Correlation
A measure of how strongly two variables move together, and a caution against inferring causation.
Definition
Correlation quantifies the degree to which two variables vary together. The Pearson correlation coefficient ranges from -1 (perfect inverse) through 0 (no linear relationship) to +1 (perfect direct relationship).
Rank-based measures such as Spearman's coefficient capture monotonic relationships that Pearson's linear coefficient can miss, and are more robust to outliers. Choosing the right correlation measure depends on whether the expected relationship is linear or merely order-preserving.
Correlation is a fast, cheap first look at how variables relate, but its famous limitation, that it does not imply causation, is a recurring source of error in analysis and public reasoning alike. A lurking third variable can produce strong correlation between quantities with no direct link. Rank-based measures capture monotonic relationships that the linear coefficient misses, and choosing the right measure depends on the shape of the relationship expected.
Cautions
- Correlation measures linear association; nonlinear links can be missed.
- It is sensitive to outliers.
- Correlation does not imply causation.
- A lurking third variable can create spurious correlation.
Why it matters
Correlation is a first, cheap look at how variables relate, guiding which relationships merit deeper modeling. Its famous pitfall, mistaking correlation for causation, is a recurring source of error in data analysis and public reasoning alike.
Fusion connection
Correlation screening across simulation variables flags candidate dependencies in Hyperion data, which are then tested against physics rather than accepted on the correlation alone.