What: PhD Data visualization workshop
When: 2023-11-20 @ 13.30 - 17.00
Where: S8
Who: Matti Vuorre (instructor), y'all
Why: What makes a good graph? In this workshop, we try to sketch an answer to this question, and then use what we learned in making our own figures that intrigue, inform, and impress.
- Slides: https://drive.google.com/drive/folders/1pmX2hldtLY7S69b_S2C6iFq0Syw1mRbo
- Shared notebook: https://docs.google.com/document/d/1rnW8eEtkrUAJOop_PZr5udigzpKddlk3x5C26xvuZxc
- Grab a copy of the code and play with it in the cloud: https://posit.cloud/content/7071123
- Download slides from GitHub
We will talk about theoretical approaches to data communication and how to apply them in practice.
- A round of greetings
- Tell everyone who you are and why you are here
- Group exercise 1
- Please form small groups (2-4) and take a few minutes each to tell other group members what kind of quantitative information you would like to display, how, and why.
- If possible, share existing or work-in-progress figures with your group.
- Lecture 1
- Theories / concepts underlying effective data visualization.
- Group exercise 2
- Do you find the concepts discussed useful for your own visualization goals? How? If not, why not? Discuss in your group.
- Lecture 2
- Introduction to the grammar of graphics and its implementation in R
- Group exercise 3
- Work on your figures and discuss them with others in your group.
- Wrap-up
Image sources and credits are in slide notes.
- The Royal Statistical Society's "Best Practices in Data Visualization" guide: Repository & website.
- European Commission's data visualization guide.
- From data to viz, "a classification of chart types based on input data format".
-
https://r4ds.hadley.nz/data-visualize.html (Chapters 1 & 9-11)
-
Cédric Scherer's ggplot2-focused workshop: Repository & website.