[R-sig-ME] Question on fitting linear mixed model

[Jorge Enrique Avendaño]

 Hi all,

I am new with mixed models and would like to ask your advice respect to a
model I am trying to implement in lme4.

Brifely, I want to test the effect of 5 plumage treatments in the
aggression response of 13 birds (territorial males of the same species
selected randomly from a population).

All 13 males were tested against my five plumage treatments. So, I have
only one observation (response) of every individual per treatment. So, I
tried to run the following model:
model = lmer(Aggresion ~ Treatment + (1|Individual_ID), data=data)

Then, I obtained this message: Error: number of levels of each grouping
factor must be < number of observations

One colleague told me that this model does not work because I only have one
observation/response of every individual per treatment, whereas the number
of treatments are five. Moreover, my model should not be nested because
individuals were tested across all five treatments, instead of being
distributed or restricted to particular treatments.

I would like to include in my model the Individual ID because some
individuals were more aggresive than others, and more importantly I want to
know if one plumage treatment elicited more aggression than others. I
understand that Individual ID is a random factor. However, I don't know
what kind of mixed model I am facing.

I would be very glad if you could help me with this issue.
My best,
Jorge


-- 
Jorge Enrique Avendaño., M.Sc.
Estudiante de Doctorado
Laboratorio de Biología Evolutiva de Vertebrados
Departamento de Ciencias Biológicas
Universidad de los Andes, Bogotá, Colombia.
Tel: +571 3394949 ext. 3755

	[[alternative HTML version deleted]]