ERIC Number: ED223688
Record Type: RIE
Publication Date: 1979-Sep
Reference Count: 0
Post-MANOVA Data Analysis: Six Alternatives and Criteria for Selection.
Gabriel, Roy M.
Although there is a healthy trend toward employment of multivariate statistical methods in the analysis of data from experiments in psychology and education which involve multiple dependent variables, there remains a need to demonstrate the wide variety of data analytic techniques available for post hoc analysis of statistically significant multivariate results. This paper attempts to: (1) clearly illustrate the use of six pertinent statistical analysis methods on a common sample data set; (2) compare the results both in terms of sensitivity to the treatment differences and interpretive clarity with respect to the research hypothesis; and (3) present guidelines to the applied researcher for selecting among the alternatives in a particular research context. Post hoc data analytic methods presented are: univariate F tests, univariate multiple comparison methods, step-down F tests, linear discriminant analysis, multiple profile analysis, and multivariate simultaneous test procedures. (Author/PN)
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Authoring Institution: N/A
Identifiers: Post Hoc Methods
Note: Paper presented at the Annual Meeting of the American Psychological Association (87th, New York, NY September 1-5, 1979).