Spearman rank correlation
Measures how strongly two measurements move together using ranks, so skew and curves that keep one direction are fine.
Clinical example
How strongly a pain score from 0 to 10 and the number of analgesic doses are related.
When to use it
- You have two measurements per patient and want to know how strongly they move together.
- The variables are skewed or ordered scores, or the relationship is curved but keeps going one way.
- You describe an association. You are not showing that two methods give the same values.
Assumptions to check
A relationship that keeps one direction
It should always rise or always fall, not both. Plot a scatter diagram.
Independent observations
One pair of values per patient.
What to report
- ρ with a scatter plot
Common mistakes
Using correlation to show agreement
Alternatives
| If your situation is… | Use |
|---|---|
| Both variables are roughly normal, with a straight-line relationship | Pearson correlation |
| I actually want to show two methods give the same values | Bland-Altman plot and ICC |
In your software
# R
plot(d$pain, d$doses)
cor.test(d$pain, d$doses, method = "spearman", exact = FALSE)
NONPAR CORR
/VARIABLES=pain doses
/PRINT=SPEARMAN TWOTAIL NOSIG.