ABSTRACT
Structural equation modeling (SEM) is a broad designation that encompasses a number of different techniques with which researchers may model the relationships among different variables that are observed (e.g., IQ score, GPA, years in school) and/or unobserved (e.g., motivation, need for achievement, academic self-concept). SEM may also be referred to as covariance structure analysis, latent variable analysis, causal modeling, and simultaneous equation modeling. This chapter will introduce three SEM techniques, including observed variable path analysis, confirmatory factor analysis (CFA), and latent variable path analysis. The presentations in this chapter will be more conceptual than statistical and, thus, will not include the respective system of equations for the models presented. Readers interested in learning more about the equations associated with each SEM technique are encouraged to consult Bollen’s (1989) text.
