Clever Legends
Hi all,
I’ve been thinking about how to make my chart legends more engaging. I prefer to directly label my data rather than using legends that are separated from the data, but when I need a legend—say, on a map or a denser bar chart or scatterplot—are there ways to really attract the reader’s attention?
“Clever legends” are those that take an interesting approach to defining the data elements in the chart. Instead of a standard set of squares or circles or lines with labels, ‘clever legends’ use a unique charting device or design to engage us while also simply providing us with the data definitions. Perhaps my favorite clever legend is this famous bubble chart from Hans Rosling. Notice the map-as-legend in the bottom-right part of the image where color shows each region of the world, which is used in the bubble chart itself.
It’s worth noting that I’m distinguishing these kinds of clever legends from toehr legend types that may also be useful and engaging, but, say, use icons or shapes in the actual graph. As an example, the legend in this graph—from a really nice scrollytelling piece by Karim Douieb—defines the values of each of the icons.
I also don’t mean legends that have a smaller version of the graphic that are used to directly explain the visualization. For example, this tile grid map from the US Geological Survey has an area chart of streamflow for every state in the country in July 2021. The legend over to the side is a larger version of each tile with a legend built in. A good legend to be sure, but not the class of ‘clever legends’ I have in mind.
The legend in this bivariate choropleth map from FiveThirtyEight is not necessarily clever in the way I’m defining it, but does a great job explaining how to read the map. The three small multiple versions at the top help you understand each dimension of the map and is then paired with full legend below the menu of state names.
Here’s a good, simple example of a clever legend. Will Chase from Axios created this tile grid map where the houses are sized by the share of renters who are not current on their rent. In addition to using color in the legend, this legend has a little house icon, scaled to the value. It’s a little play on the icon-as-legend technique we saw earlier.
I’m a big fan of using bar-charts-as-legends in maps. This is a good example from my colleagues at the Urban Institute where the four groups shown in the map are presented as a bar chart in the legend. We don’t even have to examine the map to see that a majority of counties fall in the 10.1% to 20% bucket.
Finally, the little gauge-as-legend in this cool piece from Possible. on noise in the New York City area also strikes me as clever. Here, because we often see measures of sound on a radial gauge-type visual, it works really well here. (Side note: this piece was inspired by work from Karim Douieb showing noise in Brussels. I don’t know Karim and it’s completely coincidental that two of their pieces are showing up in this post—well, maybe not totally coincidental as I’m getting a sense they try to play with clever legends. Kudos, Karim.)
I don’t have a ton of examples of clever legends, which is probably whey I think they’re clever. I’m on the lookout for more and to consider ways that I might incorporate these kinds of things into my own work in the future. If you see an example of a clever legend, be sure to let me know!
Take care and thanks,
Jon
If you want to see my collection of data visualizations, please check out my Data Visualization Catalog. It’s free to explore online and you can download a .zip file of the entire collection for free as well.
PolicyViz at Home 🏡
After my last newsletter, I chatted with a few people who had their own stories about Edward Tufte or had questions about this and that. I was glad to catch up because it had been a while since talking last. Turns out that they like the newsletter so they can see what’s going on with me. So, I’m adding a little personal update to the newsletter going forward—a little peak into my personal life and things going on. Here goes.
Umpiring. We’re getting into spring here in the Washington, DC area, which means baseball is starting up. I umpire for my kids’ (former) little league and did my first game last weekend. I was asked/told to do the highest level—this is 11- and 12-year olds—which I generally don’t do because, well, the parents and coaches get a little amped up and I don’t need to be yelled at. I only ump because it’s fun and gets me out of the house. Well, first game actually went smoothly. Best part? I helped kick off one of the coaches 56th season! Unbelievable. He’s pretty much a legend here and I’m guessing 90% of the coaches involved in the league aren’t even 56 years old.
Conferences. I’m excited about a few upcoming conferences. The big one is the Tableau Customer Conference in Las Vegas in May. I haven’t been to TC since 2013 when it was in Washington, DC and with my goal of learning more about the tool, I’m excited to sit down and really try to up my Tableau game. Some folks on Twitter were annoyed with me that I don’t want to abide by the speaker rules and end my presentation with a “Thank You!” slide, but I’m holding fast here. See some of you in Vegas!
PolicyViz Podcast with Jen Christiansen
Jen Christiansen is author of Building Science Graphics: An Illustrated Guide to Communicating Science through Diagrams and Visualizations (CRC Press) and senior graphics editor at Scientific American, where she art directs and produces illustrated explanatory diagrams and data visualizations. We talk about her new book and the challenges of bringing academic work to life.
You may see some changes in the podcast in the next few weeks. I’m working on changing my hosting service for the podcast and am hoping to add some more sponsors. Of course, you can help me out by rating or reviewing the show on iTunes, Stitcher, Spotify, or your favorite provider—this can help me bring more big names onto the show—or even donating using PayPal or Patreon.
Things I’m Reading and Watching
Books
I’m terribly behind on reading books right now. I do have I, Robot next to me, which I hoped my daughter might read, but looks she’s not interested so I’ll hopefully re-read it.
Articles
Do No Harm Guide: Applying Equity Awareness In Data Privacy Methods from Claire Bowen and Joshua Snoke
Chart Reader: Accessible Visualization Experiences Designed with Screen Reader Users by Thompson et al.
Declutter and Focus: Empirically Evaluating Design Guidelines for
Effective Data Communication by Ajani et al.
Observing Race and Ethnicity through a New Lens by Gonzalez et al.
Network Theory, Plot Analysis by Franco Moretti
Blog Posts & Twitter Threads
Modeling the Effects of Land-Use Policy on Housing Construction in the Washington, DC, Region by Yonah Freemark
Breaking down ridiculousness in food security (me):
Let's break down this ridiculousness. 1. Nearly 436,000 people in Minnesota received SNAP (Supplemental Nutrition Assistance Program) benefits in FY 2022. That represents 1 out of 13 people in Minnesota (or 8%). Source: cbpp.org/sites/default/…Minnesota Republican state Sen. Steve Drazkowski on bill providing free school breakfast and lunch: "I have yet to meet a person in Minnesota that is hungry. Yet today. I have yet to meet a person in Minnesota that says they don't have access to enough food to eat." #mnleg https://t.co/H7JsyfsGWw
Aaron Rupar @atrupar
Data Visualizations
Most trans adults say transitioning made them more satisfied with their lives from the Washington Post. (Interesting styling on the bar charts here.)
The Ghosts of Housing Discrimination Reach Beyond Redlining from the Urban Institute
Every Country Has Its Own Climate Risks. What’s Yours? from the New York Times
Build My Viz by Mohit Kumar
TV, Movies, and Miscellaneous
Friday Night Lights, Netflix (I’ve watched this entire series twice, but just had a hankering to see some Riggins this week)
Ted Lasso, AppleTV
Old, HBOMax
Note: As an Amazon Associate I earn from qualifying purchases.
Do No Harm Guide Call for Papers
Last call for the two call for papers currently open for submissions to the next set of Do No Harm Guides from the Urban Institute. One report will focus on the US experience of creating more inclusive data products and the other will focus on the international experience. If you’re interested in submitting to either, please check out the links below. The submission deadline is April 1, 2023.