We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Development of a polar extreme method for use in partial least squares SEM.
- Authors
George, Benjamin; Prybutok, Victor
- Abstract
The use of polar extremes as a data segmentation approach is a common practice to maximize variance within multivariate analysis. In this research, we propose the first application of the polar extremes methodology within partial least squares structural equation modeling (PLS-SEM). Discussed is the impact of unequal weights of indicators among latent constructs within PLS-SEM and the importance of this methodology to the current established PLS-SEM analysis approach. Using a case study approach, we demonstrate the method. Developing two different models within the case allows comparison of the path coefficients and percentage of variance explained between the current PLS-SEM application and the proposed polar extremes methodology.
- Subjects
LEAST squares; STRUCTURAL equation modeling; MULTIVARIATE analysis; RESPONSE styles (Examinations); COMPARATIVE studies
- Publication
Quality & Quantity, 2015, Vol 49, Issue 2, p471
- ISSN
0033-5177
- Publication type
Article
- DOI
10.1007/s11135-014-0004-0