Abstract
Computerized adaptive testing for cognitive diagnosis (CD-CAT) is a method that combines cognitive diagnosis testing and computerized adaptive testing. Compared with traditional cognitive diagnostic tests, it is more efficient and offers more tailored insights into examinees’ knowledge state. As a core part of CD-CAT, the item selection method directly affects the accuracy and efficiency of the overall assessment. This study aims to construct two nonparametric item selection methods based on the Hamming distance nonparametric difference index (HD-NDI) and the general Hamming distance nonparametric difference index (GHD-NDI). Subsequently, the performance of the two proposed item-selection methods in CD-CAT is examined through a simulation study. The key findings are summarized as follows: (1) Under most conditions, GHD-NDI performs better than other item selection methods, especially when the calibration sample size is small.(2) When w_ij is unavailable for each item, CD-CAT should use the HD-NDI method; when w_ij is available, CD-CAT should use GHD-NDI method.
Recommended Citation
junjie, Li; Huijing, Zheng; Chunhua, Kang; and Zhang, Yimu
(2026)
"Two New Non-parametric Item Selection Methods for CD-CAT,"
Chinese/English Journal of Educational Measurement and Evaluation | 教育测量与评估双语期刊: Vol. 7:
Iss.
2, Article 1.
DOI: https://doi.org/10.59863/XXJH1777
Available at:
https://www.ce-jeme.org/journal/vol7/iss2/1
DOI
https://doi.org/10.59863/XXJH1777
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
