NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED458812
Record Type: Non-Journal
Publication Date: 2001-Oct
Pages: 15
Abstractor: N/A
Reference Count: N/A
ISBN: N/A
ISSN: N/A
Rasch Measurement and Item Banking: Theory and Practice.
Nakamura, Yuji
The Rasch Model is an item response theory, one parameter model developed that states that the probability of a correct response on a test is a function of the difficulty of the item and the ability of the candidate. Item banking is useful for language testing. The Rasch Model provides estimates of item difficulties that are meaningful, irrespective of the ability level tested. This paper focuses mainly on how the model can contribute to the feasibility of item banking in terms of language testing. The present research deals mainly with the following basic aspects of item banking: calibration of items for storage; the measurement of student abilities; and the advantages and limitations of item banking. Item characteristics can be determined either by traditional item statistics (Classical Test Theory) or a newer method of estimating item statistics called Item Response Theory. In this paper, Item Response Theory is used for dealing with item characteristics. The sample data for the research in this paper were taken from a multiple choice test that consisted of 10 items and was given to 105 students. It is concluded that Item Response Theory facilitates item banking by allowing all of the items to be calibrated and positioned on the same latent continuum by means of a common metric, and allows additional items to be added subsequently without the need to locate and retest the original sample of examinees. Furthermore, an item bank permits the construction of tests of known reliability and validity based on appropriate selection of item subsets from the bank without further need for trial in the field. (Contains 12 references.) (KFT)
Publication Type: Reports - Research
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A