---
{
  "schemaVersion": 1,
  "version": "1.2.0",
  "id": "E-CM1984",
  "kind": "evidence",
  "title": "Cleveland and McGill: graphical perception",
  "summary": "Original paper; sections 4.3–4.5, especially p. 543 and Figure 16.",
  "evidenceType": "experimental",
  "citation": "William S. Cleveland and Robert McGill. 1984. Graphical Perception: Theory, Experimentation, and Application to the Development of Graphical Methods. Journal of the American Statistical Association 79(387), 531–554.",
  "url": "https://doi.org/10.1080/01621459.1984.10478080",
  "reviewed": "2026-09-13"
}
---

# Cleveland and McGill: graphical perception

**Type:** experimental. **Reviewed:** 2026-09-13.

## Source

William S. Cleveland and Robert McGill. 1984. Graphical Perception: Theory, Experimentation, and Application to the Development of Graphical Methods. Journal of the American Statistical Association 79(387), 531–554.

[Original / authoritative source](https://doi.org/10.1080/01621459.1984.10478080) · [Full paper](https://faculty.washington.edu/aragon/classes/hcde511/s12/readings/cleveland84.pdf)

## Scope

Original paper; sections 4.3–4.5, especially p. 543 and Figure 16.

## Limitations

Two proportion-judgment experiments inform quantitative encoding selection. The guide reports the log-error confidence interval, not a general dashboard effect. Paper stimuli and controlled comparisons do not establish performance on interactive pricing tasks.