Pairwise t-tests
Manually running separate comparisons using individual t-tests can be a pain as the number of groups gets larger. Thankfully, the pingouin package's .pairwise_tests()
method can make things easier.
You will explore the differences in average time-on-page metric between four different landing page variants loaded in the homepage DataFrame.
The dataset homepage
is available and has the columns signup
and time_on_page
. Every row in the DataFrame corresponds to a unique user visiting the respective landing_page
. The signup
column consists of binary data: '1' means the user signed up and '0' means abandoned the page, and the time_on_page
column represents the time (in seconds) each user spent on the landing page before abandoning or signing up . pingouin
has been loaded along with pandas
and numpy
.
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.
# Perform a pairwise t-test on signup, grouped by landing-page
pairwise = pingouin.pairwise_tests(data=homepage,
dv="____",
between="____",
padjust="____")
print(pairwise)