International Journal of Multivariate Data Analysis · 2017 · 1.4K citations · 0 references
Quality Of LifeEngineeringSocial IndicatorSocial PsychologyPsychometricsSocial SciencesPsychologyData ScienceSystems EngineeringData IntegrationData ManagementComposite ReliabilitySocial ImpactStructural Equation ModellingPsychosocial FactorApplied Social PsychologyCross-sectional StudyNumerous Statistical MethodsSociologyData Modeling
Numerous statistical methods are available for social researchers, and selecting the appropriate technique, such as choosing between CB‑SEM and PLS‑SEM, can be challenging. The study directly compares CB‑SEM and PLS‑SEM using identical models and data. The authors use the same theoretical measurement and structural models and dataset to perform the comparison. CB‑SEM requires removing many indicators to achieve acceptable fit, whereas PLS‑SEM yields higher composite reliability and convergent validity, comparable discriminant validity and beta coefficients, and substantially better variance explained in dependent variables.
Numerous statistical methods are available for social researchers. Therefore, knowing the appropriate technique can be a challenge. For example, when considering structural equation modelling (SEM), selecting between covariance-based (CB-SEM) and variance-based partial least squares (PLS-SEM) can be challenging. This paper applies the same theoretical measurement and structural models and dataset to conduct a direct comparison. The findings reveal that when using CB-SEM, many indicators are removed to achieve acceptable goodness-of-fit, when compared to PLS-SEM. Also, composite reliability and convergent validity were typically higher using PLS-SEM, but other metrics such as discriminant validity and beta coefficients are comparable. Finally, when comparing variance explained in the dependent variable indicators, PLS-SEM was substantially better than CB-SEM. Updated guidelines assist researchers in determining whether CB-SEM or PLS-SEM is the most appropriate method to use.