Correct option is A
Samples t-test → Median Test;
One-Way ANOVA → Kruskal-Wallis;
Dependent Samples t-test → Sign test;
Pearson's correlation → Kendall's Tau.
Information Booster:
• Parametric tests assume normal distribution and interval/ratio data; non-parametric tests don't require these assumptions
• Median Test (or Mann-Whitney U) compares two independent groups on ordinal data—alternative to independent t-test
• Kruskal-Wallis test compares three or more independent groups—non-parametric alternative to one-way ANOVA
• Sign test compares paired observations by counting positive vs. negative differences—alternative to paired t-test
• Kendall's Tau measures monotonic association between variables—alternative to Pearson's correlation
• Non-parametric tests are used when data violate parametric assumptions (non-normal distribution, ordinal scales, small samples)