[R-sig-ME] Variance structure in mixed models

[v_coudrain at voila.fr]

Yes I am interested in Pearson residuals. I had a look at MASS::glmmPQL, but I could not find the way to specify my variance structure. I thought about a second 
possibility: I use the function lme, which allowes using varIdent, and I accounted for overdispersion using an individual random effect. Do you think, this is a suitable 
method? I still wonder if fitting a variance structure really improves my model, because the residuals vs fitted plot looks worse. 

Valérie
___________________________________________________________
Quiz TV : Vous êtes fan de la série "Friends" ? 5 questions ici http://tv.voila.fr/quiz/quiz-special-friends_14538959.html