Variation Sharing

Abstract

The most fundamental problem currently associated with structural equation modeling employing the partial least squares method is that it does not properly account for measurement error, which often leads to path coefficient estimates that asymptotically converge to values of lower magnitude than the true values. This attenuation phenomenon affects applications in the field of business data analytics; and is in fact a characteristic of composite-based models in general, where latent variables are modeled as exact linear combinations of their indicators. The underestimation is often of around 10% per path in models that meet generally accepted measurement quality assessment criteria. The authors propose a numeric solution to this problem, which they call the factor-based partial least squares regression (FPLSR) algorithm, whereby variation lost in composites is restored in proportion to measurement error and amount of attenuation. Six variations of the solution are developed based on different reliability measures, and contrasted in Monte Carlo simulations. The authors' solution is nonparametric and seems to perform generally well with small samples and severely non-normal data.

Forecasting the schedule performance resulting from the adoption of social distancing in construction projects

“…Additionally, in order to ensure the robustness of the results, the model was assessed for endogeneity using the instrumental variable as recommended by Kock (2020). Kock and Sexton (2017) suggest conducting such an analysis using the single stochastic variation sharing method. The result of the analysis conducted showed that the link from the instrumental variable to the dependent variable was non-significant ( p -value = 0.143, β = −0.081).…”

Do women have a say? A moderated mediation model's influence on the leverage policy toward corporate sustainable growth

“…IV selectively shares variability with another variable. Utilizing the "Explore analytic composites and instrumental variables" menu in WarpPLS 8.0 with the "Single stochastic variation sharing" sub-option facilitates the creation and testing of IV for endogeneity control (Kock and Sexton, 2017). If the IV-link's path coefficient to the dependent variable is small and nonsignificant, it indicates successful implementation of the Heckman procedure, signifying no significant endogeneity in the model (Certo et al, 2016).…”

Exploring drivers of business performance in web-enabled supply chains: the role of web technology training and supply chain integration

“…To test the robustness of our structural model results, we relaxed the covariance-based structural equation modeling assumption of multivariate normality (Kock and Sexton, 2017). We used the factor-based PLS Type CFM3 algorithm in WARP PLS 8.0 to perform this procedure, which enables the modeling of non-linear relationships using both compositebased and factor-based SEM algorithms (Memon et al, 2021) and estimates a measurement model (Kock, 2022b).…”