Causation

A relationship in which (1) two variables are related/correlated, (2) one of them comes before the other in time, and (3) their relationship is not due to (caused by) any third variable.

• Ex: Patients are randomly assigned to receive either a drug or a placebo, and then their symptoms are measured to see changes. If the drug group has fewer symptoms (negative correlation), it’s clear that the drug caused the change in symptoms because the groups were the same at the start, ruling out third variables.
• Ex: Ice cream sales and drownings rise together every summer (are correlated), but ice cream sales do not cause drowning. Hot weather (a third variable) causes both.

---

Last updated: September 25, 2026 at 3:04 PM