Because validity is universally accepted as one of the most important aspects of sound
assessment practices, it is important that educators understand the concept and know how to use it as a quality control measure in their teaching. This paper outlines the evolving definition of the concept of validity, considers different types of evidence of validity and suggests how the
concept might be applied in outcomes-based education. It argues that the common view that test items are valid if they measure what they are intended to measure, provides a necessary, but insufficient basis for considering validity. The paper develops the idea that test item per se can never be valid or invalid and explains why it is necessary to focus on evidence from which valid inferences can be made about learning.