Lissitz and Samuelsen (2007) have proposed an operational definition of validity that shifts many of the questions traditionally considered under validity to a separate category associated with the utility of test use. Operational definitions support inferences about how well people perform some kind of task or how they respond to some kind of stimulus. Such interpretations can be useful for some purposes, but they do not support any richer interpretations of scores. The author argues that validation should involve an evaluation of the proposed interpretations and uses of test scores. Test developers may choose to adopt a narrow operational interpretation of test scores, but if they do so, they should label the test accordingly.
Stored only when both papers are in the corpus — 67 citations exist upstream.