Daft (Chart)junk

by Vivian Ng

I discovered my love for data storytelling quite early in my journey to data analytics and am always pushing myself to figure out how best to make insights easy to understand. That’s why I was stoked to learn best practices for data visualization today!

One principle I learned about was to avoid chartjunk. Coined by Edward Tufte, chartjunk consists of all the visual elements that aren’t necessary to understand what’s going on in the visualization. Some examples of chartjunk are:

  • Heavy gridlines
  • Unnecessary text
  • Unnecessary color
  • 3D effects
  • Images
  • Axis ticks
  • Redundant labels
  • Redundant titles
  • Shading and gradients

Given what I know now about chartjunk, I thought I’d look through one of my old Tableau Public visualizations and point out the chartjunk in them, as well as what I’d do differently.

Let’s take a look at a story point of a Tableau story I made before joining the Data School New York, on electricity capacities and renewable energy sources. And That’s Watt’s Up! looked at percent growth in electricity capacities for renewable vs. nonrenewable energy sources over time. The second story point specifically examined renewable energy source capacities over time.

Upon revisiting this line graph, I immediately spotted several elements of chartjunk:

  • Color legend on the right side: I could instead add labels next to the lines for each energy source
  • Colors: There are quite a few colors for each energy source here—a set action would allow the user to hover or select one line to be colored in, while the rest would be grayed out; decreasing the number of colors on the chart would prevent the design of the chart from competing for attention with the information conveyed by the chart itself
  • Gridlines: Removing them would give the visualization a cleaner look
  • Y-axis: “in Kilowatts, thousand” is difficult to understand because the audience has to do some math to mentally convert the unit to Megawatts. In addition to changing the scale, I could also decrease the number of axis ticks
  • Annotation: While the Three Gorges Dam did contribute greatly to how much global electricity is hydroelectricity, it’s not relevant to the overall story, which moves from the global electricity generation capacity landscape to the U.S. soon after this story point.
  • Tooltips: If you hover over the data points, you’ll see default tooltips 😅 I’ve learned my lesson with them now, though! If I were to rework it, I would format the tooltip to read more easily

I can’t wait to see what I’ll be learning next at DSNY! Until next time!

[cross-posted here]

Fri 02 Jun 2023

Thu 01 Jun 2023

Thu 25 May 2023