汉语阅读理解测验的认知诊断研究--以中国政府奖学金本科来华留学生预科教育汉语综合统一考试为例  

A Study of Cognitive Diagnostic Assessment Based on Fusion Model For Reading Comprehensive Test:A Case Study of CPT(Science and Engineering)

作　　者：陈璐欣 王佶旻[1] 

机构地区：[1]北京语言大学汉语考试与教育测量研究所,北京100083

出　　处：《中国考试》2016年第2期23-34,共12页journal of China Examinations

基　　金：国家社科基金青年项目(12CYY028);教育部新世纪优秀人才支持计划(NCET-13-0690)的支持

摘　　要：中国政府奖学金本科来华留学生预科教育汉语综合统一考试是北京语言大学汉语考试与教育测量研究所研制的新型汉语考试。该考试用以衡量预科生结业的语言水平,它的特点是集基础汉语和专业汉语为一体,对预科的语言能力进行全面测评。为更好地了解预科生阅读能力的认知模式,为考生和学校提供诊断性评价,我们采用融合模型(Fusion Model,FM)对2014年5月参加该考试的532名理工类被试在阅读理解部分的作答情况进行诊断性研究。研究发现:1该批被试对信息查找能力掌握得较好,对信息转译能力掌握得不理想;2不同高校在汉语教学上存在一定偏向;3该部分试题中第93题对于被试来说较难,第89题、91题、93题和94题在属性完备性上不够理想,建议进行调整和修改。Chinese Proficiency Test （CPT） is the new exam which is researched and developed by Chinese Examination and Education Survey Institute in Beijing Language and Culture University, and is used to understand language skill situation of preparatory student. It is the basic Chinese and the specialized Chinese as one to assessment the language ability of student. So, we adopt Fusion Model （FM） to do cognitive diagnostic studies to 532 sciences ＆ engineering subjects＇ response on reading understand cognitive mode of preparatory student in comprehension section of CPT in May, 2014 in order to reading comprehension skill and provide diagnostic evaluation for student ＆ teacher. The study found that： Firstly, these subjects master ＂searching ability＂, but non-master ＂translation ability＂. Secondly, the subjects in different school have some different knowledge structure in reading comprehension. Thirdly, we suggest adjust and amend some items, for example, the item 93 is too difficult to the subjects, and the item 89, itemgl, item 94 are lacking some attributes in Q matrix.

关 键 词：融合模型 预科教育 阅读理解 认知诊断 

分 类 号：G405[文化科学—教育学原理]