Adapted from my previous, unanswered question:

I'm analyzing the results of a hormone manipulation experiment. I measured a number of variables at three times in three groups. The groups are different sizes and not all individuals were measured every time, so I'd like to use GLMM rather than a repeated-measures ANOVA. I created the model then tested the significance of the terms (time, treatment, and time x treatment) with ANOVA.

I'm quite new to GLMM, but after doing the tests, further reading suggests that my approach may be inappropriate, particularly with small data sets (I have ~seven animals per group). One reason given against using this method is that it seems that there is disagreement about what the degrees of freedom should be.

Is this method of testing the significance of factors in the model appropriate? I'm aware that the place for me to go is probably the Pinheiro and Bates book, but I don't have access to it at this time. Thanks in advance for any advice.




#and now the GLMMs on each variable - I'll show just one here

var1.glmm<-lme(var1~var.time + var.treatment + var.time*var.treatment, data=datums, random = ~1| id)



No, that isn't really it. One side point is you should consider looking into multi-modal inference a la Anderson and Burnham. However, there are plenty of resources, many of which also refer to the P+B book; such as this one, with code: http://glmm.wikidot.com/basic-glmm-simulation

Also, try investigating these calls:

fit.sim<-sim(fit, n=1000)
coef.sim <- coef(fit.sim)
fixef.sim <- fixef(fit.sim)
ranef.sim <- ranef(fit.sim)
sigma.sim <- sigma.hat(fit.sim)
dotplot(ranef(gm1, postVar=TRUE))

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.