Abstract
The usefulness of analysis of variance (ANOVA) estimates of variance components is impaired by the frequent occurrence of negative values. The probability of such an occurrence is therefore of interest. This probability depends on the design used and the true values of the variance components. In the present article, we propose a method for modeling this probability in order to gain a better insight into the effect of data imbalance on the quality of ANOVA estimation. Generalized linear models techniques are used for this purpose. A demonstration of the proposed methodology is made using the unbalanced random one-way model.
Keywords
Get full access to this article
View all access options for this article.