Nonprofit Leader Guide· a Bicycle Guide

Coming soon · Book Profile

Factor Analysis Sem Joreskog

A collection of foundational papers by Karl Jöreskog and Dag Sörbom that establishes a general statistical framework (LISREL) for confirmatory factor analysis and structural equation modeling, enabling researchers to specify, estimate, and test complex causal models involving latent variables and measurement error.

A profile of this book is on the way.

Get the book →

What it’s about

This book compiles the seminal articles by Karl Jöreskog and Dag Sörbom that revolutionized the analysis of nonexperimental data in the social sciences. It moves beyond traditional exploratory methods to a powerful confirmatory approach for causal modeling, known as the LISREL framework. Readers will learn how to specify, estimate, and test complex structural equation models that incorporate latent variables (unobserved constructs), measurement error, and reciprocal causation. The book provides the theoretical underpinnings and practical applications for a wide range of problems—from confirmatory factor analysis and the study of group differences to the analysis of longitudinal data—equipping researchers with the tools to build more precise, testable theories and gain a deeper understanding of the causal relationships embedded in their data.

The through-line

Who it’s for
A social or behavioral science researcher, student, or data analyst who uses quantitative methods and wants to move beyond simple descriptive statistics or traditional regression to test complex causal theories and better understand the relationships in their nonexperimental data.
The problem
The researcher's current statistical tools (like regression, ANOVA, or exploratory factor analysis) are inadequate for testing complex theories that involve unobservable constructs, measurement error, and reciprocal causation. They feel frustrated that their methods are a poor match for their rich theories, leading to biased estimates, ambiguous interpretations (e.g., factor rotation), and an inability to rigorously test their hypotheses, thus undermining their confidence in making causal claims from observational data.
The plan
  1. Master the principles of confirmatory factor analysis to specify and test robust measurement models.
  2. Learn the general LISREL framework that integrates measurement models with a structural model of causal relationships among latent variables.
  3. Apply the framework to advanced applications, including the analysis of longitudinal data and the rigorous comparison of different groups.
The payoff
The reader can confidently specify, estimate, and test sophisticated causal models that were previously intractable. · Their research becomes more theoretically precise and empirically rigorous, leading to more impactful publications. · They can make stronger, more defensible causal claims and accurately estimate relationships between latent constructs, corrected for measurement error.

See our guide

Related profiles we’ve built

Additional reading