Linear mixed models in R
25/03/2024
26/03/2024
General context
This course is the perfect follow-up to the Basic statistics in R course for people who are doing experiments with multiple grouping variables, dependent groups and/or mixed designs. Where the Basic statistics in R course ends with a brief description of the use of lm() and lme() for comparing groups, these topics form the starting point for this course.
Linear mixed models have rapidly gained popularity in the last 20 years. They became the most used tool in research, particularly in life sciences. Yet, they are often not well understood and misused.
This short course will provide the background of the models but will be as applied as possible with examples and R-code.
The items that we foresee to cover:
- The linear model: its virtues and its limits
- Observations in live science studies are seldom independent
* Case studies to illustrate these dependencies - Mixed models are developed to handle this structure and dependencies
- How can we deal with these dependencies and put it to our advantage
- When things are not normal
Linear models assume your errors are normal. Often, they are not.
What are the common cases in biology where this occurs?
What is the problem when we ignore this issue and proceed with our standard tools?
What are the proper tools to deal with this issue? - How do we solve (generalized) mixed models in R?
* Like frequentists : nlme, lme4
* Like Bayesians: brms, Rinla - How to interpret the output of a mixed model, statistically and practically
Those who have little or no background in statistics are encouraged to follow Statistical thinking and those who work with plants also design and analysis of plant experiments.
If you have no experience with R, you should follow the Basic statistics in R training first.
Trainers
Joris De Wolf
Joris De Wolf has a long track record in biostatistics. He worked at Crop-Design, later BASF, as team leader of the biostatistics group. He was involved in the experimental design and start-up of the high-throughput phenotyping system TraitMill, design of the databases and streamlining of statistical analyses. Furthermore, he was responsible for the field testing pipeline and statistics on transgenic yield improvement. In 2016, he joined GSK vaccines. Currently he works full time as a biostatistics consultant.
Practical info
25 March 2024 - 26 March 2024
Ghent - Clemenspoort
Overwale 3
9000 Ghent
Belgium
25 March 2024 - 26 March 2024
Ghent - Clemenspoort
The train station Ghent St-Pieters is only 250 m away. We have 40 bike lots.
25 March 2024 - 26 March 2024
Ghent - Clemenspoort
We offer parking lots (1.20 € per hour). GPS address: Sint-Denijslaan 251.
Please take the neem the exit SPORTHAL HoGent (not Campus HoGent) and drive all the way to the end, so past the parking slots of HoGent.
25 March 2024 - 26 March 2024
Ghent - Clemenspoort
+32 476 46 10 40