Dear R users,
I know that it was considered many times but having searched through dozens
of posts I have only added to my confusion. My question is - is it possible
to correctly analyse simple mixed (fully)crossed (factorial) 2-way ANOVA?
Assume we have factor A (fixed) and B (random). In the model with an
interaction A should be tested against interaction, B and A:B against
residual variance. Specifying aov(response~A*B) gives wrong answer (wrong F
ratios, especially for fixed term as it's tested using residuals).
Specyfying aov(response~A*B+Error(A:B) specifies correct error term for
fixed effect, but wrong for random effect. And now I'm a bit confused - is
it possible to analyse such data using aov (or any alternative to aov) - or
it's better to switch to lme or lmer (which are better for mixed models but
are at first difficult to understand for students, especially in terms of
random interactions)?
Sorry for posting the same/similar question again - maybe at least it would
clarify the problem for future visitors.
Cheers,
sz.
--
Szymon Drobniak || Population Ecology Group
Institute of Environmental Sciences, Jagiellonian University
ul. Gronostajowa 7, 30-387 Kraków, POLAND
tel.: +48 12 664 51 79 fax: +48 12 664 69 12
www.eko.uj.edu.pl/drobniak
[[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.
