Abstract

Developing and evaluating clinical outcome assessments (COAs) requires a framework for understanding validity. The validity framework reflected in the most recent draft guidance from the US Food and Drug Administration is the argument-based approach. In this approach, a researcher should state how they would like to interpret or use scores from some measure, identify key assumptions that need to be true for the proposed interpretation/use to be justified, and evaluate evidence for or against those key assumptions. If the collection of assumptions, known as the rationale, has convincing evidence, then a decision is made that the proposed interpretation or use of scores is valid. In this article, I briefly review how this approach to validity that has been developed within educational and psychological testing has recently made its way into COAs. I then discuss several notable features of the argument-based approach that have implications for how COAs are developed and evaluated.