ERIC Number: ED388691
Record Type: RIE
Publication Date: 1995-Apr
Reference Count: N/A
A Comparison of the Type I Error and Power of Selected MANOVA Simultaneous Test Procedures.
Sheehan, Janet K.
A Monte Carlo study was conducted using the Statistical Analysis System IML computer program to compare the multivariate analysis of variance (MANOVA) simultaneous test procedures of Roy's Greatest Root, the Pillai-Bartlett trace, the Hotelling-Lawley trace, and Wilks' lambda, in terms of power and Type I error under various conditions, including violations of MANOVA assumptions. The Type I error rates of moderately-restricted contrasts in simultaneous test procedures following a significant omnibus MANOVA were robust to violations of MANOVA assumptions, such that the actual alpha remained below the nominal alpha. However, the power of even Roy's Greatest Root is unacceptably low in moderately restricted contrasts under most conditions. Therefore, the results of this study do not generally support using moderately restricted contrasts to follow-up significant MANOVA tests, unless the number of dependent variables is limited to two, or the noncentrality structure is known to be concentrated in one group and one variable. (Contains 1 table, 12 figures, and 35 references.) (Author/SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A
Identifiers: Hotellings t; Pillais Trace; Roys Largest Root; Type I Errors
Note: Paper presented at the Annual Meeting of the American Educational Research Association (San Francisco, CA, April 18-22, 1995).