Course description
Overview
Mixed modelling is a modern and powerful data analysis tool for modelling clustered data, typically used for modelling data collected in trials where the levels of a factor are considered to be a random selection from a wider pool, or in the presence of a multi-level structure with different levels of variability. Such models offer potential benefits such as: the ability to cope with modelling complex data structures, greater generalisability of results, accommodation of missing values and the possibility of increasing the precision of treatment comparisons. In particular, mixed models have been extensively used to analyse repeated measurements where, for example, measurements taken over time in a clinical trial naturally cluster according to patient. In general, the course will focus on medical and health related applications of mixed modelling. Specific applications include multi-centre trials and cross-over trials in addition to the analysis of repeated measurements.
The course focuses on the linear mixed model, assuming normally distributed data, and on how to fit linear mixed models and interpret the results for a range of common medical and health related applications. Only essential theoretical aspects of mixed models will be summarised.
Examples used will be drawn from a variety of applications in medicine and health.
Practical work will be based around the statistical software R.
Duration
2 days
Delivery Mode
All training is online and will be delivered live each day between 09:00 and 17:30 (GMT+1). Delivery platform: Zoom, which may be freely accessed. Questions may be asked using Zoom's chat box. Note our online courses are delivered by a team of two presenters, meaning at least one presenter is always available to provide additional support. During presentations, the team member who is not speaking can take questions in addition to the presenter.
Who Should Attend?
Data analysts and statisticians working in medicine, health and related areas, who wish to have a practical introduction to linear mixed models. It will be assumed that participants are R users and are familiar with the practical use of linear models, covering regression models and ANOVA.
How You Will Benefit
The course will give you the skills to formulate, fit and interpret linear mixed models for a range of practical situations, as well as an appreciation of some of the benefits of mixed modelling.
What Do We Cover?
- Concept of fixed versus random effects
- Simple random effects and variance components models for modelling clustered data
- A summary of the important theoretical aspects of mixed models: maximum likelihood versus REML for fitting mixed models, estimating and testing fixed effects, degrees of freedom options and the Kenward-Roger method
- Model checking
- Multilevel modelling for hierarchical data structures
- Nested vs crossed random effects
- Multi-centre analyses
- Mixed models for cross-over designs
- Repeated measurements analysis: random coefficient models
- Practical experience: fitting models and interpreting R output
- Convergence issues
- lmerTest CRAN package, which extends lme4, for fitting mixed models; use of other CRAN packages including emmeans for summarising results from a mixed model.
Software
Practical work will be done in R.
Note: For practical work, participants must download and install a number of CRAN packages in R. This must be done prior to the start of the course.
Do you work at this organisation and want to update this page?
Is there out-of-date information about your organisation or courses published here? Fill out this form to get in touch with us.