Five key features of a new approach to assessment design serve as the organizing themes for this chapter:

  • A model of cognition and learning, or a description of how people represent knowledge and develop competence in a subject domain, is a cornerstone of the assessment development enterprise. Unfortunately, the model of learning is not made explicit in most assessment development efforts, is not empirically derived, and/or is impoverished relative to what it could be.

  • To increase the chances of collecting evidence that supports the types of inferences one wants to draw, the design and selection of assessment tasks, along with the procedures for evaluating students’ responses, should be guided jointly by the cognition and interpretation elements of the assessment triangle. An assessment should be more than a collection of tasks that work well individually. The utility of assessment information can be enhanced by considering how to design and/or select tasks so that the information derived from them can be combined to support the desired inferences.

  • The process of construct validation during test design should rest, in part, on evidence that tasks actually tap the cognitive content and processes intended.

  • Although reporting of results occurs at the end of an assessment cycle, assessments must be designed from the beginning to ensure that reporting of the desired types of information will be possible and effective. When results are reported, well-delineated descriptions of learning in the domain are key to their effectiveness for communicating about student performance.

  • Fairness in testing is defined in many ways, but at its core is the idea of comparable validity: a fair assessment is one that yields comparably valid inferences from person to person and group to group. One way of thinking about fairness is to take into account individual learners’ instructional histories when designing an assessment.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement