Mill's Methods: Method of Difference

A second test to determine whether a given factor plays a causal role requires us to take away that factor, holding everything else constant, and to see whether the effect still occurs. Mill called this the method of difference.


If you car makes a funny noise when you accelerate, take your foot off the pedal and see whether the noise goes away. Scientists use the same technique when they do controlled experiments. In testing the efficacy of a new medicine, for example, they would use two carefully matched groups. One group would get the drug, and the other would get a placebo. The only difference between the groups would be the presence or absence of the drug, so that any difference in results could then be attributed to that factor.

Method of agreement | Method of difference |
Joint method of agreement and difference |
Concomitant variations | Residues

Mill's methods

© Copyright 1998, W.W. Norton & Co.