[R-sig-ME] Binary data - zero random effects or not depending on a small change in the level 2 groups

[Federico Calboli]

On 11 Jun 2009, at 13:14, Fiona Mary Underwood wrote:

> <snip>
>
>
> I get very different results for my random effects depending on  
> whether
> I use season or year.


Unfortunately I cannot help much, but, does the result change if year  
is coded as factor or as a continuous variable? what about season?

Best,

Federico

--
Federico C. F. Calboli
Department of Epidemiology and Public Health
Imperial College, St. Mary's Campus
Norfolk Place, London W2 1PG

Tel +44 (0)20 75941602   Fax +44 (0)20 75943193

f.calboli [.a.t] imperial.ac.uk
f.calboli [.a.t] gmail.com