Correlation (Pearson / Spearman)
Analyze the correlation between two continuous variables. Reports both the Pearson product-moment correlation (with a Fisher-z 95% confidence interval and a t-test p-value) and the Spearman rank correlation (suited to non-normal or ordinal data).
① Enter data
One pair per line; two numbers (x and y) separated by a space or comma.
How to use & methodology
Pearson or Spearman?
Use Pearson when both variables are approximately normal and the relationship is linear; use Spearman rank correlation when the data are skewed, ordinal, or contain outliers. This tool reports both.
How is the confidence interval computed?
The 95% CI for Pearson r uses the Fisher z transform (transform r to z, build the interval on the z scale, then back-transform), appropriate for n≥4.
Does a significant coefficient mean a strong relationship?
Not necessarily. A significant p only means the correlation is unlikely to be due to chance; strength is read from the coefficient itself. With a large sample even a very weak correlation can be significant.
Can correlation show causation?
No. Correlation only reflects how strongly two variables move together; confounding or reverse causation may be present. Causal inference needs a well-designed study.