MEPX
Chapter 1 of 10All chapters
  1. 1What correlation measures
  2. 2The five explanations
  3. 3Confounders
  4. 4Experiments
  5. 5Natural experiments
  6. 6Weighing causal evidence
  7. 7Selection effects
  8. 8Regression to the mean
  9. 9Language and reporting
  10. 10Acting under uncertainty

Chapter 1 of 10

What correlation measures

Movement together, nothing more.

The number

A correlation coefficient runs from -1 to 1 and measures how tightly two variables follow a straight line. It says nothing about direction of influence or about why.

  • A strong curved relationship can score close to zero.
  • Always plot the data: identical statistics can hide wildly different shapes.

Spurious correlations

In any large dataset, some unrelated series will track each other closely. With enough variables, striking correlations are guaranteed by arithmetic rather than meaning.