Chapter 3: Validating the Interpretations and Uses of Test Scores

In most discussions of educational and psychological measurement, validity has been defined broadly, to cover the evaluation of interpretations and uses of test scores. If the proposed interpretation of test scores is limited, as it is for some observable attributes, the requirements for validation can be very modest. If the proposed interpretations are more ambitious, as they are for traits and theoretical constructs, more evidence and more kinds of evidence are required for validation. Because validity addresses the wide range of issues associated with the interpretation and uses of test scores, it is viewed as the most fundamental consideration in evaluating testing programs. We can redefine validity as a more specific, technical property of tests or test scores, but if we do so, we should probably not think of it as the most fundamental consideration in evaluating testing programs.