Item response theory

Abstract

Abstract Since the 1960s, there has been a revolution in the approach to scale development. Called item response theory (IRT), this approach challenges the notion that scales must be long in order to be reliable, and that psychometric properties of a scale derived from one group of people cannot be applied to different groups. This chapter provides an introduction to IRT, and discusses how it can be used to develop scales and to shorten existing scales that have been developed using the more traditional approach of classical test theory. IRT also can result in scales that have interval level properties, unlike those derived from classical test theory. Further, it allows people to be compared to one another, even though they may have completed different items, allowing for computerized adaptive testing. The chapter concludes by discussing the advantages and disadvantages of IRT.