Which statistic is used to determine differences among three or more groups?

• A. T-test.
• B. Mann-Whitney U.
• C. ANOVA.
• D. Correlation.

Answer: (C)

When you have three or more groups, you want a test that looks at all group means at once rather than making multiple pairwise comparisons. ANOVA (analysis of variance) does this by comparing how much of the total variability in the data comes from differences between the group means versus variability within each group. If the between-group differences are large compared to within-group variation, the result is significant, indicating that not all group means are equal.

The null hypothesis in ANOVA is that all group means are the same. A significant result tells you at least one group differs from the others, but it doesn’t specify which ones. To find out exactly which groups differ, you perform post-hoc comparisons (like Tukey or Bonferroni tests). This approach helps control the overall risk of a false positive that would occur if you ran many separate t-tests.

ANOVA assumes that data are independent, normally distributed within each group, and have similar variances across groups. If these assumptions aren’t met, alternatives like a nonparametric method (Kruskal-Wallis) can be used.