ETF5922 Chap.7 Graphical Excellence and Human Perception
Graphical Excellence and Human Perception
Geometry can distort quantity
Data density evaluates how much useful information a display carries relative to its footprint. The data-to-ink principle removes non-essential decoration without deleting necessary context.
Attention follows strong signals
Position and length support more precise judgments than angle, area or colour hue.
Gestalt similarity, enclosure, connection and proximity shape the groups a viewer perceives.
Measure graphical integrity
Scaling both height and width makes area grow with the square of the linear factor.
A doubled icon therefore occupies four times the area and can imply a much larger change than the data contain.
Measure graphical integrity in context
Graphical excellence joins economy with truthful perception. Removing chartjunk is useful when non-data marks compete with evidence, but context such as baselines, labels and uncertainty may be indispensable.
Preattentive contrast can direct attention quickly, while Gestalt grouping can imply relationships before labels are read. These powers require restraint. A scaled pictogram demonstrates the risk: changing two dimensions makes area grow quadratically and exaggerates the represented ratio.
Quantifying both data and graphic changes converts an ethical concern into an auditable comparison and points toward aligned length or position as a repair.
What this chapter covers
- 01
Data density evaluates how much useful information a display carries relative to its footprint.
- 02
The data-to-ink principle removes non-essential decoration without deleting necessary context.
- 03
Position and length support more precise judgments than angle, area or colour hue.
- 04
Gestalt similarity, enclosure, connection and proximity shape the groups a viewer perceives.
- 05
Use density and ink to frame the reader's task
- 06
Check chartjunk against position before styling
- 07
Explain how length changes the visible comparison
- 08
Audit angle without removing necessary context
Compare graphical change with data change
- 2Calculate the data change as 25% and the displayed area change as 300%.
- 3Replace two-dimensional icon scaling with aligned one-dimensional length.
- 3Explain that perceptual magnitude must remain proportional to evidential magnitude.
Key terms
- Data ink
- Marks that represent data or essential guides, considered relative to all visible ink in the display.
- Chartjunk
- Decoration or redundant visual matter that competes with the evidence without improving interpretation.
- Preattentive cue
- A feature such as strong contrast or orientation that can be noticed before deliberate search.
- Lie factor
- The ratio of apparent graphical change to actual change in the represented quantity.
Graphical Excellence and Human Perception FAQ
How does a two-dimensional icon exaggerate change?
Scaling both height and width makes area grow with the square of the linear factor. A doubled icon therefore occupies four times the area and can imply a much larger change than the data contain.
How is a lie factor interpreted?
Compare the proportional change suggested by the graphic with the proportional change in the data. A value far from one indicates distortion. State whether the visual uses length, area or another geometry, because a doubled width and height creates a fourfold area even when the data rise much less.
What is the difference between data ink and minimalism?
Data ink focuses attention on marks and guides that carry evidence. Minimalism can remove elements indiscriminately, including labels or context needed for interpretation. The proper test is whether a component supports accurate reading, not whether the page contains as little ink as possible.
Why quantify distortion instead of calling a chart bad?
Quantification identifies the exact mismatch between graphical and data change and allows a repair to be evaluated. In the icon example, the area grows by 300% while the data grow by 25%. That evidence explains why aligned length is preferable more clearly than an unsupported claim about taste or professionalism.
Exam move
Take a graphic containing a truncated axis, a scaled icon and redundant decoration. Compute the underlying data ratio, estimate the visible geometric ratio, and identify which preattentive feature dominates. Rebuild the comparison with aligned position and minimal guides. Explain the repair through the reader's numerical judgment, not through a claim that the new version simply looks cleaner.
Working through Graphical Excellence and Human Perception in ETF5922? Sia is AskSia’s AI Computer Science tutor — ask any ETF5922 Graphical Excellence and Human Perception question and get a clear, step-by-step explanation grounded in how ETF5922 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
