Making Meaning out of MANOVA: The Need for Multivariate Post Hoc Testing in Gifted Education Research. (January 2020)

• Record Type: Journal Article
• Authors: Smith, Kendal N.; Lamb, Kristen N.; Henson, Robin K.
• Is Part Of: Gifted child quarterly. Volume 64:Number 1(2020)
• Page Start: 41
• Page End: 55
• Publication Date: 2020-01
• Subjects: MANOVA -- discriminant analysis -- multivariate analysis; Gifted children -- Education -- Periodicals; Gifted children -- Periodicals; 371.9505
• DOI: 10.1177/0016986219890352
• Languages: English
• ISSNs: 0016-9862

Abstract

Multivariate analysis of variance (MANOVA) is a statistical method used to examine group differences on multiple outcomes. This article reports results of a review of MANOVA in gifted education journals between 2011 and 2017 ( N = 56). Findings suggest a number of conceptual and procedural misunderstandings about the nature of MANOVA and its application, including pervasive use of univariate post hoc tests to interpret MANOVA results. Accordingly, this article aims to make MANOVA more accessible to gifted education scholars by clarifying its purpose and introducing descriptive discriminant analysis as a more appropriate post hoc technique. A heuristic data set is used to demonstrate the procedures for running a descriptive discriminant analysis, both in place of a one-way MANOVA and as a post hoc analysis to a factorial design. SPSS and R syntax are provided.