Which statistical test is used to compare means across three or more groups with a continuous outcome?

Which statistical test is used to compare means across three or more groups with a continuous outcome?

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

Answer: (C)

When you want to compare a continuous outcome across three or more independent groups, you use a One-Way ANOVA. It tests whether the group means are all equal by comparing how much of the total variation in the data is due to differences between groups versus variation within groups. If the between-group differences are large relative to the within-group variation, you get a significant F statistic and conclude that at least one group mean differs. This approach relies on observations being independent, the residuals being roughly normally distributed within each group, and similar variances across groups. If those assumptions don’t hold, or if the data aren’t normal, a nonparametric alternative like Kruskal-Wallis can be used. A paired t-test is for comparing two related groups, not three or more. The chi-square test handles relationships between categorical variables, not a continuous outcome. The Mann-Whitney U test compares two independent groups nonparametrically; for three or more groups you’d use Kruskal-Wallis (or One-Way ANOVA if assumptions are met).

When you want to compare a continuous outcome across three or more independent groups, you use a One-Way ANOVA. It tests whether the group means are all equal by comparing how much of the total variation in the data is due to differences between groups versus variation within groups. If the between-group differences are large relative to the within-group variation, you get a significant F statistic and conclude that at least one group mean differs. This approach relies on observations being independent, the residuals being roughly normally distributed within each group, and similar variances across groups. If those assumptions don’t hold, or if the data aren’t normal, a nonparametric alternative like Kruskal-Wallis can be used.

A paired t-test is for comparing two related groups, not three or more. The chi-square test handles relationships between categorical variables, not a continuous outcome. The Mann-Whitney U test compares two independent groups nonparametrically; for three or more groups you’d use Kruskal-Wallis (or One-Way ANOVA if assumptions are met).