*We are longer accepting applications for this course*


Richard Traunmüller is a visiting professor of quantitative methods in the social and behavioural sciences the University of Mannheim and an assistant professor of empirical democracy research at Goethe University Frankfurt. He has previously held positions at the University of Konstanz, the University of Berne, and the University of Essex. Traunmüller has taught semester long courses on data visualisation at these universities and has been invited to teach statistical visualisation at the European University Institute (EUI) in Florence, the German Institute of Global and Area Studies (GIGA) as well as the Essex Summer School in Social Science Data Analysis. His work has been published in British Journal of Political Science, Comparative Political Studies, European Journal of Political Research, Political Analysis, and Sociological Methods & Research, amongst others.

Course Content
Data visualisation is one of the most powerful tools to explore, understand and communicate patterns in quantitative information. At the same time, good data visualisation is a surprisingly difficult task and demands three quite different skills: substantive knowledge, statistical skill, and artistic sense. The course is intended to introduce participants to a) key principles of graphical perception and analytic design, b) useful visualisation techniques for the exploration and presentation of univariate, multivariate, time series and geographic data and c) new developments of data visualisation for the social sciences, such as interactive data visualization, visual inference, and visualising statistical models.

Course Objectives
This course is highly applied in nature and emphasizes the practical aspects of data visualisation in the social sciences. Students will learn how to evaluate data visualizations based on principles of analytic design, how to construct compelling visualisations using the free statistics software R, and how to explore and present their data and models with visual methods. In short, students will get hands-on experience producing modern visualisations for their practical problems. This will be especially helpful to those students with own datasets related to their research.

Course prerequisites are a basic understanding of statistics and bivariate linear regression. Some experience in the use of a statistical software package would help but no prior exposure to R is required. I will provide detailed code examples and spend much time in the lab to get students up to pace.

Core Reading
Few, Stephen (2009). Now you see it: Simple Visualization Techniques for Quantitative Analysis. Analytics Press.
Tufte, Edward (2001). The Visual Display of Quantitative Information. (Second Edition). Graphics Press.
Murrell, Paul (2011). R Graphics. (Second Edition). Chapman & Hall/CRC.
Unwin, Anthony (2015). Graphical Data Analysis with R. CRC Press.

Background knowledge required
OLS = e
Maximum Likelihood = e

e = elementary, m = moderate, s = strong

Part I: Foundations and Principles of Data Visualisation

Day 1: Intro
Statistical Graphs vs. InfoVis
A Brief History of Visualisation
Graphs vs. Tables
Basic Graphs & Basic R

Day 2: Data Visualisation as a Methodology
Exploration vs. Presentation
Comparison, Comparison
Component Graph Design in R

Day 3: Graphical Perception
Attributes of Preattentive Processing
Gestalt Principles of Visual Perception
General Graph Design

Part II: Tools and Methods of Data Visualisation

Day 4: Visualisation of Bivariate and Time Series Data
Enhancing Scatterplots
Overplotting Reduction
loess: Scatterplot Smoothers
Slope Graphs

Day 5: Visualisation of Multivariate Data
Small Multiples
Mosaic Plots
Parallel Coordinate Plots

Day 6: Visualisation of Geographical Data
Map Variations (Choropleth, Dot, Flow)
Linked Micro Maps

Part III: New Developments and Extensions of Data Visualization

Day 7: Interactive Data Visualisation
Selection and Identification
Zooming and Filtering
Highlighting and Linking

Day 8: Visual Inference
Is what we see really “there”?
Simulation Inference
The Line-up Protocol

Day 9: Visualisation of Statistical Models
Graphs instead of Regression Tables
Visualising Quantities of Interest
Visualising Inferential Uncertainty
Visual Diagnostics & Model Checking

Day 10: Student Visualisation Projects