Visualizing data
Philosophy and psychology of good plots
2016-07-04 — 2025-06-07
A grab-bag of links about making data visually comprehensible. For now, this is mostly a list of people I want to link to, but I might also share some insight into good communicative graphic design if, heaven forfend, I end up needing to pretend to be a graphic designer in some dire strait. A weird corner of this discipline is the specialization, data dashboards, which appeals to executives and thus has money in it, so it has its own notebook.
1 Membership, sets
You know about Venn diagrams, right? They are very hard to scale to many sets. As far as I knew the best option to plot them was vqf/nVennR which implemented the nVenn algorithm (Pérez-Silva, Araujo-Voces, and Quesada 2018) in R. Not only is that package discontinued, the Venn diagram is just not that useful for more than seven sets. I think ggVennDiagram (Gao, Yu, and Cai 2021) still works if you like Venn.
However! there exists a better option, the UpSet algorithm (Lex et al. 2014). I use UpSetPlot.
2 Resources
- cxli233/FriendsDontLetFriends: Friends don’t let friends make certain types of data visualization—What are they and why are they bad.
- Data cuisine visualises statistics via Michelin-star-lookin’ food
- A tour through the visualization zoo
- rawgraphs unifies a lot of hip ideas about graphing in a web context.
3 Practitioners
- Mitchell Whitelaw is a practitioner and also a critic of this area.
- Small multiples is a fun agency in Sydney doing visualisations of intriguing data sets.
- Miriam Quick
- Omid Kashan
- David McCandless founder of …
- Information is Beautiful is a classic blog of datavis projects
- FlowingData likewise blogs tasty links in this area.
- The works of data visualization granddaddy, French engineer Charles-Joseph Minard, are online.
- Average Happiness | Short film by Maja Gehrig