Tutorial on Structural Equation Modeling in Medical Research: A Review Article

Document Type : Review

Authors

1 Center for healthcare Data modeling, Departments of biostatistics and Epidemiology, School of public health, Shahid Sadoughi University of Medical Sciences, Yazd, Iran

2 PhD student, Department of Health Services Management, School of Health Management and Information Sciences, Iran University of Medical Sciences, Tehran, Iran

Abstract

Structural equation modeling (SEM) as a quantitative approach can greatly help the researcher in the analysis of multivariate experimental data. This issue can be more significant in medical research. The results of SEM are expected to be more realistic than the results of regression analysis. SEM is a combination of measurement and structural models. The measurement model in SEM is divided into two types of reflective indicators and constructive indicators. Fitting indicators of measurement models (external model) include index reliability, convergent validity and divergent validity. Structural models indices (internal model) include Significant numbers t (T - values), Coefficient of determination (R2), The effect size (f2), Predictive Relevance (Q2 criterion), Quality Index, Communality Criterion (sharing index) and Redundancy Criterion. Overall fit indicators also include GOF criterion. In the present study, various methods and indicators of structural equation modeling have been developed in full detail, which can help researchers in conducting their studies in the field of structural equations.

Keywords