Categorical Structural Equation Modeling

Structural equation modeling (SEM; also known as covariance structure analysis or latent variable analysis) belongs to the family of two statistical models and techniques of multivariate data analysis, based on interdependence technique (factor analysis [FA]) and dependence (multiple regression analysis [MRA]). It represents a unique combination, in which FA explores the interdependence of observed variables (indicators, items) and develops theoretical constructs (latent—or unobserved—variables), while the MRA (i.e., SEM) explains their mutual relationships. In particular, SEM aims to test the hypothesized causal relationships between the dependent (exogenous) latent variables and independent (endogenous) latent variables that are reflected in theoretical constructs and developed in the process of operationalization, based on data derived from, for example, tests, measurement scales, self-reports, or questionnaires. Examples of theoretical constructs include intelligence, ...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles