Abstract
Item response theory (IRT) is a modeling approach that links responses to test items with underlying latent constructs through formalized statistical models. This article focuses on how IRT can be used to advance science and practice in organizations. We describe established applications of IRT as a scale development tool and new applications of IRT as a research and theory testing tool that enables organizational researchers to improve their understanding of workers and organizations. We focus on IRT models and their application in four key research and practice areas: testing, questionnaire responding, construct validation, and measurement equivalence of scores. In so doing, we highlight how novel developments in IRT such as explanatory IRT, multidimensional IRT, random item models, and more complex models of response processes such as ideal point models and tree models can potentially advance existing science and practice in these areas. As a starting point for readers interested in learning IRT and applying recent developments in IRT in their research, we provide concrete examples with data and R code.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call