This research aims to study methods that can be used to test paired sample data in parametric and nonparametric statistics. Parametric statistics are used if the data sample meets certain assumptions, namely normal distribution and a minimum interval measurement scale. Meanwhile, if these assumptions are not met, you can use nonparametric. This research data was generated randomly to be applied to the paired sample test and the Wilcoxon test. These tests were applied to test the difference in averages before treatment and after treatment. The test results show that for normally distributed data using the paired t test it is proven that there is a difference between the averages before and after treatment. The same thing applies to data with an exponential distribution which is tested using Wilcoxon, that there is a difference in the average before treatment and after treatment.
