基于可达阵的补偿模型Q矩阵标定方法
The Method for Compensatory Model's Q-Matrix Specification Based on the Reachability Matrix

Q矩阵标定是认知诊断评估中研究的热点问题,Q矩阵的好坏决定了认知诊断评估的准确性.根据确定性输入噪声“与”门模型(DINA)中可达阵R与简化Q矩阵存在布尔“与”的关系,提出基于确定性输入噪声“或”门模型(DINO)的可达阵R与简化Q矩阵在列向量上存在布尔“或”的关系,并由此推导出基于可达阵的补偿模型Q矩阵标定方法.实验结果表明:当可达阵失误与猜测小于0.20且待标定项目参数小于0.25时,该方法所得Q矩阵元素返真率达到90%以上,且在可达阵失误与猜测参数均小于0.25时真实Q矩阵与估计Q矩阵之间的差异较小.
It is very important that the calibration method of Q-matrix in cognitive diagnosis,which directly determines the correctness of the classification of individual.The augment algorithm provides fact that any column of the reduced Q-matrix can be expressed by the columns of the reachability matrix under the logical OR operation.The purpose of this study is to propose a method for compensatory model's Q-matrix specification based on the reachability matrix.Simulation results show that the performance of the new method is promising in terms of correct classification rates of examinees' attributes