Data-to-Visual Encoding
Choosing which visual channel represents which variable is the central decision in visualization, governed by how accurately people read each channel.
The core mapping
Every visualization is a set of choices: this variable becomes position, that one becomes color, another becomes size. These are visual encodings. Because human perception reads some channels far more accurately than others, matching the most important variables to the most accurate channels is the foundation of a truthful chart.
Channel accuracy
Empirical studies rank how precisely people compare values by channel. Position along a common scale is read most accurately, followed by length, then angle and slope, then area, then color and density. This ordering is a practical guide: put the quantity that must be judged precisely on position or length.
- Quantitative data: prefer position and length; use color intensity only when position is taken.
- Ordered categories: position, or an ordered lightness ramp.
- Unordered categories: distinct hues or shapes, not a continuous ramp.
Data types drive choices
Match the encoding to whether data is quantitative, ordinal, or nominal. Encoding a nominal category with a continuous color ramp implies an order that does not exist; encoding a quantity with distinct hues loses magnitude. The type of the variable narrows the honest options.
Expressiveness and effectiveness
An encoding is expressive if it shows all and only the facts in the data, and effective if it uses the channels people read best for the most important facts. Aim for both; a chart can be pretty and still fail either test.
Kronos use
Kronos figures assign position and length to the quantities that must be compared precisely and reserve color for perceptual, colorblind-safe maps.