A structural equation model (SEM) provides a unified statistical framework for analyzing complex relationships among variables, especially when key variables cannot be measured without error. In this Statistically Speaking, Ken Bollen introduces SEMs to an interdisciplinary audience familiar with statistics and regression analysis but new to this approach.

Because SEMs apply to both experimental and observational data, they are widely useful across disciplines. This talk covers model estimation, evaluation of fit, and interpretation of coefficients, while highlighting the flexibility of SEMs for diverse research questions and data types.

Best suited for those interested in:

• How to improve empirical research by addressing measurement error and capturing complex causal structures.
• Understanding the intuition and core concepts of SEMs, including latent variables, measurement error, and causal pathways with mediation.
• How SEMs can prove a valuable resource when analyzing complexities.Meet the speaker

Ken Bollen
Professor, Department of Psychology and Neuroscience and Department of Sociology, University of North Carolina at Chapel Hill

Ken Bollen is the Henry Rudolph Immerwahr Distinguished Professor in the Department of Psychology and Neuroscience and Department of Sociology at UNC-Chapel Hill. He is a faculty member in the Quantitative Psychology Program in the Thurstone Psychometric Laboratory. Bollen is a Fellow at the Carolina Population Center, the American Statistical Association, and the American Association for the Advancement of Science. He was the Director of the Odum Institute for Research in Social Science from 2000-2010.

He has authored several books and over 200 papers, which have been widely cited over the years. His best-known publication, Structural Equations with Latent Variables, has been cited more than 43,000 times. It integrated a diverse body of literature from several disciplines and helped define the area of structural equation modeling (SEM), an area in which he is considered a world-renowned expert. Bollen’s ground-breaking work has resulted in many respected awards, including the Psychometric Society’s Career Award for Lifetime Achievement, the Lazarsfeld Award for Distinguished Contributions to Sociological Methodology, and the Sells Award for Distinguished Multivariate Research. In 2025, he was elected to the American Academy of Arts and Sciences.