This paper describes a model that integrates an item response theory (IRT) Rasch model and a hierarchical linear model and presents a method of estimating model parameter values that does not rely on large-sample theory and normal approximations. The model resulting from the integration of a hierarchical linear model and the Rasch model allows one to estimate all model parameters simultaneously and thus incorporate the standard errors of the latent trait estimates into the total variance of the model. The Rasch hierarchical measurement model (HMM) can allow one, for example, to model the variances of person-level and school-level error while estimating latent trait parameters of student ability estimates or student attitudes from student responses to a questionnaire of dichotomous items. Two different simulated data sets, both having a hierarchical structure, were created to illustrate the Rasch HMM. To illustrate how the Rasch HMM performs relative to a traditional two-step approach, the simulated balanced data set was reanalyzed. Results show that the Rasch HMM is very specialized because it is only appropriate for dichotomous responses and does not allow the incorporation of any level-1 or level-2 covariates. The usefulness of hierarchical measurement models hinges on the degree to which these models can be generalized. (Contains 4 figures, 6 tables, and 29 references.) (SLD)