Guidelines
Suggest edit

Prefer color-saturation over area for a secondary quantitative field during exact value reading

For exact value lookup and pairwise comparison, prefer color-saturation for a secondary quantitative field on point-based multivariate views to improve fidelity and mitigate interference from varying mark area for readers inspecting individual values.

  • purpose:refine
  • basis:empirical
  • quality:fidelity:use
  • lever:encoding
  • operator:lookup
  • reading-mode:exact
  • measure:multi
  • channel:color-saturation:use

advice

Encode the secondary quantitative field with color-saturation

Encode the secondary quantitative field with color-saturation when the main quantitative field must be read from position. For example, keep the main quantitative field on x or y and map the other quantitative field to color-saturation instead of varying mark area when readers must read exact values or compare two values.

reason

Why color-saturation helps here

Varying mark area acts as a distractor when the reader is trying to decode a different quantitative field from position. Color-saturation preserved better value-reading performance for the positioned field in these exact-reading tasks.

Mechanism: Color-saturation leaves the positioned readout visually stable, while changing mark area makes the positioned field harder to decode quickly and accurately.

Evidence: In controlled tests of trivariate point-based views, the designs with the main quantitative field on position and the secondary quantitative field on color-saturation outperformed the corresponding area-based designs for retrieve-value and compare-values tasks in both accuracy and completion time (Zeng & Battle, 2023; Kim & Heer, 2018).

Notes: This rule is about protecting the readability of the primary positioned field, not about maximizing readability of the secondary field.

context

Use when exact value reading is the priority

  • User Goal: Read exact values or compare two individual values of a main quantitative field.
  • Task: Exact lookup or pairwise value comparison.
  • Data: One categorical field and two quantitative fields.
  • Chart Setting: A point-based multivariate view already uses position for the main quantitative field and still needs to show a second quantitative field.
  • Success Criterion: Lower error and faster responses on individual-value tasks.

exceptions

Do not use this as a summary-task default

Break it when: The main task is to find group maxima or compare group averages. Why: The value-task advantage of color-saturation over area does not define the best choice for those summary tasks, where size-based encodings performed well.

costs

What you give up

Sacrifice: You give up the size-based summary cue that can help on some summary judgments. Risk: Applying this rule to summary tasks can miss the benefit that area-based encodings showed there. Mitigation: Use this rule only when exact reading of the positioned field is the main requirement.

mistakes

Common failure mode

Mistake: Encode the secondary quantitative field with varying mark area while expecting fast, exact reading of the primary positioned field. Why it fails: The changing area interferes with decoding the positioned field.

check

Compare the two encodings directly

Failure Sign: Readers hesitate or misread the main positioned value when mark sizes vary. Quick Check: Make two versions of the same chart, changing only the secondary quantitative field from area to color-saturation, and compare one lookup question and one pairwise comparison question. Stronger Test: Measure answer time and error on a small set of representative exact-reading prompts.

fix

Edit the secondary encoding

  • Remap the secondary quantitative field from area to color-saturation.
  • Keep the main quantitative field on x or y.
  • Revisit the encoding only if the task shifts from exact value reading to group-level summaries.

References

Kim, Y., & Heer, J. (2018). Assessing Effects of Task and Data Distribution on the Effectiveness of Visual Encodings. Computer Graphics Forum, 37(3), 157–167. https://doi.org/10.1111/cgf.13409
Zeng, Z., & Battle, L. (2023). A Review and Collation of Graphical Perception Knowledge for Visualization Recommendation. Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, 1–16. https://doi.org/10.1145/3544548.3581349