Two fundamental ways of analyzing a multivariate analysis of variance (MANOVA) problem in more detail are the stepdown analysis (Roy, 1958; Bock, 1963), and the use of simultaneous confidence intervals (Gabriel, 1968; Morrison, 1967). Assuming a global test of significance, such as Wilk's A, has indicated non- ohance association between the independent variable and the set of dependent variables (this paper will limit discussion to the one- way design), the aforementioned methods may be used to deter- mine which of the dependent variables and/or groups are responsible for the global significance.
No takes yet. Share an insight, caveat, or question.
James P. Stevens (1973) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: