Validity is concerned with the clarification and justification of the intended interpretations and uses of observed scores. It has not been easy to formulate a general methodology set of principles for validation, but progress has been made, especially as the field has moved from relatively limited criterion-related models to sophisticated construct models. The watershed event in the development of validity theory has been the development of a well-articulated version of construct validity by L. Cronbach and P. Meehl (1955). The general principles they, and others, have articulated, that validation requires an extended analysis of evidence based on an explicit statement of the proposed interpretation and involving the consideration of competing interpretations, are applicable to all validity arguments. These principles fit into an argument-based approach to validation. The validity argument evaluated the plausibility of the proposed interpretation by examining the inferences and assumptions in the interpretive argument critically. The validity argument will typically involve different kinds of evidence and is most likely to be effective in improving the measurement procedure and its interpretive argument to the extent that it identifies the weak points in the interpretive argument. A proposed interpretation is most effectively evaluated by challenging its most questionable assumptions and thereby pitting it against the most plausible alternate interpretations of the observed scores. (Contains 44 references.) (SLD)