Covariance Structure Models for Fixed and Random Effects

Fixed- and random-effects models are often used to control for unmeasured personspecific effects when examining the influence of measured covariates on an outcome. To date, different estimators are used for fixed- and random-effects models, making it difficult to choose between these two alternatives. In this article, the authors outline a covariance-structure model that captures both fixed and random effects and provides a simple test for comparing them.