适用于二级与多级评分项目组成的混合测试的加权最大似然潜在特质估计

摘要

提出适用于二级与多级评分项目组成的混合测试的加权最大似然潜在特质估计。使用N-R算法获得WML估计，并给出相关方程的详细推导。为探讨WML的性能，进行了模拟研究，所得到的结果表明， WML的估计比最大似然估计（MLE）具有更好的性能。最后，基于一个实际例子对该方法进行了实证研究。%In this article, a weighted maximum likelihood (WML) latent trait es-timator is proposed for the mixed-type tests composed of both dichotomous and polytomous items. The N-Ralgorithm is used to obtain the WML estimator, and the relevant equations required are given in detail. To evaluate the performance of the WLE, a simulation study is conducted, and the obtained results show that the WML estimator has better properties than the maximum likelihood estimation (MLE). Finally, an empirical study is used to demonstrate the methodology.