Re: [R] Anova's in R

From: Simon Blomberg <>
Date: Fri 29 Jul 2005 - 09:39:56 EST

If I understand you correctly, this looks like a split-plot design.

The anova is:

aov(response~burning*temperature+Error(site), data=dataset)

Alternatively in lme:

lme(response~burning*temperature, random=~1|site, data=dataset)

At 03:09 PM 29/07/2005, you wrote:
>I am looking for some help using anova's in RGui.
>My experiment ie data, has a fixed burning treatment (Factor A) 2 levels,
>Nested within each level of Factor A are 2 random sites (Factor B).
>All sites are crossed with a fixed temperature treatment (Factor C) 2
>levels, 0 degreesC/2 degreesC, with many replicates of these temperature
>treatments randomly located at each site.
>I am trying the following
>variable~burning*temperature*site+Error(replicate),data=dataset) and
>variations on that, however can't get it quite right.... the F ratios are
>not correct. I imagine this is a fairly common experimental design in
>ecology and would ask that anyone who has some advice please reply to this
> mailing list
>PLEASE do read the posting guide!

Simon Blomberg, B.Sc.(Hons.), Ph.D, M.App.Stat. Centre for Resource and Environmental Studies The Australian National University
Canberra ACT 0200
T: +61 2 6125 7800 email: F: +61 2 6125 0757
CRICOS Provider # 00120C mailing list PLEASE do read the posting guide! Received on Fri Jul 29 09:46:46 2005

This archive was generated by hypermail 2.1.8 : Fri 03 Mar 2006 - 03:34:07 EST