Which test would be used for nominal or categorical data?

Which test would be used for nominal or categorical data?

• A. Chi-square
• B. t-test
• C. ANOVA
• D. Pearson correlation

Answer: (A)

Nominal or categorical data describe categories without inherent numeric order, such as gender, ethnicity, or diagnostic category. To determine if the observed frequencies across these categories differ from what would be expected by chance or to see if two categorical variables are related, the chi-square test is used. It compares observed counts in each category to expected counts under the null hypothesis and can be applied as a test of independence in a contingency table or as a goodness-of-fit test. Other tests don’t fit categorical data as naturally: the t-test compares means of a continuous variable between two groups; ANOVA compares means across groups for a continuous outcome; Pearson correlation assesses linear association between two continuous variables. Note that chi-square assumes a sufficiently large sample and expected counts in each cell (typically at least five); with small samples, Fisher’s exact test may be more appropriate.

Nominal or categorical data describe categories without inherent numeric order, such as gender, ethnicity, or diagnostic category. To determine if the observed frequencies across these categories differ from what would be expected by chance or to see if two categorical variables are related, the chi-square test is used. It compares observed counts in each category to expected counts under the null hypothesis and can be applied as a test of independence in a contingency table or as a goodness-of-fit test.

Other tests don’t fit categorical data as naturally: the t-test compares means of a continuous variable between two groups; ANOVA compares means across groups for a continuous outcome; Pearson correlation assesses linear association between two continuous variables. Note that chi-square assumes a sufficiently large sample and expected counts in each cell (typically at least five); with small samples, Fisher’s exact test may be more appropriate.