- Subject index
Validity is the hallmark of quality for educational and psychological measurement. But what does quality mean in this context? And to what, exactly, does the concept of validity apply? These apparently innocuous questions parachute the unwary inquirer into a minefield of tricky ideas. This book guides you through this minefield, investigating how the concept of validity has evolved from the nineteenth century to the present day. Communicating complicated concepts straight forwardly, the authors answer questions like: What does ‘validity’ mean? What does it mean to ‘validate’? How many different kinds of validity are there? When does validation begin and end? Is reliability a part of validity, or distinct from it? This book will be of interest to anyone with a professional or academic interest in evaluating the quality of educational or psychological assessments, measurements and diagnoses.
Chapter 5: The Deconstruction of Validity: 2000–2012
The Deconstruction of Validity: 2000–2012
The earliest definition of validity seemed fairly straightforward: the degree to which a test measured what it was supposed to measure. From the mid-1950s, three main kinds of validity were recognized: content, criterion and construct. Content validity and criterion validity still seemed fairly straightforward; if anything, they now seemed more straightforward than their parent concepts, logical validity and empirical validity. Construct validity was the exception, having been invented to deal with exceptional circumstances (when neither content validity nor criterion validity could be relied upon as a gold standard), and having been grounded in (as then) state-of-the-art philosophical thinking. From the mid-1970s, it was acknowledged increasingly that both content validity and criterion validity were over-simplistic, and that neither constituted a gold standard of any sort. In due course, ...