Which test would determine whether there is an association between gender and whether or not college-age drivers were ticketed?

Which test would determine whether there is an association between gender and whether or not college-age drivers were ticketed?

• A. 2-proportion z-test
• B. Chi-square test of independence with 1 degree of freedom
• C. Linear regression
• D. ANOVA

Answer: (B)

Testing whether gender is related to being ticketed uses a chi-square test of independence. It examines two categorical variables in a contingency table (gender and ticketed status) and asks if the distribution of ticketing is the same across genders or if there’s an association. If there’s a real relationship, the observed counts in the cells differ from what we’d expect if gender and ticketing were independent. For a 2x2 table, the degrees of freedom are (2−1)(2−1) = 1, which matches this scenario. Other methods aren’t suited for this question: a 2-proportion z-test compares a single proportion between two groups, not the full cross-tabulation; linear regression and ANOVA focus on numeric outcomes or mean differences, not the association between two categoricals (though logistic regression can model a binary outcome, the standard test for independence between two categoricals is the chi-square test).

Testing whether gender is related to being ticketed uses a chi-square test of independence. It examines two categorical variables in a contingency table (gender and ticketed status) and asks if the distribution of ticketing is the same across genders or if there’s an association. If there’s a real relationship, the observed counts in the cells differ from what we’d expect if gender and ticketing were independent. For a 2x2 table, the degrees of freedom are (2−1)(2−1) = 1, which matches this scenario. Other methods aren’t suited for this question: a 2-proportion z-test compares a single proportion between two groups, not the full cross-tabulation; linear regression and ANOVA focus on numeric outcomes or mean differences, not the association between two categoricals (though logistic regression can model a binary outcome, the standard test for independence between two categoricals is the chi-square test).