Coined by Edward Tufte in "The Visual Display of Quantitative Information", chartjunk refers to charts with redundant data-ink and non-data-ink. "Data-ink is the non-erasable core of a graphic, the non-redundant ink arranged in response to variation in the numbers represented." To keep this post within the discussion, I will be using the following 3 characteristics for chartjunk, as taken from the book that chartjunk has been termed:
- Vibrations - moiré vibrations (effects) create noise and clutter
- Gridlines
- Ducks - elements not relevant to the data that distract the audience
1. Vibrations
The image above is an attempt to create the idea of what a sci-fi character would see from the spaceship of a warp drive. This is the original format I have submitted for a #DiversityinData theme. It may also be classified as a chartjunk despite it looking like art. While the lines moving into the centre as no value in terms of data represented, only the squares moving out from the centre have meaning. Each square represents a winner and timing, with the fastest at the centre to the slowest at the outer square, Click on the image on the Tableau Public page to interact with the chart(junk). It is perhaps best to remove the extra lines moving into the centre like below:
2. Gridlines