Randomization effects
Random assignment plays a critical role in A/B testing. Imagine you are a Data Scientist setting up an experiment to test the effects of different checkout page designs on business metrics.
You will explore sampling a percentage of users to simulate random enrollment of traffic into the experiment and check the distributions of certain attributes across randomly assigned users in each group. This allows you to verify the ability to generalize to the general traffic population, and isolate the impact of the one variable we are changing across groups: the check out page design.
checkout
DataFrame is loaded for you. Assume every row in the DataFrame corresponds to a unique user visiting the respective checkout_page
along with their actions and additional attributes.
This exercise is part of the course
A/B Testing in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Determine the normalized distribution of browser counts
checkout['browser'].____(normalize = ____)