[R-sig-ME] Need some help with glmer output

[Hans Ekbrand]

On Sat, Feb 18, 2012 at 09:13:04AM -0800, lopez toledo wrote:
> Dears lme4 users:
> I'm doing a repeated measurement analysis, which seems pretty obvious to 
> me, but I'm getting some strange results, which I do not 
> understand. Hope you can help me!! Thanks in advance!
> 
> I'm evaluating the effects of Defoliation Treatments (4 levels), Gender 
> (Male/Female) and Time on several responses (Growth , leaf production, 
> inflorescence production, etc) of a palm dioecious species 
> 
> I have about 550 palms total which have been measured 3 times. My main question is whether the effects depends on the intensity of defoliation, palm gender and how this has changed through time. I am 
> considering the effects of repeated measurement as Palm/Time
> I am exploring with the model below, which include as fixed factors DT*G*Time and as 
> random factor Palm/Time to consider the effect of repeated measurements. 
> 
> In this case, I'm using glmer, 'cos I've have counts as response variables (number of leaves, number of inflorescences, etc).
> 
> Model1<-glmer(Total leaves ~ DT * G * Time + (1 | palm/Time), family=poisson)
> 
> 
> I've got two questions:
> 1) Am 
> I considering the random effects correctly or not? as when I run the model there is the following message
> 
> 
> "Number of levels of a grouping factor for the random effects
> is *equal* to n, the number of observations"

Is it not enough to define one simple random effect for palm? Wouldn't
that take care of the problem of repeated measurements of the same
palm?

Try:

Model1<-glmer(Total leaves ~ DT * G * Time + (1 | palm), family=poisson)