ABSTRACT
This chapter will introduce the main ideas of confirmatory factor analysis and structural equation modelling with latent variables. As will become clear from the discussion later, confirmatory factor analysis (CFA) is mainly concerned with testing hypotheses about the values of factor loadings (usually, that some are zero) whereas structural equation modelling (SEM) is concerned with estimating (linear and non-linear) relationships between factors (latent variables). The reader should not be deterred by noticing that many of the models discussed in this chapter are more complicated than those that have gone before. An appropriate software package will take care of all the technicalities. Later versions of SEM software packages even allow one to specify the model graphically rather than in terms of equations. As always our emphasis here will be on the formulation of the model and the interpretation of the results – not on the mathematical analysis which lies between the two.
