Understanding validity and fairness issues in high-stakes individual testing situations

This policy brief discusses validity and fairness issues that could arise when test-based information is used for making “high stakes” decisions at an individual level, such as, for the certification of teachers or other professionals, or when admitting students into higher education programs and colleges, or for making immigration-related decisions for prospective immigrants. To assist test developers, affiliated researchers and test users enhance levels of validity and fairness with these particular types of test score interpretations and uses, this policy brief summarizes an “argument-based approach” to validation given by Kane.

This policy brief is based on a synthesis of conference proceedings and review of selected pieces of extant literature. To that synthesis, the authors add practitioner-friendly examples with their own analysis of key issues. They conclude by offering recommendations for test developers and test users.

The authors conclude that validity is a complex and evolving construct, especially when considering issues of fairness in individual testing contexts. Kane's argument-based approach offers an accessible framework through which test makers can accumulate evidence to evaluate inferences and arguments related to decisions to be made with test scores. Perspectives of test makers, researchers, test takers and decision-makers must all be incorporated into constructing coherent “validity arguments” to guide the test development and validation processes.

Standardized test use for individual-level decisions is gradually spreading to various regions of the world, but understandings of validity are still uneven among key stakeholders of such testing programs. By translating complex information on test validation, validity and fairness issues with all concerned stakeholders in mind, this policy brief attempts to address the communication gaps noted to exist among these groups by Kane.