Keywords

workshop, Sociolinguistic research, marginal models

Abstract

Sociolinguistic studies generally gather multiple instances from a single speaker. This introduces repeated measures that need to be taken into account when carrying out statistical analysis in order for the outcome to be statistically valid. One way of doing this is to apply a mixed-effects model. In the workshop, I present a lesser-known, but equally valid method for dealing with repeated measures—marginal models.

Original Publication Citation

2016. “Marginal Models: An Alternative to Mixed-effects Models in Sociolinguistic Research.” Workshop given at New Ways of Analyzing Variation, Vancouver, CA