Master’s-level course in physical geography at the University of Marburg
Data analysis is an essential skill for professional geographers. Technological advances have made it easier to collect and distribute large datasets, increasing the need for both statistical and computing skills.
Working with data involves organizing, processing, describing, and interpreting datasets. This course uses R to build data-analysis workflows and automate common tasks with a flexible programming language.
Intended learning outcomes
By the end of this course, you should be able to:
- organize datasets and intermediate analysis results in a structured way,
- use Git for version control and GitHub to document workflows, collaborate, and manage issues and tasks,
- build data-analysis workflows in R using scripts and functions from third-party packages,
- model relationships among variables and estimate uncertainty in the results, and
- critically evaluate your analyses.
Setting
The course meets synchronously in a hybrid format, with in-person sessions in room F 14 | 00A19. Optional tutor sessions will also be offered regularly. Details will be shared at the first class meeting on Tuesday, 13 October 2026, at 9:15 a.m. (German time), in room F 14 | 00A19. Online participants can join through ILIAS; the virtual room link will be sent by email. Tutor sessions are voluntary.
Syllabus
The course consists of 13 sessions, running from 13 October 2026 to 9 February 2027.
| Session | Date | Topic | Content |
|---|---|---|---|
| Data basics | |||
| 01 | 13.10.2026 | First things first | Expectations; data and information; R, RStudio, R Markdown, GitHub, and Classroom 50 |
| 02 | 20.10.2026 | First things second | Data types and structures, indexing, logical operators, control structures, and simple functions |
| Data exploration | |||
| 03 | 27.10.2026 | Look at your data | Local files and function help, tabular data I/O, visual exploration, and descriptive statistics |
| 04 | 03.11.2026 | Clean your data | Missing values and source codes, text cleaning, reshaping, grouped summaries, and matching tables by keys |
| Data modeling | |||
| 05 | 10.11.2026 | Describe Your Linear Data | Linear regression, confidence intervals, hypothesis tests, and analysis of variance |
| 06 | 17.11.2026 | Predict your data | Cross-validation |
| 07 | 24.11.2026 | Select your variables | Multiple linear models and feature selection |
| 08 | 01.12.2026 | Tune Your Models | Generalized additive models |
| 09 | 08.12.2026 | Predict your temporal data | Autocorrelation, AR, and ARIMA models |
| – | 15.12.2026 | Built-in hold | No class; buffer date |
| – | Christmas break | – | |
| 10 | 12.01.2027 | Explain your temporal data | Decomposing time series |
| – | 19.01.2027 | Built-in hold | No class; use this date to prepare for the MOHA |
| Marburg Open Hackathon | |||
| 11 | 26.01.2027 | MOHA session | Marked assignment |
| Visualization | |||
| 12 | 02.02.2027 | Visualize your data | Data visualization overview, questions, and feedback |
| Wrap up | |||
| 13 | 09.02.2027 | Wrap up | Questions and feedback |
Deliverables
The graded course certificate is based on an individual portfolio in your personal GitHub repository. Each assignment describes the portfolio item and indicates whether it will be graded.
To create your repository, follow the Classroom 50 setup steps. Use the same repository for all assignments.
Preparation and prerequisites
The course assumes basic knowledge of and skills in R and geographic information science.
This course also uses additional material for teaching basic R skills, which can be found here.
Instructor
