In Mplus Program 3.2, we have demonstrated how to test interaction between two observed variables by creating a new variable, which is the product of the two observed variables. This approach is called product indicant. In this section we will discuss and demonstrate how to test interactions that involve latent variables.^{6} Testing interactions involving latent variables has been a challenge. Fortunately, Mplus has made this mission possible and easy. Here, we will limit our discussion to interactions that involve only continuous latent variables. Some applications of testing interactions involving categorical latent variables (latent class variables) are available in Appendix 6.A, in which interaction between a covariate and the baseline latent class variable in a latent transition analysis (LTA) model is demonstrated.

The model shown in Figure 3.6 is a modification of the model shown in Figure 3.4. In the current model, the interaction involves the latent variable ANX () and the observed endogenous variable Crack (). The interaction between and is shown in Figure 3.6 as a filled circle. The Mplus program for the model is shown below. ...

Start Free Trial

No credit card required