Stage 2 · Lesson 8 of 8
Correlation, confounders and causation
Correlation measures how two variables move together, from minus one to one. correlation captures straight-line relationships; Spearman works on ranks and handles curves and better. Correlation is not causation: a confounder, a third driving both variables, can create a strong link with no cause behind it.
Heavy shoppers use wishlists and spend more, but the wishlist may not cause the spending. To claim a cause you need an experiment, which the experiments stage covers, or a careful design that rules out confounders. Always plot a scatter chart before quoting a correlation.
Common traprecommending a business change because two numbers move together, then seeing nothing happen when the change ships.
A question to check yourself, an AI practice session and 1 cheat sheet entry come with the full roadmap. Unlock
Resources for this stage
Start with· follow one of these
- VideoFreeStatQuest with Josh Starmer
StatQuest · Beginner
Free: short, clear videos on each concept, from the normal distribution to p-values and power
- CourseFreeKhan Academy Statistics and Probability
Khan Academy · Beginner
Free: the probability, sampling distributions, confidence intervals and significance tests units, with practice
Practise
Also worth your time
Interactive visuals for probability, distributions, inference and regression: play with each one