# Repeated measures ANOVA, unbalanced and missing values in R

I have chemical compositions of individuals collected in 6 different sites in monthly intervals over a period of 6 months.

I would like to determine whether there is significant variation among sites as well as among months within sites. I was advised to use a repeated measures ANOVA to address this. However my design is unbalanced and has missing values. So, for example not all the sites are represented at all times and the number of individuals per site and time also varies.

Is it possible to do a repeated measures ANOVA with this dataset? Should I use the lmer function from the lme4 package?

• I think you should use good old lme from package nlme, which also allows to model temporal autocorrelation of residuals. 6 time points make a short time series but you are still dealing with time series. – Roland Sep 19 '12 at 14:41