Disaggregating Person- and Situation-Specific Heterogeneity: a Categorical Latent State-Trait Model

Disentangling heterogeneity in psychological constructs remains vital for identifying homogeneous subgroups that hold practical utility and theoretical importance. The increased popularity of intensive longitudinal designs enables the use of novel statistical methods for identifying classes that incorporate considerations for temporal dynamics. Despite advances in latent state-trait theories and methods, traditional methods have rarely differentiated between situation-specific and person-specific classes. The current study describes a novel latent state-trait model with a discrete state and a discrete trait using Bayesian estimation. An artificial data example illustrates parameter estimation and inference in a large sample. A real data example follows to demonstrate model interpretation. The proposed model can account simultaneously for heterogeneity in stable traits as well as transitory states, given a continuously measured observed variable across multiple persons and multiple occasions. Theoretically, differentiating between time-varying and time-invariant components of class membership can harness the rich information in longitudinal data to disaggregate situation- and person-specific heterogeneity. Methodologically, our model complements existing methods for identifying mixture structure in latent states and traits.

The online version contains supplementary material available at 10.1007/s42113-023-00181-6.