Data Skeptic Visualizing Uncertainty
Mar 20, 2020
Lonnie Besançon, a postdoc studying HCI and visualization of statistical uncertainty, explores how design shapes interpretation. He discusses the arbitrary 0.05 threshold, the cliff effect around p-values, and how novel visuals like gradient rectangles and violin-style CIs can soften binary thinking. He also covers audience literacy, aesthetics, and cautious use of new visuals for teaching and transparency.
AI Snips
Chapters
Transcript
Binary Interpretation Distorts Findings
- Researchers often interpret statistical results in overly binary ways driven by publication pressures.
- This dichotomy leads to misleading yes/no conclusions instead of nuanced evidence statements.
The Cliff Effect Explained
- The 'cliff effect' is a sharp loss of confidence when p crosses a threshold like 0.05.
- People treat p-values dichotomously instead of viewing evidence as a continuous gradient.
Use Visuals To Soften Dichotomies
- Use visual estimation displays (confidence-interval based visuals) to encourage nonbinary interpretation.
- Visuals can show effect sizes and fuzziness, prompting more cautious conclusions.