On Mon, 16 Feb 2004 demiurg at post.tau.ac.il wrote:
> I could not find any documentation about aov() and Error() usage.
> Both help pages (for aov and Error) does not specify it.
My copy says
If the formula contains a single 'Error' term, this is used to
specify error strata, and appropriate models are fitted within
each error stratum.
and gives an example *and* a reference. (If you know about this sort of
thing, that is sufficient information. If not, read the reference.)
> I'm mainly interested in aov with random effects.
Depends what you mean by that: it is not what the help pages says it does,
and depending what you mean it may be better to use lme in package nlme.
There are worked examples in MASS (see the FAQ) and this was actaully
implemented in R to support those examples.
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595