When and how to use set-exploratory structural equation modelling to test structural models: A tutorial using the R...

Area of Science:

• Psychometrics
• Statistical Modeling

Background:

• Confirmatory Factor Analysis (CFA) is a standard for measurement models.
• Exploratory Structural Equation Modeling (ESEM) extends CFA but can face estimation challenges in complex structural models.
• Set-ESEM was developed to balance ESEM flexibility with CFA's practicality.

Purpose of the Study:

• To demonstrate the application and advantages of set-ESEM over full-ESEM and CFA for structural models.
• To provide practical guidance and code for implementing set-ESEM using the R package lavaan.
• To compare the performance of set-ESEM with CFA-based structural models using real-world data.

Main Methods:

• Utilized two applied examples with real data, avoiding simulation studies.
• Employed the lavaan R package to provide practical code for set-ESEM implementation.
• Compared goodness of fit, factor correlations, and path coefficients between set-ESEM and CFA models.

Main Results:

• Set-ESEM structural models exhibited better goodness of fit and more realistic factor correlations compared to CFA.
• Path coefficients in set-ESEM models were more accurate, with some previously non-significant effects becoming significant.
• Set-ESEM models suggest a lower Type II error rate due to more precise parameter estimates.

Conclusions:

• Set-ESEM provides a valuable and practical approach for analyzing structural models, outperforming traditional CFA.
• The method enhances model fit and parameter accuracy, leading to more reliable statistical inferences.
• Researchers are encouraged to consider set-ESEM for complex structural modeling tasks to improve analytical outcomes.