ANOVA_robust: A SAS Macro for Parametric Tests of Mean Differences in One-Factor ANOVA Models

Abstract

Testing the equality of several independent group means is a common statistical practice in the social sciences. The traditional analysis of variance (ANOVA) is one of the most popular methods. However, the ANOVA F test is sensitive to violations of the homogeneity of variance assumption. Many alternative tests have been developed in response to this problem of the F test. These tests include some modifications of the ANOVA F test and others based on the structured means modeling technique. This paper provides a SAS macro for testing the equality of group means using thirteen methods including the regular ANOVA F test. In addition, this paper summarizes the results of a simulation study that compares the performance of these tests in terms of their Type I error rate under different conditions, especially under violations of the homogeneity of variance assumption.

Keywords

• analysis of variance
• homogeneity of variance assumption
• simulation study
• homoscedasticity
• heteroscedasticity
• sas macro