Parametric and Non Parametric Approach in Structural Equation Modeling (SEM): The Application of Bootstrapping

Abstract

Lately, there was some attention for the Variance Based SEM (VB-SEM) against that of Covariance Based SEM (CB-SEM) from social science researches regarding the fitness indexes, sample size requirement, and normality assumption. Not many of them aware that VB-SEM is developed based on the non-parametric approach compared to the parametric approach of CB-SEM. In fact the fitness of a model should not be taken lightly since it reflects the behavior of data in relation to the proposed model for the study. Furthermore, the adequacy of sample size and the normality of data are among the main assumptions of parametric test itself. This study intended to clarify the ambiguities among the social science community by employing the data-set which do not meet the fitness requirements and normality assumptions to execute both CB-SEM and VB-SEM. The findings reveal that the result of CB-SEM with bootstrapping is almost similar to that of VB-SEM (bootstrapping as usual). Therefore, the failure to meet the fitness and normality requirements should not be the reason for employing Non-Parametric SEM.

Covariance-Based Structural Equation Modeling in theJournal of Advertising: Review and Recommendations

“…Only a handful of the articles report reliance on an SEM with other estimation techniques, such as generalized least squares (GLS) or distribution-free estimation. The heavy reliance on maximum likelihood is justified based on the relative robustness of the approach (Awang, Afthanorhan, and Asri 2015;Hox and Bechger 1998).…”

Structural equation modelling assessment of personal and social factors on hajj crowding among african pilgrims in Mina

“…To ensure appropriateness, that statistical values of fitness for the measurement and structural model have to align within any three (3) fit index categories namely; absolute fit, incremental fit and parsimonious fit [59,61,62,64,65]. The minimum thresholds of indices is p-value ≥ 0.00, X2/df (CMIN) ≤ 2.0, GFI, AGFI, NNFI and CFI 0 no fit) to 1 (perfect fit), RMSEA<0.05 (very good fit); 0.05-08 (Fairly good fit); 0.08-0.1 acceptable; > 0.1(unacceptable) by previous researchers [61], [62], [64]. The statistics of the measurement model had a RMSEA value at 0.64, CFI at 0.934 and close to a perfect fit and CMIN of 1.627.…”

A Quantitative Analysis on Leisure Participation of Turkish Seafarers by Structural Equation Modeling

“…If the research"s data meet all the requirement of parametric assumptions, the finding will be meaningful by using the Covariance Based Structural Equation Modeling (CB-SEM) rather than Variance Based Structural Equation Modeling [54]. All parametric assumptions are satisfied for each variables of conceptual model.…”