Visual encoding should express all of—and only—the information in the dataset.
Ordered data should be shown in a way we perceive as ordered.
Match channel and data characteristics.
Effectiveness
Encode most important attributes with highest-ranked channels
[Automating the Design of Graphical Presentations of Relational Information. Mackinlay. ACM Trans. on Graphics (TOG) 5:2 (1986), 110–141]
Where Does the Ranking Come From?
Accuracy
Discriminability
Separability
Popout
Accuracy: Fundamental Theory
Accuracy: Visualization Experiments
Effectiveness Tips
Discriminability
How many steps are usable? Line width: Not many 🤷🏼♂️
Separability
Popup
Find the odd ones.
Which ones pop up?
Grouping
Containment
Connection
Containment
Similarity
Grouping (cont.)
Proximity 👉 use spatial separation
Similarity 👉 use categorical channels
Proximity and Similarity
Relative vs. Absolute Judgments
Perceptual 👉 mostly relative judgments, not absolute
That’s why accuracy increases with:
Common frame
Scale and alignment
after [Graphical Perception: Theory, Experimentation, and Application to the Development of Graphical Methods. Cleveland and McGill. Journ. American Statistical Association 79:387 (1984), 531–554.]