调查问卷的变量合成及切分点计算方法研究——来自国际大规模测评项目的启示  

An Introduction to the Methods of Variable Synthesis and Cut-off Score Setting for Questionnaires:Experience from ILSA Program

作　　者：李潇 陈平[2] LI Xiao;CHEN Ping(Beijing Institute of Educational Supervision and Evaluation,Beijing 100036,China;Beijing Normal University,Beijing 100875,China)

机构地区：[1]北京教育督导评估院,北京100036 [2]北京师范大学,北京100875

出　　处：《中国考试》2023年第9期52-59,共8页journal of China Examinations

基　　金：2022年度中国基础教育质量监测协同创新中心基础教育质量监测科研基金项目“教育质量监测过程中的异常作答检测方法研究”(2022-01-082-BZK01)。

摘　　要：问卷法作为主流的实证研究方法被广泛应用于教育评价领域,其使用的变量合成与分数切分点计算方法直接关系到测评结果的科学性和解释力。目前,PISA、TIMSS和PIRLS等国际大规模测评项目主要采用基于项目反应理论的量尺化方法合成变量,应用分数链接技术将趋势量表分数转换到基准量尺,实现不同测验周期的测评结果可比。另外,TIMSS 2019在充分运用Rasch簇模型性质的基础上,根据专家评定的反应类别组合来确定分数切分点,而PISA 2018则基于被试在群体中的相对位置确定切分点。Questionnaire data has been widely used in the field of educational evaluation.This paper introduces the way of variable synthesis and cut-off score setting in PISA 2018 and TIMSS 2019.Both ILSA programs adopt the IRT scaling method to composite scores and apply linking technology to convert the scores to the benchmark scale,which can achieve the longitudinal comparability of the questionnaire scores in each test cycle and then characterize the development trend of the examinee's latent trait.In addition,the cut-off scores are calculated in terms of combinations of response categories employing the property that"the raw score and ability value have a conversion relationship"in the Rasch-family model in TIMSS 2019,while the cut-off scores in PISA 2018 are determined based on the quantile of subjects'latent trait in the population.

关 键 词：问卷法 变量合成 项目反应理论 国际大规模测评项目 

分 类 号：G405[文化科学—教育学原理]