Structural Equation Modeling: Applications Using Mplus

2.1 Basics of CFA Model

CFA is often used to determine and confirm the factorial structure of an already developed measuring instrument in an application among a target population. In other words, CFA tests whether the theoretically defined or hypothesized factorial structures of the scales in the measuring instrument under study are valid. If the hypothesized CFA model fits data, we confirm the factorial structure is valid for the population. This is called testing for factorial validity of the measuring instrument (Byrne, 2006). In this section, we discuss some basics of the CFA model with an example of a well- known psychiatric measuring instrument, the Brief Symptoms Inventory 18 (BSI- 18) (Derogatis, 2000, 2001).

The BSI- 18 is a shorter version of the instrument Brief Symptoms Inventory 53 (Derogatis, 1993; Derogatis and Spencer, 1983). The BSI- 53 is widely used to assess psychiartric disorders in clinical and nonclinical populations. It has good psychometric properties including high internal consistency and test– retest reliability (Derogatis and Spencer, 1983; Derogatis, 1993). BSI- 53 has nine well- defined psychometric subscales. However, the nine subscales are usually computed as composite scores, and their factorial structure has not been confirmed using factor analyses (Boulet and Boss, 1991; Ruiperez et al., 2001). As a result, Derogatis (2000) developed a shorter version of the instrument – the BSI- 18 – to be used as a screening tool for the most common psychiatric ...