Traditionally, linear models (linear regression and ANOVA) assume all observations are independent. In this course you will learn how to analyze more complex datasets
- individuals measured repeatedly over time
- individuals nested within another observational unit (e.g. students in a school, patients in a hospital, animals in a litter)
This course covers the basics of linear mixed models (aka multilevel modeling, longitudinal data-analysis, repeated measures analysis, analysis of panel data), to deal with the research questions of linear regression and ANOVA, but accounting for dependencies within the data. The course is offered live and consists of a starters package, followed by 2 in-person sessions. For the self-study part of the course, you need to have the software R or JMP Pro installed on your PC/Laptop (see installation instructions). The starters package should be completed before the live session.
For this course, we offer the possibility to take an exam. For the PhD students in the faculties IOB and Applied Economics, this is a requirement to obtain a credits for these courses, but people from other faculties are allowed as well. If you are interested in taking the exam, check the wants-to-take-exam-box in the registration form. Participating in the exam costs 10€, which is deduced automatically from your educational credit.
Course contents
At the end of this course you are able to recognize a dataset / research question that requires analysis through a linear mixed model. You can fit and evaluate a linear mixed effect, inserting the correct set of random effects (random intercept and random slope) depending on the type of data. You can draw the correct conclusions from the JMP output, regarding effect sizes and significance. You are aware of the assumptions underlying the model and the robustness against missing values.
Theory and exercises will be taught using the statistical software packages JMP Pro.
For R users, The R-code will be made available and briefly discussed after each session.
Target audience / prerequisites
Doctoral and post-doctoral researchers who are already familiar with elementary statistical data analysis techniques.
Linear regression is assumed to be known.
No previous experience with JMP/R is required.
Instructors
Prices
PhD student UAntwerpen : € 50
UA-affiliated : € 90
Academic non-UA : € 160
Publicand non-profit sector : € 250
Private sector : € 500
Course structure and time schedule
This 2-day course will take place in March 2024, in Campus Groenenborger (TBC).
March 19 from 10.00 - 16.00, 2024 room G.Z.446
March 20 from 10.00 - 16.00, 2024 room G.Z.422
All rooms are equipped with a desktop PC. There is no need to bring your own laptop.