Color in Data Visualization
Using color to communicate numbers and categories accurately.
Categorical vs Sequential scales
Data visualization requires specific color strategies. Categorical scales use distinct, unrelated hues (red, blue, green) to represent different categories (e.g., Apples, Oranges, Bananas). They don't imply magnitude.
Sequential scales use a single hue that varies in lightness or saturation (e.g., light blue to dark blue) to represent numerical magnitude (e.g., Population density). The darker the color, the higher the value.
Categorical
- Data type
- Unordered categories
- Color strategy
- Distinct hues
Sequential
- Data type
- Ordered, continuous data
- Color strategy
- Lightness steps of 1-2 hues
If you are mapping median income from lowest to highest across a country, which scale should you use?
Diverging scales
A Diverging scale is used when data has a critical midpoint, usually zero. It combines two sequential scales joined by a neutral color (white or light grey) in the center.
For example, a map of profit/loss might use a diverging scale where 0 is white, negative values become increasingly dark red, and positive values become increasingly dark green.
What color is typically used at the midpoint of a diverging scale?