Chapter 12. Probabilistic multifactorial grammar and lexicology
In this chapter you will learn how to model the speaker’s choice between two near synonymous words or constructions on the basis of contextual features. The most popular statistical tool that is used to create such models is logistic regression. The approach is illustrated by a case study of two Dutch causative auxiliaries. As in the case of linear regression, you will learn how to create, test and interpret a logistic model with the help of different R tools.