December 1, 2002
Psychometric Principles in Student Assessment
Authors:
Robert J. Mislevy, Mark R. Wilson, Kadriye Ercikan, and Naomi Chudowsky
In educational assessment, we observe what students say, do, or in a few particular circumstances make, and attempt to infer what they know, can do, or have accomplished more generally. Some links in the chain of inference depend on statistical models and probability-based reasoning, and it is with these links that terms such as validity, reliability, and comparability are typically associated—psychometric principles, as it were. Familiar formulas and procedures from test theory provide working definitions and practical tools for addressing these more broadly applicable qualities of the chains of argument from observations to inferences about students, as they apply to familiar methods of gathering and using assessment data. This report has four objectives: It offers a framework for the evidentiary arguments that ground assessments, examines where psychometric principles fit in this framework, shows how familiar formulas apply these ideas to familiar forms of assessment, and looks ahead to extending the same principles to new kinds of assessments.
Mislevy, R. J., Wilson, M. R., Ercikan, K., & Chudowsky, N. (2002). Psychometric principles in student assessment (CSE Report 583). Los Angeles: University of California, Los Angeles, National Center for Research on Evaluation, Standards, and Student Testing (CRESST).