Bivariate inductive statistics - interval and binary variables - dependent groups

Paired t-test

Student's paired t-test provides a hypothesis test of the difference between population means for a pair of random samples. The null hypothesis specifies that difference between means equals 0. Paired t-test assumes that differences are approximately normally distributed. If normality is not valid, one of the two nonparametric tests, the Wilcoxon Signed Rank test or the sign test, can be applied. They test median instead of the mean.

Example: Is sales increase after launching advertising campaign statistically significant or only by chance? (Was the advertising campaign effective?)