5: Validity

Validity assessment is both at the core of measurement theory and perhaps the most difficult aspect of applied measurement and psychometrics to investigate. Quite simply, the validation of an instrument involves determining the extent to which obtained scores represent the intended measured trait (e.g., academic achievement, motivation) and building an argument based on theory and empirical evidence to support inferences. Early in the development of measurement theory, validation was treated as primarily a statistical problem involving the correlation of a target measurement to be validated with a criterion that was known, or believed strongly, to be associated with the construct. Later, this statistical methodology extended to the use of factor analysis to investigate whether an instrument’s internal structure based on obtained data corresponded to its theoretical model. For example, empirical questions pertaining to whether a measure of reading aptitude was three dimensional as theory said it should be could be investigated. Under this paradigm, therefore, discussion of validity assessment was primarily the discussion of statistical methodology.