【作者】 陈琳； 张敏强；

【Author】 Chen Lin Zhang Minqiang Psychological Application Research Center,South China Normal University,South China Normal University,No.55, Zhongshan Road,Guangzhou,China,

【机构】 华南师范大学心理应用研究中心；

【摘要】 将题组GIRM(Testlet Response Thoery)方法和传统GT(Generalizability Theory)方法应用于实际研究,通过对实测数据进行分析,将题组GIRM方法与传统GT方法进行对比,来探讨题组GIRM方法的特点,以检验在GT和TRT模型的基础上估计方差分量的题组GIRM方法在多大程度上适用于实测数据。结果用WinBUGS软件进行处理,通过MCMC方法计算题组GIRM方法下的各方差分量,并计算其与传统GT方法下各方差分量的差异,比较两种方法下所得出的方差分量、相对误差、绝对误差、概化系数及可靠性系数,考察题组GIRM方法在应用领域的估计效果。更多还原

【Abstract】 This study applies both Testlet Response Thoery(TRT) approach and Generalizability Theory(GT) approach to practical application,exploring the feature of the TRT approach by analyzing real data,to examine the extent to which estimates of variance components based on GT and TRT models are the same using real data.Using the software WinBUGS with Markov Chain Monte Carlo(MCMC) method,comparisons are made between the two approaches on the estimates of variance components,relative error variances,absolute error variances,generalizability coefficients,and dependability coefficients,to inspect the performance of the TRT approach in application.更多还原