Measures of Goodness of Fit for Group-Based Trajectory Models

Daniel S. Nagin, Carnegie Mellon University

This paper describes a series of diagnostics for measuring the adequacy of model fit for a group-based trajectory model. While much has been written on the determination of the optimal number groups in finite mixture modeling, which forms the analytical basis for group-based trajectory modeling, little attention has been given to developing criteria for judging the adequacy of the selected model. Based on simulated data, this paper proposes four diagnostics for judging model adequacy. The diagnostics focus on the accuracy of the estimates of the probability of group membership and the precision of assignments of population members to their correct group.

(Return to Program Resources)

Updated 05/20/2006