Heungsun Hwang, McGill University

Structural equation modelling (SEM) is widely used to examine theory-driven relationships between constructs, such as self-esteem, depression, and socioeconomic status. These constructs are abstract concepts that are not directly measurable and are represented by entities linked to empirical data or observed variables in statistical models. This representation allows researchers to test hypotheses about their interrelationships. In SEM, constructs are typically represented as factors (latent variables) or as weighted composites of observed variables (components).

As psychology and many other sciences become interdisciplinary, there is an increasing need to consider distinct types of constructs simultaneously to understand human behaviour and cognition from more diverse perspectives. Some constructs can be represented as factors, while others as components. For instance, researchers are increasingly interested in examining how genetic variation or altered brain activity influences psychological constructs in cognition, personality, or mental disorders. Psychological constructs are typically represented as factors, whereas genetic or neuroimaging constructs, such as genes or brain regions, can be represented as components.

To address this need, I proposed an SEM method, integrated generalized structured component analysis (IGSCA), that enables researchers to estimate models that simultaneously include both factors and components. In this talk, I will discuss the conceptual background of IGSCA and demonstrate its potential through a real data application investigating the effects of multiple genes on depression severity. I will also briefly outline ongoing methodological extensions and illustrate how to implement the method using the free, user-friendly software GSCA Pro (www.gscapro.com).

about the speaker

Dr. Heungsun Hwang is a Professor of Psychology at McGill University, where he also completed his Ph.D. in Quantitative Psychology. His research focuses on developing and applying advanced quantitative methods to measure and analyze human characteristics, behaviors, and processes. His current work integrates statistics, psychology, and machine learning to incorporate individuals’ multifaceted information—such as psychological, physiological, imaging, and genetic data—to improve the understanding and prediction of behavioral and cognitive differences. He has served on the editorial boards of several journals, including Psychometrika, Psychological Science, Behaviormetrika, and the British Journal of Mathematical and Statistical Psychology.