Evaluating statistical significance
Now that you know the personalization variant outperformed the control, it's time for you to determine whether the result is statistically significant. Remember, statistical significance is vital to understanding whether a test showed a positive result by chance or if it is reflective of a true difference between the variants. This will enable your marketing team to make an informed choice about whether to roll out the feature or not.
Test out the stats.ttest_ind()
using control
and personalization
as the inputs.
Is the difference between the control and personalization statistically significant?
This exercise is part of the course
Analyzing Marketing Campaigns with pandas
Hands-on interactive exercise
Turn theory into action with one of our interactive exercises
