In order to establish that a relationship between two variables is indeed causal (i.e. that the independent variable causes, or affects, the dependent variable), one must demonstrate the following:
- There must be a correlation between the two variables (i.e. as one variable varies, so does the other).
[HOW: Bivariate tabular analysis/regression analysis]
- The cause must precede the effect (i.e. the independent variable has to happen before the dependent variable).
[HOW: Direct manipulation or through logic/theory]
- Other possible explanations for the relationship must be tested and rejected.
[HOW: Multivariate tabular analysis/regression analysis]