Natural Experiment

A natural experiment is a real-world situation where some outside force — a lottery, a policy change, a geographic border, a weather event — assigns treatment in a way that is effectively random, even though no researcher planned it. It hands you randomization that you could never ethically or practically arrange yourself.

Natural experiments power many famous causal studies and underlie methods like instrumental variables, difference-in-differences, and regression discontinuity. The art lies in arguing convincingly that the assignment really was as-good-as-random and not secretly tangled with the outcome — a claim that demands knowledge of the context, not just the data.