ABSTRACT

This chapter considers models that postulate specific explanatory relationships among latent constructs as structural regression (SR) models. In terms of their path diagrams, SR models have at least one path leaving a putative explanatory latent variable and ending at another construct. The chapter presents an example to demonstrate SR model which focuses on two fluid intelligence subabilities, induction and figural relations. Induction relates to one's capability to reason using analogies and rules of generalization to more comprehensive contexts. Figural relations pertains to one's ability to see patterns of relationships between parts of figures, mentally rotate them, and also use forms of inductive reasoning with figural elements. Imposing the condition of invariance deals with the requirement that the same construct be measured at all occasions, and in particular that its structure remains the same. Since a necessary condition for measuring the same constructs at repeated assessments is the identity of the associated factorial structure, the term factorial invariance is commonly used.