[R-sig-ME] Multilevel modeling-corpus linguistics

[Taha Omidian]

Greetings, 

I am a researcher in the field applied linguistics. My colleagues and I are currently working on a corpus based study. We have decided to use mixed effect modelling due to the hierarchical and unbalanced design of our data. However, we are having trouble specifying the best random effects structure that can reflect the nested design of our data. We would be very grateful if we could have your expert opinion on this. 

I have pasted our R script and data below. 

Thank you.

Kindest regards, 

Taha