Advances in composite-based structural equation modeling

Abstract

Structural equation modeling (SEM) has become a quasi-standard tool for analyzing complex inter-relationships between observed and latent variables. Two conceptually different approaches to SEM have been proposed: factor-vs. component-based SEM. Factor-based SEM approximates latent variables by common factors as in common factor analysis, whereas component-based SEM regards them as weighted composites of observed variables as in multivariate statistics such as canonical correlation analysis and principal component analysis. Factor-based SEM is represented by covariance structure analysis, whereas composite-based SEM includes generalized structured component analysis (GSCA; Hwang and Takane 2004), partial least squares (PLS; Lohmller 1989), regularized generalized canonical correlation analysis (Tenenhaus and Tenenhaus 2011), and several others. Although factor-based SEM remains prevalent in practice, numerous methodological advances (e.g.,

Topics & Concepts