Confounding Variable

A hidden third variable that influences both observed variables, creating a false impression of a direct relationship between them. In the lecture, temperature drives both ad spend and revenue.

When to use

Whenever you see a correlation, ask what the hidden third variable is. Lecture examples: ice cream sales and drownings (hot weather), stores with more employees (store size), Nobel laureates and chocolate (wealth).

See Lec 1.

Appears in

No references yet.