Information Design, Data Visualization, and Statistical Ethics
Designing for Evidence
The communicative role of visualization
Information design transforms structured data into perceptible patterns. A strong visualization should preserve data integrity, support comparison, reveal uncertainty, and minimize perceptual distortion. In research contexts, design choices have epistemic consequences because they shape what can be seen, inferred, and misunderstood.
Visualization goals
What is a common ethical risk in data visualization?
Scale manipulation can visually exaggerate or diminish differences.
Correct answer: Misleading scale choices that distort magnitude
Why is uncertainty important to visualize?
Uncertainty is part of the evidence and should not be hidden.
Correct answer: It helps users understand the limits and confidence of the data.
Chart Selection as Research Design
Choosing an appropriate chart
Bar chart
- Best for comparing discrete categories
- Supports ordered comparison
- Less suited to continuous distributions
Scatterplot
- Best for relationships between two quantitative variables
- Reveals clusters and outliers
- Can become cluttered with large datasets
Which chart is usually best for showing the relationship between two continuous variables?
Scatterplots are the standard choice for two quantitative variables.
Correct answer: Scatterplot
State one reason pie charts are often discouraged in research communication.
Angle and area judgments are less accurate than position along a common scale.
Correct answer: They make precise comparison of slices difficult.