R statistics Mixed ANOVA but outcome Poisson distributed - which model in R?

Question

I have a mixed ANOVA with one between and one within factor:

between factor: control versus treament

within factor: ad1, ad2 (measures click rates on 2 ads)

aov(repdat~(within*between)+Error(subjcts/(withincontrasts))+(between),data.rep)

However my outcome repdat has a Poisson distribution and 0s (since sometimes users just dont click). How do I model this in R?

I considered:

1.) using Zelig package with model Poisson, but I dont know how to properly write the mixed model formula here. Is it the same as for the mixed ANOVA?

2.) using lme4 package with model Poisson. Can I use both a between and within factor or even treat "ads" as a random variable?

It would be great if someone could quickly show me example code for both 1. and 2.

thanks

Answer 1

You can use the lme4 package for this I'm pretty sure. However, I don't understand your aov() example well enough to try to replicate your situation.