Information Technology Reference
In-Depth Information
We have identified elsewhere (Meyers et al., 2006) several benefits or
advantages to using MANOVA designs. We can briefly note some of them
here.
Multiple dependent measures provide the design with useful redun-
dancy and greater conceptual clarity.
MANOVA reduces Type I error rate (falsely rejecting the null hypoth-
esis) by avoiding separate univariate
F
tests that may inflate the
operational alpha level.
MANOVA considers the intercorrelation of dependent measures by
an examination of the variance-covariance matrices.
MANOVA can pinpoint group differences that sometimes become
masked at the univariate level of analysis.
While the topic of MANOVA is certainly beyond the scope of the
present topic, we encourage the interested reader to explore the following
publications: Huberty and Morris (1989), Hummel and Sligo (1971),
Meyers et al. (2006), Stevens (2002), and Weinfurt (1995).
