Categorical Data Analysis with Structural Equation Models: Applications in Mplus and Lavaan

Multivariate categorical outcomes, such as Likert scale responses and disease diagnoses, require specialized structural equation modeling (SEM) software to be analyzed properly. Providing needed skills for applied researchers and graduate students, this book leads readers from regression analysis with categorical outcomes to complex SEMs with latent variables for categorical indicators. The initial section sets the stage by demonstrating regression analyses for binary, ordered, or count outcomes using R. Chapters then reanalyze the same data using Mplus and R lavaan to show how univariate models for categorical outcomes can be estimated and interpreted with SEM programs. Subsequently, the book turns to multivariate models, discussing path models, confirmatory factor models, and latent variable path models with categorical outcomes. Concluding chapters cover advanced SEM with categorical outcomes, including growth models, latent class models, and survival models. Worked-through examples are featured throughout. The companion website provides R (including lavaan), Mplus, and SAS code, as applicable, for the examples.

Kevin J. Grimm, PhD, is Professor of Psychology at Arizona State University. His research interests include multivariate methods for the analysis of change, multiple group and latent class models for understanding divergent developmental processes, categorical data analysis, machine learning techniques for psychological data, and cognitive/achievement development. Dr. Grimm teaches graduate quantitative courses, including Longitudinal Growth Modeling, Machine Learning in Psychology, Structural Equation Modeling, Advanced Categorical Data Analysis, and Intermediate Statistics. He has also taught workshops sponsored by the American Psychological Association’s Advanced Training Institute, Statistical Horizons, Instats, Stats Camp, and various departments and schools across the country.