Causality

There are 4 elements necessary to establish a causal relationship. These are:

1. Theory: there should be some logical reason to suggest that variable x is the cause and variable y is the effect.

2. Time order: if x causes y, then x must precede y in time.

3. Correlation: If x causes y, then changes in x should result in changes in y, and if changes in y have occurred, we would expect there to have also been changes in x. Correlation can be positive (x and y both increase; x and y both decrease) or negative (y increases when x decreases; y decreases when x increases).

4. No plausible alternative explanations: There is no z variable that affects both x and y in a way that makes it appear that x and y correlate when this is not actually the case.

References:

Mill, John Stuart. A system of logic, ratiocinative and inductive: Being a connected view of the principles of evidence and the methods of scientific investigation. Vol. 1. Longmans, green, and Company, 1884.

Cook, Thomas D., Donald Thomas Campbell, and William Shadish. Experimental and quasi-experimental designs for generalized causal inference. Boston: Houghton Mifflin, 2002.