Seminar: Multilevel-Models

General overview

This seminar provides basic knowledge about linear and generalized linear multilevel models. Approaches from the frequentist and Bayesian paradigms are discussed. Course objectives include building a conceptual understanding of the various application areas, mathematical models, and the general analysis workflow of hierarchical modelling. Concerning the implementational aspects, we will use the programming language R (particularly the R-packages lme4 and brms).

The seminar takes place as a block seminar at the end of the lecture period. The examination format consists of a presentation and the preparation of a report (corresponding to the presentation).

The course language is English. All up-to-date information will be presented on this website.

Grading

The course evaluation includes: (1) a 45-minute presentation and a 15-minute discussion, as well as (2) a written report of 7-9 pages. Both the presentation and the report are conducted in pairs. Each group member’s presentation is individually graded, while the report is evaluated as a group project.

More information about grading criteria will follow.

Prerequisits

It is advantageous if participants have a basic understanding of (generalized) linear models and Bayesian statistics. Furthermore, we will be using the programming language R, thus, basic programming skills in R are expected.