[R] Mixed models fixed effects

[Emma Stone]

Hi Thierry,

Thanks again! You are a great help!!

I have taken habitat out, and then run it with style but still the same 
problem exists, so I have taken both style and habitat out. The problem 
here is it leaves me with only 3 parameters and because they are all 
correlated I cant use them in the same models, so I just have 3 single 
models (one for each parameter) - which isnt ideal.

G3Pgaps<-glmer(passes~pgaps+(1|Site),family=poisson)
C1Hvol<-glmer(passes~hvolume+(1|Site))
C2OHArea<-glmer(passes~oharea+(1|Site))

Also, when I run the C1 and C2 model, it wont work if I state family 
poisson.

Is there anyway I can get rid of the correlations in the parameters so that 
I can use them in the same model? Or is there another option to boost n, ie 
bootsrapping??

Thanks again

Emma

--On 13 March 2009 11:21 +0100 "ONKELINX, Thierry" 
<Thierry.ONKELINX at inbo.be> wrote:

> poisson



----------------------
Emma Stone
Postgraduate Researcher
Bat Ecology and Bioacoustics Lab
& Mammal Research Unit
School of Biological Sciences,
University of Bristol, Woodland Road,
Bristol, BS8 1UG
Email: emma.stone at bristol.ac.uk