[R] Random effects aov

Christine Griffiths Christine.Griffiths at bristol.ac.uk
Wed May 20 15:41:08 CEST 2009

Dear All

I have a repeated measures design in which abundance was measured 
repeatedly over 10 months in three treatments (Tortoise A; Tortoise B and 
control) established in 6 blocks, i.e. crossed fixed effects. My original 
design incorporated two tortoises per treatment, however as fieldwork goes 
I ended up losing some animals. Rather than lose a couple of enclosures in 
the analysis and have to do a lmer, I thought I could include tortoise 
weight as an explanatory variable. For my treatments, tortoise weight in 
the control always = 0, while in general Tortoise A is twice as large as 
Tortoise B except when I lost animals. Is this the correct model?


Or should tortoise weight be nested in Treatment, i.e not included as a 
fixed factor but including the fact that tortoises species may have an 
effect? I am utterly confused now as to whether that should be the case as 
to some extent Tort.W and Treatment are correlated.
Any help would be much appreciated.
Many thanks

Christine Griffiths
School of Biological Sciences
University of Bristol
Woodland Road
Bristol BS8 1UG
Tel: 0117 9287593
Fax 0117 925 7374
Christine.Griffiths at bristol.ac.uk

More information about the R-help mailing list