I answer myself, perhaps it can be useful to anyone.
The problem is that I tried to run a psot hoc test before i get the minimal
model by deleting the interaction,
if I do it:
> ancova2<-aov(log(peso)~edadysexo+log(lcc))
> anova(ancova,ancova2)
Res.Df RSS Df Sum of Sq F Pr(>F)
1 509 10.085
2 511 10.148 -2 -0.063484 1.6021 0.2025
It's correct, there is no differences between models. And now i can run a
post hoc test
> summary(glht(ancova2, linfct=mcp(edadysexo="Tukey")))
Estimate Std. Error t value Pr(>|t|)
M - H == 0 0.15120 0.01761 8.588 <1e-05 ***
SUB - H == 0 -0.01584 0.02659 -0.596 0.812
SUB - M == 0 -0.16704 0.03248 -5.144 <1e-05 ***
(Adjusted p values reported -- single-step method)
>Two question:
>1) Why I don't get differences in edadysexo when i got it in the ancova
Wthi the minimal model i get it
>2) About the warning message. Why R said "covariate interactions found" of
i didn't get it in the ancova
Because in the first model "edadysexo" was in an interaction.
I hope this will be useful Mario
-----
Mario Garrido Escudero
PhD student
Dpto. de Biología Animal, Ecología, Parasitología, Edafología y Qca. Agrícola
Universidad de Salamanca
--
View this message in context:
http://r.789695.n4.nabble.com/problems-with-glht-for-ancova-tp4284908p4288226.html
Sent from the R help mailing list archive at Nabble.com.
______________________________________________
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.
