Fragmented treatment in the literature of the problem of unbalanced data suggests an examination within a single framework of some of the more important issues. (JH)

The multivariate analog of Hays omega-squared for estimating the strength of the relationship in the multivariate analysis of variance has been proposed in this paper. The multivariate omega-squared is obtained through the use of Wilks' lambda test criterion. Application of multivariate omega-squared to a numerical example has been provided so as…

The repeated measures design is of importance to those interested in doing learning studies concerned with repeated trials on a single type of task, repeated trials on different tasks, or both together crossed with and following different treatments. In doing analysis of variance with such data it is assumed that the data fits an additive model.…

Methods for the calculation of eta coefficient, or correlation ratio, squared have recently been presented for examining the strength of relationship in univariate analysis of variance. This paper extends them to the multivariate case in which the effects of independent variables may be examined in relation to two or more dependent variables, and…

An alternate procedure is presented for testing the trial-by-subject interaction in the Shine Combined analysis of variance test. This new procedure is designed to get around the independence and distributional problems of the orginal F test. (Author)

Previous arguments against the use of Analysis of Variance (ANOVA) to determine test bias are extended. Explains why ANOVA may obscure test bias when it exists and also create a false impression of bias when it does not exist. (RB)

Problems encountered in univariate analysis of variance for repeated measures can be circumvented without recourse to less generally familiar multivariate methods. (Authors)