Model Setting and Interpretation of Results in Research Using Structural Equation Modeling: A Checklist with Guiding Questions for Reporting

Abstract

This study develops a checklist with guidelines for the methods and important factors to consider in research using structural equation modeling (SEM). Method: The paper discusses the factors to consider in the process across the three stages of 1) model setting, 2) model evaluation and modification, and 3) interpretation and reporting of SEM-based studies. Results: The authors present a checklist for researchers during the stages of model setting, model evaluation and modification, result analysis, and reporting, along with examples of figures and tables with explanations. Conclusion: A checklist will help to improve the reporting quality of SEM-based studies.

Choosing a package carrier during COVID-19 pandemic: An integration of pro-environmental planned behavior (PEPB) theory and service quality (SERVQUAL)

“…This research analyzed the collected data using the multivariate analysis, specifically structural equation modeling (SEM). There are many benefits in applying SEM that it became one of the most used data analysis tools in various studies, some of which include control on measurement errors, utilization of mediating variables, and statistical evaluation of the theoretical model ( Kang and Ahn, 2021 ). Other studies cited that SEM is also convenient to use when testing a research theory ( Savari and Gharechaee, 2020 ) and when modeling the cause-and-effect relationship of different exogenous and endogenous factors/variables of the theoretical hypothesis ( Yin and Huang, 2021 ).…”

Structural equation modelling of COVID-19 knowledge and attitude as determinants of preventive practices among university students in Ghana

“…On the other hand, latent variables are unobserved variables that cannot be measured directly. The theory of SEM simplifies complex relationships between variables by utilizing a path model or analysis for explaining effects resulting from observed and latent variables [ 19 , 20 ]. The technique clearly determines the relationship between cause-and-effect variables.…”

Underestimated Risk Perception Characteristics of Drivers Based on Extended Theory of Planned Behavior

“…In the third step, the influence model of the underestimation of the driving risk was tested using the SEM, which is a method for establishing, estimating, and testing causal relationships between variables [ 23 ]; it comprises a measurement model and a structural model. Compared to conventional methods, the SEM has controlled measurement error and allows a statistical evaluation of the theoretical models [ 46 , 47 ]. In addition, studies have shown that SEM can help build more accurate models for driving behavior analyses [ 48 ].…”