ABSTRACT
Factorial analysis of variance deals with models with one dependent variable (DV) and more than one independent variable (IV). It is extremely useful for investigating the effects of different factors on a single measure, but very often a single measure fails to capture the complexity of a psychological construct. One of the examples in Chapter 2 concerned the unpacking of IQ into a number of related, but discrete components. In that example, the three components were treated as covariates in an ANOVA design with a single DV. This is an appropriate procedure when the DV is the only variable of interest. If, however, we were mostly interested in IQ, we would either have to collapse the three components into a rather artificial single value, or examine the effect on the three values simultaneously. The latter procedure is known as multivariate ANOVA, or MANOVA for short.
