Key Concepts in Measurement
Gillespie, David F., and Brian E. Perron. 2015. Key Concepts in Measurement. New York: Oxford University Press.
Measurement refers generally to the process of assigning a numeric value to, or ordering characteristics or traits of, entities under study. Measurement is necessary for building and testing theory, specifying problems, and defining goals. It is arguably one of the most important and diffcult tasks in social work research. Social work researchers who are not expert in developing, selecting, and using measures will not be able to contribute maximally to the social work knowledge base. Such knowledge and skills related to measurement ultimately determines the extent to which social work research can effectively inform social policy and social work interventions.
This book is to serve as a guide for developing, selecting, and using measures in social work research. In particular, this book provides a detailed review of contemporary validity theory; an update on the major issues of reliability; common errors in measurement of latent variables; and suggestions on measurement of social networks and collectives.
An important theme of this book is the focus on the creative potential of measurement - that is, helping social work researchers think about the wide variety of ways that social work concepts can be measured. Reflecting on these differences raises questions about underlying assumptions that in turn inspires creative theoretical insights. Rather than seeing measurement as simply a task to be completed in the research process, we will encourage the reader to think creatively about measurement and theory.
This book also addresses the interdependency of measurement and theory construction. In other words, this book covers how measurement and theory are connected in two different ways. First, every measure has its own working theory that relates the measure to the concept being measured. Second, theory construction is dependent on measurement. What we learn using a given measure could be different if a concept was measured in a different way.
measurement, validity, reliability, classical test theory, latent variables, scale construction, psychometrics