Which description best defines MANOVA?

• A. A multivariate analysis of variance across multiple dependent variables
• B. A univariate ANOVA on a single dependent variable
• C. A non-parametric comparison of medians
• D. A regression with a single predictor

Answer: (A)

MANOVA is about comparing groups on several dependent variables at once. It tests whether the vector of means across those outcomes differs between groups, taking into account how the outcomes relate to each other. This multivariate approach is helpful because it can detect group differences that might be missed when you look at each variable separately and it controls the Type I error that would inflate if you conducted multiple separate tests. When there is more than one outcome that you care about, MANOVA provides a single, coherent test of the overall difference in the combined outcomes. If you only have one dependent variable, you would use a univariate ANOVA instead. For non-parametric data, you’d use non-parametric alternatives, and if your goal is to predict a continuous outcome from predictors, regression would be the appropriate framework.