A Use of the Information Function in Tailored Testing

Several important and useful implications in trait theory, with direct implications for individualized or tailored testing, are pointed . A way of using the information function in testing in connection with the standard error estimation of the ability level using maximum estimation is suggested. It is emphasized the standard error of estimation be considered as the major index of dependability, opposed to the reliability of a test. The of weak parallel forms is expanded to testing in which different sets of items are to different examinees. Examples are . have tended to use latent trait theory than classical test theory in research on individualized or tailored testing. This is natural, since latent trait theory has definite over classical test theory in many crucial matters. of the lack of opportunities to really the theory, however, these researchers tend to some important implications in latent trait . As a result, its full use has not yet . Not only are information functions used to maximum advantage, but also those have tried to use latent trait theory still use popular concepts in classical test theory, such reliability. expanded to testing in which different sets of items are to different examinees. Examples are . In paper, the author points out some important in latent trait theory which are fully understood and appreciated among researchers, gives some practical suggestions for use.