Roland Schäfer. 2018 (maybe). Generalised Linear Mixed Models. In: Stefan Gries & Magali Paquot. Practical Handbook of Corpus Linguistics. De Gruyter. Read more for a link to a better version.

Abstract Mixed effects modeling – alternatively called hierarchical or multilevel modeling is a straightforward extension of (generalized) linear modeling as discussed in the previous chapter. A common characterization of mixed-effects modeling is that it accounts for situations where observations are clustered or come in groups. In corpus linguistics, there could be clusters of observations defined by individual speakers, registers, genres, modes, lemmas, etc. Instead of estimating coeffcients for each level of such a grouping factor (so-called fixed effects), in a mixed model they can be modeled as a normally distributed random variable (a so-called random effect) with predictions being made for each group. This chapter introduces readers to the situations where mixed effects modeling is useful or necessary. Thee proper specification of models is discussed, as well  as some model diagnostics and ways of interpreting the output. Readers are assumed to be familiar with the concepts covered in the previous chapter on (Generalized) Linear Models.

Comment My first draft of the chapter was evaluated as being of very high quality by two anonymous reviewers for the handbook. However, in two rounds of comments, the editors insisted on a huge number of stylistic changes and changes to the way the methods are presented despite the fact that I made it clear from the beginning that I was planning to follow Gelman & Hill (2007). This has alienated me from my own text, and I cannot recommend the published version of my chapter as an introduction to multilevel/hierarchical models.

Click the following link to download the first draft, which represents my views on GLMMs and hierarchical/multilevel models much better than the published text: Roland Schäfer. 2018. Generalised Linear Mixed Models. Original draft of Chapter 22 in: Stefan Gries & Magali Paquot. Practical Handbook of Corpus Linguistics. De Gruyter. An extended version will be part of my own text book on statistical inference for linguists.

So far, I have not given away my copyright to Springer. Let’s see whether anybody even notices that I didn’t sign and return the form.