SL 4.11—Expected, observed, hypotheses, chi squared, gof, t-test

There are two categorical variables here: satisfaction level and store branch.

When testing whether two categorical variables are related, the chi-squared test for independence is used with a contingency table.

The correct test is the chi-squared test for independence, option B.

GOF compares one variable to a theoretical distribution, not two variables to each other.

This compares means from the same subjects measured twice, not categorical relationships.

This compares means of two groups, not categorical association.

p-value $=0.041$, $\alpha =0.05$.

Since $0.041<0.05$, the p-value is less than the significance level.

By the rule p < α → reject $H_0$.

We reject $H_0$, option A.

This would apply only if p ≥ α, which is not the case here.

We never 'accept' $H_0$ in hypothesis testing, this phrasing is always wrong.

The decision rule gives a definite outcome once p and α are compared.