ABSTRACT

Structural Equation Modelling (SEM) currently enjoys widespread popularity in the behavioural, educational, medical and social sciences. Contributions to the literature in terms of books, book chapters and journal articles applying SEM or developing new SEM methodology are appearing at an incredible rate. There is even a scholarly journal devoted exclusively to SEM, entitled Structural Equation Modeling: A Multidisciplinary Journal. According to one review of the literature, SEM is the fastest growing and dominant multivariate statistical technique of the past two decades (Hershberger 2003). A major reason for this popularity is that SEM permits researchers to study complex multivariate relationships among observed and latent variables whereby both direct and indirect effects can be evaluated. Another reason for this popularity is the availability of specialized SEM programs. For example, programs such as Amos (Arbuckle 2006), EQS (Bentler 2004), LISREL (Jöreskog and Sörbom 1996), Mplus (Muthén and Muthén 2006), Mx (Neale et al. 1999), SAS PROC CALIS (SAS Institute 1989), SEPATH (Steiger 1995) and RAMONA (Browne and Mels 1990) are all broadly available for the analyses of various models.