Interrelationships Between Latent State-Trait Theory and Generalizability Theory Within a Structural Equation Modeling Framework

Over recent years, latent state-trait theory (LST) and generalizability theory (GT) have been applied to a wide variety of situations in numerous disciplines to enhance understanding of the reliability and validity of assessment data. Both methodologies involve partitioning of observed score variation into systematic and measurement error components. LST theory is focused on separating state, trait, error, and sometimes method effects, whereas generalizability theory is concerned with distinguishing universe score effects from multiple sources of measurement error. Despite these fundamental differences in focus, LST and GT share much in common. In this article, we use data from a widely used personality measure to illustrate similarities and differences between these two frameworks and show how the same data can be readily interpreted from both perspectives. We also provide comprehensive instructional online supplemental materials to demonstrate how to analyze data using the R package for all LST models and GT designs discussed. Translational Abstract This article is intended for researchers, practitioners, instructors, and students who wish to deepen their understanding of latent state-trait theory (LST) and generalizability theory (GT) and how the theories align and differ. We accomplish this by describing both theories within a structural equation modeling framework and showing how common GT designs can be considered special cases of LST. Although the same data can be analyzed from both perspectives, labeling of effects, interpretation of indices, key units of analysis, and importance of model fit indices differ across the frameworks. We illustrate examples of LST and GT analyses using both first- and second-order factor models and expand LST models to allow for additional factors and correlated method effects. In separate online supplemental materials, we provide detailed instruction and computer code in R for analyzing all models discussed and testing assumptions underlying the models using data from the conscientiousness subscale of the Big Five Inventory (BFI; John et al., 1991).