[R] Random Effects for One-Way Anova

[Prof Brian Ripley]

On Wed, 30 Nov 2005 Giuseppe.Palermo at bo.infn.it wrote:

> Hello to All.
> I'd want to use a one-way ANOVA. This means that I have only one factor, with,
> lets say, 5 levels.
> I made a dataframe, called "DATA", with two Columns:
> A, that is my response, and it is "class numerical".

There is no class 'numerical': I presume you mean 'numeric'.

> B, that defines the different levels of my factor, and it is "class factor".
>
> If I want to use a fixed effect model,
> I know that the formula I have to use is:
> lm.1 <- lm(A~B,data=DATA)
> anova(lm.1)
>
> My questions is:
> which formula should I use if I want to use a random effects model?
> I think I should use "lme", but I don't know how.

aov() is the function for anova modelling. so you want

aov(A ~ B, data = DATA)         # fixed effects
aov(A ~ Error(B), data = DATA)  # random effects

but a 1-way ANOVA with fixed or random effects is the same analysis, and 
only the interpretation differs.

-- 
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