Factor Analysis and Structural Equation Modeling

Structural equation models (SEMs) simultaneously combine confirmatory factor analysis (CFA) and path analysis. SEM includes a measurement model, which creates latent constructs from observed variables, and a structural model, which assesses relationships between variables. The ability to create measures of latent constructs has additional benefits, one being increased validity of the measure of interest. All measures comprise two forms of measurement error: random and systematic. CFA requires an exact number of hypothesized factors and the arrangement of indicators across factors to be specified prior to the analyses. The structural portion of an SEM relies on a method of multiple regression where systems of relationships are specified between multiple independent and dependent variables. For criminologists, thoughtfully applied SEM can be one way to combine creativity and analytical rigor, a pairing essential for unraveling the complexities of human behavior.