Current multidimensional computerized adaptive testing (MCAT) is limited only to linked multiple abilities that can compensate for one another rather than non-compensatory ones. In recognition of the usefulness of MCAT and the complications associated with non-compensatory test structure, we propose extending this model to other types of abilities (non-compensatory ones) and evaluating their performance. Three popular item selection methods are used and compared, namely, the Fisher information method, the mutual information method, and the Kullback-Leibler information method. The simulation results showed that the Fisher information and mutual information methods performed similarly, and both outperformed the Kullback-Leibler information method. Furthermore, it was found that the more stringent the termination criterion and the higher the correlation between the latent traits, the higher the resulting measurement precision and test reliability. Test reliability was very similar across the dimensions, regardless of the correlation between the latent traits and termination criterion. Generally, the difficulties of the administered items were found to be at a lower level than the examinees abilities, which shed light on item bank construction for non-compensatory items. Copyright © EcoSta 2019.
|Publication status||Published - Jun 2019|
|Event||The 3rd International Conference on Econometrics and Statistics (EcoSta 2019) - National Chung Hsing University, Taichung, Taiwan, Province of China|
Duration: 25 Jun 2019 → 27 Jun 2019
|Conference||The 3rd International Conference on Econometrics and Statistics (EcoSta 2019)|
|Abbreviated title||EcoSta 2019|
|Country/Territory||Taiwan, Province of China|
|Period||25/06/19 → 27/06/19|