Calculate the correlation between two variables controlling for a third.
Partial correlation removes the variance both variables share with the control. A large drop from the raw correlation indicates confounding by that third variable. Spurious correlations often collapse once an obvious confounder is controlled, which is exactly what this calculation reveals.
Partial Correlation
r_xy·z = (r_xy − r_xz r_yz) ÷ √((1 − r²_xz)(1 − r²_yz))
r_xy·z = (r_xy − r_xz r_yz) ÷ √((1 − r²_xz)(1 − r²_yz)) Partial correlation removes the variance both variables share with the control. A large drop from the raw correlation indicates confounding by that third variable.
Spurious correlations often collapse once an obvious confounder is controlled, which is exactly what this calculation reveals.
This calculator takes 4 inputs: Correlation between x and y, Correlation between x and the control, Correlation between y and the control, Sample size. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.