Not infrequently, investigators assume that reliability for groups is greater than reliability for persons, or that the error variance for groups is less than that for persons. Using generalizability theory, it is shown that this "conventional wisdom" is not necessarily true. Examples are provided from the course-evaluation and the…

Results and interpretations of the data from a performance examination were compared for four methods of analysis for 74 medical specialty certification candidates: (1) traditional summary statistics; (2) inter-judge correlations; (3) generalizability theory; and (4) the multifaceted Rasch model. Advantages of the Rasch model are outlined. (SLD)

Recently, there has been a significant expansion in the use of performance assessment in large scale testing programs. Although there has been significant support from curriculum and policy stakeholders, the technical feasibility of large scale performance assessments has remained a question. This report is intended to contribute to the debate by…