Mixed-effects regression models have been a popular choice of statistical analysis in experimental psychology and linguistics research. While they’ve historically been developed and taught in R (namely, via the lme4 package), a newer implementation in the Julia programming language called MixedModels.jl has emerged and matured as a successor. The MixedModels.jl Julia library preserves much of the familiar interface from lme4, while offering many advancements in the complexity and speed of fitting mixed-effects models.
This workshop is a gentle, practical introduction to mixed-effects models with Julia that is low code and requires little set up. Emphasis is on using MixedModels.jl as a drop-in replacement for lme4 while minimizing friction with existing R-centered data analysis workflows. Participants should be familiar with R and Rstudio and have prior experience with using (mixed effects) regression models.