The purpose of this chapter was to present an overview of the development of item response theory as an alternative to classical test theory. Item response theory and the models that are relevant to applications in behavioral medicine were presented. An application to a self-reported patient outcome, depression, was also presented to illustrate the results from IRT modeling and to show how item quality could be assessed. Once an adequate item bank is established and the item parameters are estimated, IRT can be used to estimate trait levels from item subsets or from adaptive testing. Unlike classical test theory, the examinee's true score is not test dependent, and the item parameters are not sample dependent, therefore the parallel test assumption is not necessary in IRT.

0 0

Post a comment