Analyses of variance (ANOVA) with the general linear model (GLM) in many st
andard statistical packages use an overparameterized model, a model unfamil
iar to most behavioral science researchers. Estimates and significance test
s with GLM procedures are calculated by computing generalized inverses and
estimates of estimable functions. Using simple examples, the authors discus
s the concepts that underlie the solutions for 1-way and 2-way ANOVAs with
overparameterized models and illustrate how these models allow one to evalu
ate the research hypotheses. The authors also extend the discussion of over
parameterized models to a more general modeling approach than GLM, the gene
ral linear mixed model.