| dc.description.abstract | The latent state-trait theory posits that a psychological construct may reflect stable influences specific to a person (i.e., trait), ephemeral influences from situations (i.e., state), and interactions between them (i.e., state-trait interactions). Researchers conventionally apply mixture modeling to explore heterogeneity in variables by identifying homogenous classes with respect to the measured variable; however, they rarely distinguish between person- and situation-specific classes. My dissertation introduces novel categorical latent state-trait models to identify subgroups in states and traits, quantifying the effects of person-specific classes, situation-specific classes, and person-situation interactions. The proposed models are applied to an empirical dataset of 174 low-income mothers with children in Early Head Start programs in North Carolina or New York, United States. Depressive symptoms were measured using the Center for Epidemiological Studies-Depression at baseline, 14 weeks, 22 weeks, and 26 weeks. Using these models, I demonstrate how to conduct statistical inference, estimate effect size, and visualize the results. Based on realistic parameter values from the empirical dataset, three simulation studies were conducted to investigate model performance in terms of relative bias, coverage rates for estimated parameters, as well as clustering accuracy for latent state and trait class identification. Proposed models complement existing mixture modeling approaches by distinguishing between person-specific and situation-specific classes from data on multiple persons across multiple measurement occasions. Bayesian estimation in the proposed models can conveniently account for missing data and test a wide range of hypotheses related to state, trait, and interaction effects. Limitations of the proposed models suggest future research directions, such as accommodating multiple observed variables and relaxing the assumption of homogeneity of variance across state- trait combinations. | |
