Sciencespeaking topic
The Bradford Hill criteria: how do you prove causation where you can't run an experiment, as with smoking and cancer? When clues like consistency, a strong association and temporal order come together, do we earn the right to say cause?
— Austin Bradford Hill, criteria for causation
practice with this topic
Set the timer (5-30 min), take 20 seconds of prep if you like, start talking. Jot your thoughts onto the sticky-note board.
similar topics
- The double-blind trial: neither the participant nor the researcher knows who got the real drug. What does it tell us that merely knowing can distort a result this much?
- Fairness and the market: why do we get angry at a seller who multiplies the price of a shovel during a snowstorm? Why does the economically correct price feel morally wrong?
- In a forest ecosystem there is no such thing as 'waste'; every dead leaf, every carcass is someone else's meal. Does the concept of garbage belong to humans alone? How did nature abolish waste?
- Pain is not the damage itself but an interpretation the brain produces. That's why a soldier wounded in battle may notice the pain only hours later, while an anxious person can writhe from a small scratch. How much of pain is 'real,' and how much is the brain's script?
- The hydrogen atoms in a glass of water are older than every star in the universe; they're left over from the Big Bang. The water in your hand is a keepsake from the universe's first minutes. How can something so ordinary be so ancient?