chapter  14
Toward a Science of Assessment
WithIsaac I. Bejar
Pages 20

This chapter briefly reports on recent efforts in developing a science of assessment. Referring to a science of assessment does not imply that current assessments are not scientifically based. Rather, it implies that, currently, a clearly articulated and general approach for designing and implementing assessments is lacking. It seems that in the past, assessments started from scratch or reinvented aspects of other assessments that could easily have been reused had we shared a common vision of the assessment design process. A science of assessment would allow us to view specific assessments as instances of wellunderstood components of any assessment. Such an approach to assessment design can have deep implications for the scientific standing of an assessment instrument but this is not to suggest that instruments designed in this manner would be exempted from the principles that we know as validation. Indeed a turning point in the road toward a science of assessment occurred with the publication of the Cronbach and Meehl (1955) paper on construct validity. They noted, “Construct validity calls for no new scientific approach. Much current research on tests of personality…is construct validation, usually without the benefit of a clear formulation of this process” (p. 282, italics added).