report the lower-level interactions and main effects. (Doesn't matter
> whether these are truly interpretable in the face of a significant
> higher-order interaction: It's standard practice to report them.)
>
> I can measure the significance of the ANOVA-style omnibus interaction by
> using model comparison. But that doesn't give me an effect size exactly.
> (One suggestion I heard recently was to use the change in AIC as an effect
> size.)
>
> Any ideas?
>
> Josh
>
> [[alternative HTML version deleted]]
>
--
Dipl. Psych. Henrik Singmann
PhD Student
Albert-Ludwigs-Universit?t Freiburg, Germany
http://www.psychologie.uni-freiburg.de/Members/singmann