Quality Assurance in the Reporting of Structural Equation Path Models: A Reviewer’s Perspective based on LISREL Analysis

Abstract

Maintaining the transparency, reliability, and replicability of knowledge in research papers and manuscripts is critical to the credibility of scientific output. This paper emphasizes the importance of preserving the integrity and transparency of path model results from structural equation modeling (SEM), a key statistical tool for researchers. SEM offers researchers an advantage in testing hypotheses across variables with simple and complex multivariate relationships. However, this popularity could lead some researchers to report results that inaccurately reflect SEM outputs, possibly due to limited knowledge of SEM and/or the urgency to produce statistically significant results. Therefore, to maintain research integrity and transparency, editors, reviewers, and the research community should be aware of this potential issue. To address this challenge, results from SEM path analysis should be verified by examining the correlation and/or covariance matrices presented in the summary data. This paper illustrated this approach in LISREL using covariance-based SEM to analyze a path model from a published paper. This paper recommends this approach as the norm for users, prospective users, editors, reviewers, and consumers of SEM tools and outputs. Additionally, several key recommendations for maintaining the quality and integrity of structural path models reported in the manuscripts submitted for possible publication in journal outlets were outlined.