Making Measures

Ben Wright and Mark Stone’s insights into the philosophical and mathematical foundations of measurement are necessary for constructing practical, objective assessments of educational, psychological, and physical functioning. Since the publication of their seminal text, Best Test Design (1979), they have collaborated on a number of works designed to make accessible the principles and procedures of objective measurement. This volume widens yet again the audience that can benefit from these ideas. By beginning with a disarmingly simple stone-lifting problem the reader is introduced step by eloquent step to the thinking and mathematics behind making measures and making measures meaningful. This volume looks back at the philosophical underpinnings of measurement and connects the prescience of Charles S. Peirce (1838–1914) to the models of Georg Rasch (1901–1980) and finally to the day-to-day needs of instrument-builders in the human sciences.

From the perspective of Best Test Design, this volume also moves forward with an emphasis on the role of substantive theory in measurement. The measurement applications, the Knox Cube Test for attention and short term memory (Chapter 7), Wolpe and Lang’s (1964) Fear Survey Schedule (Chapter 8), and the Lexile Framework for Reading (Chapter 9), make extensive use of theory in predicting why item difficulties vary. Can we imagine a more important piece of evidence for the construct validity of an instrument than a specification equation that is capable of explaining variation in observed item difficulties?

The paradox of unity and separation finds expression in the puzzle of whether a book is well comprehended by a reader because the book is easy or because the reader is skilled. Does a person succeed on an attention task because it requires fewer memory registers or because the person possesses good attention? Is a stone lifted because the stone is light or the lifter is strong? This paradox presents in various guises across the human sciences or more generally wherever measurement is contemplated. The paradox is resolved by positing a single yardstick of reading ability/readability, attention/difficulty, or strength/weight. This book explains how and why such yardsticks are built and how to ensure quality in their construction.

From the Introduction by A. Jackson Stenner