[R-sig-ME] p-value mixed model family poisson

[Ben Bolker]

magali proffit <mafgali at ...> writes:

>  I have an experiment where I want to test the effect of two
> treatment on the oviposition of an insect. Each insect individually
> in a box has been subject to 1 of the two treatment and we counted
> the number of eggs lay per insect. The replicates of experiment have
> been done on different days.  I think it´s easy to test: I need to
> do a mixed model with number of eggs layed per female as an response
> variable, treatment fixe factor, day random factor, with
> family=poisson.  I read a lot of things about p-value in mixed
> models and I´m completely confused ... Can anyone tell me how
> should I do to know here if there is an effect of the treatment on
> the number of eggs layed per female?  Thanks a lot!

  You haven't given us very much to go on.  How many days
(relevant to whether you will be able to estimate a random effect
reliably)?  How many insects?

  You can estimate the size of the treatment efect with the
model suggested above.

  You can use anova() to compare models with and without
treatment effects to get likelihood ratio test statistics.
However, LRTs assume large data sets.  You may not have a large
data set.

  You can use parametric bootstrapping: see the examples
under help("simulate-mer").

  If your data are not too extreme (i.e. not too many zeros),
you can see if square-root transforming followed by classical
mixed model ANOVA gives you a reasonable answer.