The article compares the effectiveness of two types of statistical tests, parametric and non-parametric, in situations where data is not normally distributed. The researchers found that non-parametric tests, like the Mann-Whitney U-test, are more robust when dealing with discrete data compared to parametric tests like the Student’s t-test. They also discovered that including tie correction factors in the non-parametric tests significantly improves their robustness. Overall, when working with discrete data, using non-parametric tests with tie correction factors is more reliable than parametric tests.