ABSTRACT
Samuel Messick wrote that “validity, reliability, comparability, and fairness are not just measurement issues, but social values that have meaning and force outside of measurement wherever evaluative judgments and decisions are made.” This chapter re-examines reliability, validity, comparability, and generalizability from a sociocognitive perspective, given the re-interpretation of measurement models. (Fairness gets its own two chapters.) The discussion includes recent work on validation, a conception of reliability as in terms of inference through an exchangeability structure, and the distinction between comparable data and comparable evidence.
