[R] Understanding the output from lme anova

Hedberg Peter phedberg at biol.uw.edu.pl
Mon Dec 6 12:50:09 CET 2010


I have a question regarding understanding the output from a nested linnear mixed effect model.
My model looks like the following
lme(Poaceae~Site+Ditch_block+Tree_cut+as.factor(Dist_start)+(as.factor(Dist_start)*Tree_cut)+((as.factor(Dist_ditch))/Ditch_block),random=~1|Mainplot/Transect/Obsplot)
I analyzed the outcome in lme.anova and was puzzled by the output.
The question I have is why the output from my lme.anova for the nested term as.factor(Dist_ditch)/Ditch_block is represented the same way as the result for the two-way interaction of as.factor(Dist_start)*Tree_cut.
That is both are in the output presented with a : sign even though one is a * interaction and the other one a / interaction.
Tree_cut:as.factor(Dist_start)
Ditch_block:as.factor(Dist_ditch)
Does the model interpret my / interaction as a * interaction, or is it simply so that the use of : in the output for both just denotes an interaction, and the type of interaction is the one I put into the model.
Thankful for any help with this matter.
Best regards,
Petter 

___________________________________
NOCC, http://nocc.sourceforge.net




-- 
This email was Anti Virus checked by Astaro Security Gateway. http://www.astaro.com



More information about the R-help mailing list