b. 1943, USA · Bell Labs statistician; pioneer of data-visualisation research
Also known as: William S. Cleveland
William S. Cleveland spent most of his career at Bell Labs, where he and Robert McGill ran a series of controlled experiments in the early 1980s on how accurately people read different kinds of statistical graphics. Their 1984 paper Graphical Perception gave the field its first empirical ranking of the visual encodings: position along a common scale is read most accurately, followed by position along non-aligned scales, length, angle, area, volume, and colour saturation last.
The Cleveland-McGill hierarchy is now the central piece of evidence cited against pie charts and three-dimensional decoration in statistical graphics, and underpins the design of every modern visualisation system. Cleveland's later books The Elements of Graphing Data (1985) and Visualizing Data (1993) collected the experimental work into practical guidance, and his concept of trellis displays seeded what became R's lattice and ggplot2 faceting.
Related people: Edward Tufte, Colin Ware
Works cited in this book:
- Graphical Perception: Theory, Experimentation, and Application to the Development of Graphical Methods (1984) (with Robert McGill)
- Graphical perception; theory, experimentation, and application to the development of graphical methods (1984) (with Robert McGill)
- Graphical Perception and Graphical Methods for Analyzing Scientific Data (1985) (with Robert McGill)
- Graphical perception and graphical methods for analyzing scientific data (1985) (with Robert McGill)
Discussed in:
- Chapter 14: Data Visualisation (The Cleveland-McGill Hierarchy)
- Chapter 14: Data Visualisation (Choosing Visual Encodings)