This week we looked at the body parts that people are attaching technology to. A nice simple data set this week, and the original visualization wasn’t particularly bad. The topic was interesting though, especially after some topics that required deeper thought lately.
LESSON 1: KEEP IT SIMPLE
If you’ve ever tuned into one of the Viz Review webinars, you will be fully aware that Eva and I stress simplicity in visualizations. Simplicity doesn’t mean that you need to dumb it down; simplicity means making it as easy as possible for other people to understand. Let’s compare and contrast two submission to this week’s Viz Review:
In Eric’s tweet, he says: “Straight away we can see that 59% of all devices are Wrist Devices and Most of these devices are for Fitness use.” Hmm, I’m not so sure. There’s no way to understand percentages without doing the math in your head. Secondly, there is way too much going on. There’s no title, nothing is sorted, there are too many colors. This dashboard is trying to show everything at once and it’s failing to communicate anything. We gave some feedback to Eric, so hopefully he iterates on it.
Contrast Eric’s viz that with this viz from Will Perkins:
Will has communicated clearly, simply, and we can quickly see the category that has the largest number of devices. Expressing the numbers as percentages make the relative distribution easy to understand. The icons match the topic and are nicely color coordinated. The title and subtitle set the story well.
We recommend to people that you show your work to someone who doesn’t know anything about the topic. If they can understand it without you needing to explain it, then you’ve likely done a good job. Give this a try next week and let us know how it goes.
LESSON 2: USE LINES TO CONNECT ITEMS ON AN INTERVAL SCALE
Line charts are designed to show change and patterns across continuous data such as time, age, and salary ranges. They are not, however, designed to show items that do not have a natural connectivity or order to them. This week, Claude Muller created this marginal histogram:
Note how he has used a line to connect the discrete items both across and down the view. When you use a line, it implies a pattern or order. There’s no natural order to the categories or body parts. For this specific example, I would remove the line completely as it’s about average price whereas every other part of the visualization is about the number of devices.
We provided Claude this feedback during Viz Review, so hopefully we see him iterate.
Now here are this week’s favorites.