Linear Mixed Models Workshop
A one-day workshop introducing linear mixed-effects models for psychological and behavioural data, taught in JAMOVI. Everything you need for the day — slides, worksheets, datasets, and the exercises — are below.
If you spot a typo or have a question during the day, just grab me. After the workshop, email me at alex.l.jones@swansea.ac.uk.
Morning · Concepts and theory
We start with the why and the what — going over the basics of the General Linear Model, why mixed models are useful and what issues they can solve, the differences between fixed and random effects, interpretation, and challenges in their use. The worksheet has some helpful points alongside the slides!
Lecture slides
PowerPoint · morning session.
Accompanying worksheet
Worked examples and notes to follow along with the lecture.
Afternoon · Practical
We move to fitting models in JAMOVI on a small, deliberately messy dataset (low n, missing values). We can work together on the data as we go through the slides.
Practical slides
Practical dataset
The dataset used in the practical session — low_n_missing.csv.
Your turn · Exercise
To consolidate what we’ve done, we’ll work through a new example which is about faces and reaction-times. The worksheet poses the questions; the CSV is the data to fit your models to in JAMOVI. Don’t peek at solutions until you’ve had a real go!
Exercise handout
Exercise dataset
exercise_data.csv.
Where to go next
If you want to keep going after the workshop, my free online course Statistical Modelling with Python has a longer treatment of mixed-effects models and the wider modelling toolkit (GLMs, factor analysis, clustering), all with runnable code in Python.