The server is under maintenance between 08:00 to 12:00 (GMT+08:00), and please visit
later.
We apologize for any inconvenience caused
Missing data modeling and Bayes estimation in IRT
Author(s): LI Bin, LI Xiaoyi, FU Zhihui, School of Mathematics and Systems Science, Shenyang Normal University
Pages: 216-
220
Year: 2015
Issue:
2
Journal: Journal of Shenyang Normal University, Natural Science Edition
Keyword: IRT; missing data; Gibbs sampling;
Abstract: Item response theory mainly studys the reaction of the person in the test project and the relationship between the achievements and latent traits.The application of item response theory depends on whether the parameters of the model can be effectively estimated.The integrity of the data has a certain effect on the parameter estimation,while the missing of data in the item response process is very common.The mechanism of missing data affects the processing method.As a consequence,using the graded item response model to fit the observed data and Rasch model to the missing data and according to latent variable modeling method,is a solution to the non-ignorable missing data.Finally,the Gibbs sampling method is applied to extract the parameter and provide the estimation.By simulation study,it verifies the conclusion that the method effectively reduces the deviation owing to ignoring the missing data during parameter estimation.
Citations
No citation found