For the Shine-Bower single-subject ANOVA the numerator and demoninator of all F tests based on the Shine-Bower error term are independent of each other. The same property holds for all such tests in the Shine Combined ANOVA except for the test for the trial by subject interaction. (Author/RC)

A weighted harmonic means analysis is presented that incorporates all of the available data, preserves the planned proportionality of the design, and avoids the problems associated with the replacement of missing data with sample estimates. (Author/BW)

A solution to some problems of maximized contrasts for analysis of variance situations when the cell sizes are unequal is offered. It is demonstrated that a contrast is maximized relative to the analysis used to compute the sum of squares between groups. Interpreting a maximum contrast is discussed. (Author/GK)

A major pitfall with fixed-effects analysis of variance in the nested design is illustrated. Under the null hypothesis of no intervention effect, the rejection rate for "F(sub fixed)" in the nested design is greater than the putative Type I error rate of the test, potentially to an extreme degree. (Author/TJH)

An experimental sampling procedure for communities on which coal had been surface-mined yielded missing cells and caused the number of degrees of freedom to be N instead of the usual N minus one. The apparent discrepancy is explained, and a solution to the problem is presented. (Author/JKS)

The authors argue that a parameterization in terms of cell means is a useful conceptual device for resolving problems inherent in the nonorthogonal analysis of variance. They demonstrate their argument by considering two widely-used methods for testing main effects. (Author/BW)

It is recommended that levels of analysis of variance (ANOVA) designs be applied to carrier variables to help overcome the lack of unambiguously generablizable results in many applied experiments. (Author)

Sections of this paper are: basic framework for the review; fundamentals of choosing contrasts in multivariate analysis of variance; principles and procedures for orthogonal and in nonorthogonal MANOVA; general problems and principles in the analysis of multivariate dependent variables. References. (Author/VW)

It is argued here that the choice of the appropriate method for calculating least squares analysis of variance with unequal sample sizes depends upon the question the experimenter wants to answer about the data. The different questions reflect different null hypotheses. An example is presented using two alternative methods. (Author/BW)

Describes some of the situations in which a psychologist is likely to violate the assumption of independent errors. A Monte-Carlo study of the effects of this violation is then described. (Author/RC)

The intent of this paper is to demonstrate that the standard age effect in analysis of variance provides an ineffective means of assessing developmental change when several age levels are involved and a roughly monotonic trend can be expected. (Author)

The problem of determining which of several populations has the largest mean is considered. The procedure described by Dudewicz and Dalal is extended to the case of unequal sample sizes. (JKS)