Avoid catastrophe. According to Edward Tufte, this is “the first principle” when using color in data visualization.1 Given the complexity contained in this warning, avoiding catastrophe is not an easy task to achieve.
Color is a complicated physiological phenomenon associated with symbolic, aesthetic, and emotional qualities. Each of these qualities is enough by itself to wreak havoc in data visualizations if not treated with care. Together, they make disaster almost inevitable and justify Tufte’s principle.
Corrections & Suggestions
None so far.
Files
- (XLSM) Workbook for Chapter 15
Additional Resources
- Stephen Few: “Practical Rules for Using Color in Charts”;
- Hichert color palette.
- Edward Tufte: Envisioning Information (Amazon link), chapter “Color and Information”.
- Robbins, Naomi B. and Richard M. Heiberger. “Design of Diverging Stacked Bar Charts for Likert Scales and Other Applications.” Journal of Statistical Software. Volume 57, Issue 5. 2014.
- Color Brewer.
- Adobe Color CC.
- Color blindness simulator: Color Oracle.
- How To Avoid Equidistant HSV Colors and Color picker.
- I want hue.
- Lisa Rost: Your friendly guide to colors in data visualization
Corrections? Suggestions? Leave a replay below.