Regression and Mixed Effects Modelling

In Person

This course will introduce you to regression and linear mixed-effects models (LMMs). It will help to develop your theoretical understanding and practical skills for running such models in R.

A regression is a statistical technique that relates a dependent variable to one or more independent variables. A regression model can show whether changes observed in the dependent variable are associated with changes in one or more of the explanatory variables.

Linear mixed-effects models are powerful and flexible statistical tools that help us understand the world. This is particularly useful in cases where we need to account for individual differences among people taking part in a study. For example, in language sciences, we might want to investigate how time spent on the internet per week influences people’s tendency to use internet slang in face-to-face communication, whilst acknowledging that this effect might differ from person to person. Linear mixed models allow us to do this and more!

This course includes three 2-hour taught sessions (Tuesdays 29/04, 06/05 and 13/05) and two 2-hour drop-in sessions to let you work on more exercises with tutors on hand to help (Fridays 09/05 and 16/05).

Session 1 is mainly conceptual. We will first recall linear regressions and discuss how linear mixed-effects regressions differ from basic regression and when/why we might want to use mixed-effects models.

Sessions 2 and 3 are more practical in that we will go through a real dataset from a sociolinguistic study to demonstrate how to run LMMs in R using the lme4 package (Bates, Mächler, Bolker, and Walker, 2015). You will also have the opportunity to try it out yourself by analysing a made-up dataset with extended exercises in the Friday drop-in sessions.

This is an advanced-level training course. It requires a basic understanding of R and statistical analyses. Some general knowledge of regression is not mandatory but will help you follow the content of this course. The lme4 package should be downloaded and installed prior to participating in this course.

Those who have registered to take part will receive an email with full details on how to get ready for the course.  

This course will be taught by Fang Yang and Aislinn Keogh.

After taking part in this event, you may decide that you need some further help in applying what you have learnt to your research. If so, you can book a Data Surgery meeting with one of our training fellows.

More details about Data Surgeries.

If you’re new to this training event format, or to CDCS training events in general, read more on what to expect from CDCS training. Here you will also find details of our cancellation and no-show policy, which applies to this event.

Learning Outcomes:

• Understand the concept of model structure (fixed and random effects) and how coefficients work (intercepts and slopes).

• Understand how to perform model fitting and selection and interpret the model assumptions.

• Know how to construct LMMs in R.

Knowledge refreshment pre-reading suggestions

• Introductory video on navigating RStudio
• Refresh statistical principles with the StatQuest videos
• Statistical Analysis Pathway of Learning

If you're interested in other training on data analysis, statistics, and machine learning, have a look at the following: 

• Digital Method of the Month: Statistical Methods - Linear Modelling 
• Introduction to Statistics and Descriptive Statistics
• A Gentle Introduction to Causal Inference
• Introduction to Bayesian Statistics
• Digital Method of the Month: Machine Learning
• An Introduction to Machine Learning
• Null Hypothesis Testing in R

Return to the Training Homepage to see other available events.