STABLE ITEM/TEST RESPONSE FUNCTIONS WITH UNSTABLE ITEM PARAMETER ESTIMATES
Ogasawara, H., Otaru University of Commerce, Japan
The asymptotic standard errors of item/test response function estimates are derived by the delta-method with the assumption of the three-parameter logistic model. Using a similar method, the asymptotic standard errors of item, test and score information function estimates are obtained. Numerical illustrations show that the response function estimates are rather stable in spite of the unstable item parameter estimates. However, the information function estimates are shown to be relatively unstable. From these results, it is recommended to use the item/test response function estimates in place of the separate use of the item parameter estimates when such method is available as in item response theory linking.