Your examples make me think once again about an important aspect of data visualization we do not talk enough about: data, no matter how accurate or representative, always require interpretation. We bring our own mental models to the equation. As your example explains, the exact same information can lead people to very different conclusions. Scientists, when they do their work right, do their best to dissect and refute the realities their minds construct from data, often by crafting clever follow-up experiments. But we as a society are very far from that. We do not even seem to promote that kind of self-skepticism/humility.
That's one of the core themes in 'How Charts Lie'. Usually, charts don't “lie” (they aren't designed to lie on purpose), but we tend to lie to ourselves *with* them because we project.
Your examples make me think once again about an important aspect of data visualization we do not talk enough about: data, no matter how accurate or representative, always require interpretation. We bring our own mental models to the equation. As your example explains, the exact same information can lead people to very different conclusions. Scientists, when they do their work right, do their best to dissect and refute the realities their minds construct from data, often by crafting clever follow-up experiments. But we as a society are very far from that. We do not even seem to promote that kind of self-skepticism/humility.
That's one of the core themes in 'How Charts Lie'. Usually, charts don't “lie” (they aren't designed to lie on purpose), but we tend to lie to ourselves *with* them because we project.