If we want to understand whether the outcomes of two categorical variables are associated, we can use a Chi-Square test. It is useful in situations like:

• An A/B test where half of users were shown a green submit button and the other half were shown a purple submit button. Was one group more likely to click the submit button?
• People under and over age 40 were given a survey asking “Which of the following three products is your favorite?” Did these age groups have significantly different preferences?

In SciPy, we can use the function chi2_contingency() to perform a Chi-Square test. The input to chi2_contingency is a contingency table, which can be created using the pandas crosstab() function as follows:

#create table:
import pandas as pd
table = pd.crosstab(variable_1, variable_2)

#run the test:
from scipy.stats import chi2_contingency
chi2, pval, dof, expected = chi2_contingency(table)

For example, suppose we want to know whether gender is associated with the probability of a website visitor making a purchase. The null hypothesis is that there’s no association between the variables (eg. males, females, and non-binary people are all equally likely to make a purchase on the website, so gender and purchase-status are not associated). If the p-value is below our chosen threshold (often 0.05), we reject the null hypothesis and can conclude there is a statistically significant association between the two variables (eg. men, women, and non-binary people appear to have different probabilities of making a purchase, so gender is associated with purchase-status).