Hi R people,
I'm trying to do a Anova with 3 different tests (intra subject), 6 times of
measures (intra subject) and 3 groups (inter subject). I want to obtain an
Huynh-Feldt epsilon on accuracy with >anova(lm(acc.t1~gr.dem),test="Spherical")
for test 1 # gr.dem is my group variable
>anova(lm(acc.t2~gr.dem),test="Spherical") for test 2
>anova(lm(acc.t3~gr.dem),test="Spherical") for test 3
acc.t1, acc.t2 and acc.t3 was data frames with 6 columns each
the results was something like this:
Analysis of Variance Table
Greenhouse-Geisser epsilon: 0.2746
Huynh-Feldt epsilon: 0.2915
Df F num Df den Df Pr(>F) G-G Pr H-F Pr
(Intercept) 1 7352.7017 6 150 1.640e-182 6.894e-52 6.269e-55
gr.dem 1 0.6222 6 150 0.71230 0.51170 0.52064
Residuals 25
And so what could I do with my epsilon?
Previous aov() revealed me that there no effects of my groups but is my 6
times measures factor and my test factor were always significant?
Guillaume
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
