Skip to content

lmer, p-values and all that

2 messages · Dan Bebber, Simon Blomberg

#
Hello,
I've just located the illuminating explanation by Douglas Bates on degrees 
of freedom in mixed models.
The take-home message appears to be: don't trust the p-values from lme.
Questions:
Should I give up hypothesis testing for fixed effects terms in mixed models?
Has my time spent reading Pinheiro & Bates been in vain?
Is there a publication on this issue?

Thanks,
Dan Bebber

Department of Plant Sciences
University of Oxford
#
Try using mcmcsamp() to sample from the posterior distribution of the 
parameter estimates. You can calculate a p-value from that, if that is 
your desire. Instructions are in the R wiki:  
http://wiki.r-project.org/rwiki/doku.php?id=guides:lmer-tests

HTH,

Simon.
Dan Bebber wrote: