ABSTRACT
Item response theory (IRT) has gained considerable acceptance as a basis for item analysis and test scoring in the field of educational measurement, and IRT is gaining a foothold in social measurement as a structural model for the responses to survey questions. However, with the notable exception of the development of computerized adaptive testing, the advent of item response theory has not changed the appearance of tests—the theory has been applied almost exclusively to existing material. Even within the new context of computerized adaptive testing, IRT has been used to provide a scoring mechanism for old tests presented on a computer screen, one item at a time. It does not have to be this way.
