Errors and biases in Structural Equation Modelling
Jelte Wicherts (Tilburg University, The Netherlands)
Like any statistical procedure, the application of Structural Equation Modelling (SEM) is sensitive to errors and biases. Here I discuss the most likely errors and biases in applying SEM and reporting of SEM results. I present the results of a study of reporting errors with confirmatory factor analyses in the literature. I discuss the implications and highlight potential solutions. The latter includes the mandatory reporting of correlation/covariance matrices when applying SEM and syntaxes as part of supplementary files, which would allow for verification of the SEMs in published reports.