Publications & Resources

Comparability Across Assessments: Lessons From the Use of Moderation Procedures in England

Jan 1994

Elizabeth Burton and Robert L. Linn

Although there is considerable interest in developing a system of performance-based examinations in the United States, there is a general lack of agreement on how to compare the results of different performance assessments to sets of common national standards. This paper addresses the “comparison” problem, drawing on two major approaches used in England, moderation by inspection and statistical moderation, to link performance assessments to sets of common standards. While the United States will probably not develop a program of assessments exactly like those used in England, it is likely that the procedures used to compare the assessments will be similar. “Currently English secondary school exams in various subjects are developed and administered by nine examination boards,” wrote Burton and Linn. “Individual schools are free to choose the examination board that best fits their standards. While local control and high quality of assessments are maintained, the comparison of scores across the boards is problematic,” added the researchers. In this report, Linn and Burton discuss the advantages and problems of moderation by inspection and statistical moderation, together with an explanation of why neither approach is satisfactory by itself. The authors concluded that some combination of the two approaches may be necessary. “Neither a pure moderation by inspection nor a strict statistical moderation system is likely to meet this [link between assessments and standards] need,” wrote Burton and Linn. “It seems more likely that some sort of hybrid system will be required.”

Burton, E., & Linn, R. L. (1994). Comparability across assessments: Lessons from the use of moderation procedures in England (CSE Report 369). Los Angeles: University of California, Los Angeles, National Center for Research on Evaluation, Standards, and Student Testing (CRESST).