How to use statistical tools to develop causal understanding of systems? How to shape statistical models to the requirements of a research question? What can be learned from comparing models? This course aims to provide some interesting insights about these questions.
In the lecture part (2h per week), you will learn about how to use statistical tools to answer research questions, how probability can be understood using Bayesian statistics, how to design and fit regression models - starting from simple models with one predictor, continuing with generalized linear models, which do not assume normal distributed residuals, and multilevel models to pool information across instances - and how to consider causal knowledge about processes in statistical modeling.
The main goal is to give insights and intuition about how to practically perform statistical analyses. Therefore computer code in R and examples are emphasized over mathematical formula, without neglecting the underlying mathematical intuition. In the exercise part (1h per week, assisted self (or group-)working on exercises) you can get own practical experience (and credit points for the course) via weekly exercises.
The course is intended primarily for master students who already attended an introductory course in statistics, have basic R knowledge, and want to get additional insights in how to apply statistics.
Lecture materials will be provided via learnweb.
- Lehrende/r: Henning Teickner