This paper was presented at the annual meeting of the Psychometric Society at the University of California at Los Angeles, July 1989. It describes an approach to assessing the reliability and validity of Likert-type ratings. Latent structure models for Likert-type ratings previously developed in other contexts can be adapted to the problem of rating agreement. This approach can be viewed as a special case of latent class analysis, where parameter estimation is simplified by the application of a heuristic model that imposes a desirable structure on parameters. Several advantages derive from this approach, including Bayesian models for the interpretation of multiple ratings, adaptive rating schemes, and the ability to determine optimal rating panel size.