[R] Repeated measures with categorical data

Ben Bolker bbolker at gmail.com
Tue Nov 19 14:50:56 CET 2013

sualme <susana.alberichmesa <at> osakidetza.net> writes:

> Hello,
> I am working in a longitudinal study, with a categorical variable as a
> dependent variable (alcohol consumption: no use, use, abuse and dependence)
> with repeated measures (baseline, 1 year, 2 years). Besides I have another
> variable with two groups: control and experimental. 
> I would like to analyze the evolution in each group, I have though in lme,
> but I am not sure if I can do an lme with a categorical vble as dependent
> vble, or it must be continuous. If so, which model could I do? I have read
> something about lmer, but doesn`t give p-values..
> Thanks in advance!

  I'd take a look at the clmm function in the 'ordinal' package,
or possibly the MCMCglmm package, and consider posting future questions 
to r-sig-mixed-models at r-project.org .  (Your response variable is
probably considered 'ordinal', i.e. ordered categorical, which may
be helpful terminology in searching for solutions.)

  Ben Bolker

More information about the R-help mailing list