Everyday Application of the Cell Means Model

Abstract

Abstract Statistical computer packages that routinely generate the functional form of estimable functions help the user to understand better the difference between the various similar hypotheses in an unbalanced analysis of variance. Another way to understand different hypotheses is to consider estimable functions expressed in terms of cell means. This article proposes routinely generating cell means estimable functions and gives a formula for translating estimable functions in a given parameterization to the cell means equivalent and vice versa. We discuss an example.