2004-05 Program on Latent Variable Models in the Social Sciences

Latent variables are widespread in the social sciences. Whether it is intelligence or socioeconomic status, many variables cannot be directly measured. Factor analysis, latent class analysis, structural equation models, error-in-variable models, and item response theory illustrate models that incorporate latent variables. This SAMSI program takes a broad look at latent variables and measurement error. Issues of causality, multilevel models, longitudinal data, and categorical variables in latent variable models are examples of the SAMSI topics for this program.

Program Leaders: Kenneth A. Bollen (Chair), James J. Heckman, Alan F. Karr, and Susan A. Murphy