Encode the primary quantitative field with size for grouped summary tasks in point plots

Use point size for the primary quantitative field when the job is to judge grouped summaries rather than exact single-point values. For example, in a categorical point plot used to find which group contains the highest value or which group has the larger average, encode the primary quantity with size instead of a positional axis.

Summary tasks ask readers to judge a set of marks, not decode one exact value at a time. Size supported those grouped judgments well in the tested point plots.

Mechanism: Size can support ensemble reading of a group of marks, which helps when readers need to judge maxima or averages across categories.

Evidence: For summary tasks, dot plots with the primary quantitative field encoded by size outperformed other dot-plot encodings that used position for that same primary field (Kim & Heer, 2018).

• User Goal: Decide which category has the highest value or the larger average.
• Task: Group-level summary comparison across categories.
• Data: One categorical field and a primary quantitative field shown with points.
• Chart Setting: A categorical point plot where the primary quantity could be assigned either to size or to position.
• Success Criterion: Lower error on summary judgments.

Break it when: Readers must read an individual value or compare two specific points exactly. Why: Position was the stronger choice for those value tasks.

Sacrifice: Exact value reading of individual marks becomes less direct. Risk: The same chart becomes a poor fit if the task shifts from overview judgment to exact lookup. Mitigation: Keep size for summary views and switch back to position when the task becomes exact point reading.

Mistake: Keep the primary quantitative field on position in a categorical point plot that is mainly used for average or maximum judgments. Why it fails: The study found that size-based encodings performed better for those summary tasks.

Failure Sign: Readers struggle more with “which group is larger” or “which group contains the maximum” than with point-level lookup. Quick Check: Compare the current plot against a version that moves the primary quantity from position to size. Stronger Test: Ask one maximum-finding question and one average-comparison question on both versions and keep the lower-error version.

• Move the primary quantitative field from position to size.
• Keep the category on x or y so groups remain easy to identify.
• Reserve this size-based encoding for summary views rather than exact value views.