Visualizing many categories
So far in this chapter, we've only considered the case of differences in a numeric variable between two categories. Of course, many datasets contain more categories. Before you get to conducting tests on many categories, it's often helpful to perform exploratory data analysis (EDA), calculating summary statistics for each group and visualizing the distributions of the numeric variable for each category using box plots.
Here, we'll return to the late shipments data, and how the price of each package (pack_price
) varies between the three shipment modes (shipment_mode
): "Air"
, "Air Charter"
, and "Ocean"
.
late_shipments
is available; pandas
and matplotlib.pyplot
are loaded with their standard aliases, and seaborn
is loaded as sns
.
This exercise is part of the course
Hypothesis Testing in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Calculate the mean pack_price for each shipment_mode
xbar_pack_by_mode = ____
# Print the grouped means
print(xbar_pack_by_mode)