Session Ready
Exercise

Mann-Whitney U test

Let's assume that you are an Analytics Engineer at an early-stage start-up where the traffic size to your website is small. Since in many cases you will need to run the tests for a long duration to gather a sample size that is large enough to satisfy parametric tests' assumptions, you may need to lean on non-parametric tests to make faster decisions.

One of these tests is the Mann-Whitney U test, which is a statistical significance test used to determine if two independent samples were drawn from a population with the same distribution. You will leverage this test to analyze the difference in time_on_page between two groups of the checkout dataset. The checkout DataFrame is available for you and pingouin has been loaded along with pandas and numpy.

Instructions 1/3
undefined XP
  • 1
  • 2
  • 3
  • Calculate the mean and rows count of the time_on_page column for each variant in one step.
Loading