Structural Equation Modelling in the biomedical domain: Concepts and methods revisited with the help of R and SAS statistical package

Structural Equation Modelling (SEM) is widely used in the social sciences [1, 2] analysing complex relationships among multiple variables. It is defined by two models measurement model and construct model. Measurement model contains defines the relationship observed variables and constructs or latent variables or factors or unobserved variables which are derived from observed variables. Construct model defines the relationship between the construct or latent variables. SEM is able to model the error terms in the observed variables and error while defining the relationship in the model and it is a unique property of SEM. The objective of the SEM is to verify whether a model specified prior is a best fit to the given data or not. This paper provides an overview of Structural Equation Modelling, its application in biomedical domain and illustrated with the help of R Statistical Software