Alternatives Test in ANOVA with Unequal Variance and Unequal Sample Size

This article presents a discussion about the methods used in experimental statistics, which considers a case study in which we have experimental units with unequal sample sizes. For this type of experimentation, non-observation of certain assumptions underlying the carrying out of the associated tests is observable, forcing the researcher to opt for alternative methods. With the aim of comparing the average pedagogical performance of students from different groups of students, a database was considered, made available in three groups of students according to the distances travelled to school. Given the observable violation of the assumptions of equality of variances and normality, for the analysis and discussion of the results, some non-parametric tests were described, for multiple comparisons and post-hoc tests, which are pairwise comparisons of means. Keywords:- Analysis of Variance, Unequal Sample Size, and Non-Parametric Test.