Analysis of Variance and Covariance

Abstract This article is aimed at providing some basics of analysis of variance (ANOVA) for MBA and undergraduate students, particularly those interested in marketing research. ANOVA is used to determine the probability that differences in means across multiple categories are due solely to sampling error. ANOVA must be composed of at least one metric‐dependent variable and one or more nonmetric independent variables. In this article, ANOVA is introduced with the experimental design. Then, several types of ANOVA are discussed such as one‐way ANOVA, N ‐way ANOVA, analysis of covariance (ANCOVA), and multivariate analysis of variance (MANOVA). Furthermore, ANOVA is compared with multiple regression, another common statistical technique in marketing research. The last section shows some ANOVA examples applied in marketing research.