Causation
Causation is the relation in which one factor directly produces or influences another, such that changing the cause changes the effect.
Why care about causation over correlation?
Causal claims are stronger than correlations. Saying X causes Y is a claim about how the world would change under intervention, not just how it currently looks.
Causal Factor
One of the things that affects an event.
A controlled study is the most effective way of establishing causality. The sample is split into comparable groups, the groups receive different treatments, and outcomes are compared.
Causation vs. Implication
“If I flip the switch, the lights turn on” has two readings:
- Logical/regularity: whenever switch-flip happens, light-on happens (implication)
- Causal: flipping the switch is what makes the lights turn on
Ice cream and public nudity
Ice cream sales and arrests for public nudity rise together. Banning ice cream would not reduce nudity, so the link is correlation, not cause. A causal claim says: if X were changed, Y would change.
Causal factors come in degrees:
- Smoking causes lung cancer, even though not every smoker gets it
- Exercise prevents heart disease (a negative causal factor), even though not always
Speaking in terms of causal factors makes room for other contributing factors and for hidden interactions.
Mental model
Instead of asking “is X necessary/sufficient for Y?”, ask:
- Does X increase the probability of Y?
- Under what conditions does X pay off?
- What other factors does X combine with?
- What is the bottleneck?
Success and most real-world outcomes are multi-factor systems, not single-switch logic. See also Causality.