In response to Lissitz and Samuelsen (2007) , the author reconstructs the historical arguments for the more comprehensive unitary concept of validity and the principles of scientific inquiry underlying it. Her response is organized in terms of four questions: (a) How did validity in educational measurement come to be conceptualized as unitary, and why? (b) What is construct validity, and how does it provide the basis for a unitary concept of validity? (c) Why has the focus of validity been on the interpretations and uses of test scores rather than on the test itself? and (d) What sort of guidance for test developers and evaluators has been provided within a unitary concept of validity, and how might it be enhanced? The author highlights the role that cases of programmatic validity research can play in representing validity theory and guiding validity inquiry.
