Master of International Affairs   Master of Data Science for Public Policy   Master of Public Policy  

Data visualisation with R

Data has become ubiquitous and plays an ever-increasing role both in research and policymaking. However, the real potential of data can only be unlocked if we can effectively communicate the insights we drew from analysing the data. This can be particularly challenging when working with interdisciplinary teams or with policymakers, as data literacy is oftentimes low.

To overcome these challenges, data visualisation is a powerful tool as humans are much better at processing and retaining information that is presented visually. This course gives you an introduction to data visualisation with R. The objective of this course is to learn how visualisation can be used to go through best practices for visualising different types of data.

The course consists of a brief theoretical introduction:

  • Why use visualisation? How do humans process visual information compared to numbers and text?
  • Which types of graphs exist? Which ones are suitable to visualize different types of data?
  • The dos and don'ts of data visualisation.

Then, we will delve into the practical part. You will get familiarised with the following fields of data visualisation:

  • Visualising data sets for descriptive statistics (univariate and bivariate figures such as bar charts, pie charts, scatter plots, boxplots, violin plots, heat maps, grids of plots, ridgeline plots, etc.)
  • Visualising regression outputs (point estimates, confidence intervals, comparing different models estimates visually)
  • Geographic visualisations: Choropleth maps (e.g. colouring German counties by population density) or distributional maps (e.g. displaying cities with the most bikes lanes)
  • Displaying networks (e.g. retweet networks between politicians)
  • Using Rshiny to create interactive apps that allow readers to shape the data visualisation according to their needs. This is a powerful tool to allow readers to experience the data themselves.

Along with the technical side of how to implement these graphs in R, we will always discuss how different graphical displays are perceived by the audience and which design choices are paramount for focusing the audience’s attention on what you want to communicate.

The whole course is very hands-on and there will be a lot of time dedicated for students to try it out themselves.
Upon completing this module, you will be able to:

  • Understand when data visualisation is beneficial
  • Understand how your audience perceives your visualisation
  • Choose an appropriate visualisation to communicate effectively with your audience
  • Critically evaluate visualisations and know how to improve them
  • Tell data stories with visualisations