ABSTRACT
The concepts of validity and validation of tests and test scores have implications that could require a survey of all possible uses of tests, because all have been considered relevant to validity. If there is a single meaning for the “validity of a test score,” it is captured in the defining statement, “a test score is valid to the extent that it measures the attribute of the respondents that the test is employed to measure, in the population(s) for which the test is used.” This can be restated in noun form: “The validity of a test score is the extent to which it measures ….” We can also say, “a test score is validated by evidence of the extent to which it measures ….” These are formal restatements of the old and often-repeated proposition: “A test is valid if it measures what it purports to measure.” 1 Recall, once more, that the word attribute carries its natural-language meaning—a quality, which may, according to Webster, be either inherent or incidental. Thus certainly it covers traits and states, and all examples given so far. We seek evidence to validate a given use of a test score. In principle, the evidence can come both from the original design of the test and from the intention and context of its use, which may be different from the original intent of the designer.
