This test assesses whether two categorical variables forming a contingency table are associated. What is it called?

This test assesses whether two categorical variables forming a contingency table are associated. What is it called?

• A. ANOVA
• B. t-test
• C. Mann-Whitney U
• D. Chi-square test

Answer: (D)

Testing whether two categorical variables are related using a chi-square test of independence is what this item targets. It uses a contingency table of observed counts and asks whether those counts align with what would be expected if the variables were independent. When there’s an association, the observed frequencies differ enough from the expected ones to yield a small p-value, suggesting the variables are not independent. The calculation sums, for every cell, (observed − expected)² divided by the expected count. Degrees of freedom are (rows − 1) times (columns − 1). A practical note: the chi-square approximation works best when expected counts in each cell are reasonably large; with very small counts, Fisher’s exact test is more appropriate. Other tests listed, such as ANOVA, the t-test, or the Mann-Whitney U, are designed for comparing a numeric outcome across groups, not for assessing the association between two categorical variables.

Testing whether two categorical variables are related using a chi-square test of independence is what this item targets. It uses a contingency table of observed counts and asks whether those counts align with what would be expected if the variables were independent. When there’s an association, the observed frequencies differ enough from the expected ones to yield a small p-value, suggesting the variables are not independent. The calculation sums, for every cell, (observed − expected)² divided by the expected count. Degrees of freedom are (rows − 1) times (columns − 1). A practical note: the chi-square approximation works best when expected counts in each cell are reasonably large; with very small counts, Fisher’s exact test is more appropriate. Other tests listed, such as ANOVA, the t-test, or the Mann-Whitney U, are designed for comparing a numeric outcome across groups, not for assessing the association between two categorical variables.