Global Least Squares Path Modeling: A Full-Information Alternative to Partial Least Squares Path Modeling
- PDF / 651,128 Bytes
- 26 Pages / 547.087 x 737.008 pts Page_size
- 22 Downloads / 268 Views
GLOBAL LEAST SQUARES PATH MODELING: A FULL-INFORMATION ALTERNATIVE TO PARTIAL LEAST SQUARES PATH MODELING
Heungsun Hwang
and Gyeongcheol Cho
MCGILL UNIVERSITY
Partial least squares path modeling has been widely used for component-based structural equation modeling, where constructs are represented by weighted composites or components of observed variables. This approach remains a limited-information method that carries out two separate stages sequentially to estimate parameters (component weights, loadings, and path coefficients), indicating that it has no single optimization criterion for estimating the parameters at once. In general, limited-information methods are known to provide less efficient parameter estimates than full-information ones. To address this enduring issue, we propose a full-information method for partial least squares path modeling, termed global least squares path modeling, where a single least squares criterion is consistently minimized via a simple iterative algorithm to estimate all the parameters simultaneously. We evaluate the relative performance of the proposed method through the analyses of simulated and real data. We also show that from algorithmic perspectives, the proposed method can be seen as a block-wise special case of another full-information method for component-based structural equation modeling—generalized structured component analysis. Key words: partial least squares path modeling, full-information, single optimization criterion, alternating least squares, block-wise generalized structured component analysis, component-based structural equation modeling, regularized generalized canonical correlation analysis, Lohmöller’s algorithm, Wold’s algorithm.
1. Introduction Partial least squares path modeling (PLSPM; Lohmöller, 1989; Wold, 1966, 1973, 1982) is a well-known statistical method for component-based structural equation modeling (SEM), where constructs are represented by weighted composites of observed variables (i.e., components). Component-based SEM is different from factor-based SEM, in which constructs are represented by common factors (e.g., Jöreskog & Wold 1982; Rigdon, 2012; Rigdon, Sarstedt, & Ringle, 2017; M. Tenenhaus, 2008). Covariance structure analysis (CSA; Jöreskog, 1970, 1978) has been a standard method for factor-based SEM, although other methods, such as consistent partial least squares (PLSc; Dijkstra, 2010; Dijkstra & Henseler, 2015) and generalized structured component analysis with measurement errors incorporated (GSCAM ; Hwang, Takane, & Jung, 2017), can also be used for estimating factor-based models. It has been recognized that the two SEM domains would need to conceptually differentiate from each other (e.g., Hair, Hult, Ringle, Sarstedt, & Thiele, 2017; Rigdon, 2012) and their statistical methods should be used for estimating models with their corresponding representations of constructs (i.e., CSA for factor-based models and PLSPM for component-based models), otherwise tending to yield biased solutions (e.g., Cho, Sarstedt, & Hwang, 2020;
Data Loading...
