摘要
学生的数学素养具有多维结构,素养导向的数学学业成就测评需要提供被试在各维度上的表现信息,而不仅是一个单一的总分。以PISA数学素养结构为理论模型,以多维项目反应理论(MIRT)为测量模型,利用R语言的MIRT程序包处理和分析某地区8年级数学素养测评题目数据,研究数学素养的多维测量方法。结果表明:MIRT兼具单维项目反应理论和因子分析的优点,利用其可对测试的结构效度和测试题目质量进行分析,以及对被试进行多维能力认知诊断。
The student’s mathematical literacy is inherently multi-dimensional,and the literacy-oriented mathematics assessment needs to provide the student’s performance information on each dimension,not just a single total score.Taking the mathematical literacy structure of PISA as the theoretical model and the multidimensional item response theory(MIRT)as the measurement model,using the MIRT package for the R environment to analyze the data,we study the multidimensional measurement method of students’mathematical literacy.The data is from the program for eighth-grader mathematics literacy assessment project in a district.This research shows that MIRT has the advantages of uni-dimensional item response theory and factor analysis.MIRT can be used to analyze the dimensional structure of the test and the quality of the item.MIRT can be also used to diagnose the multi-dimensional cognitive ability.Finally,the procedure on MIRT mathematical literacy measurement and several problems about the use are reviewed,and some prospects are put forward.
作者
林子植
胡典顺
LIN Zizhi;HU Dianshun(Central China Normal University,Wuhan 430079,China;Jiangxi Science and Technology Normal University,Nanchang 330013,China)
出处
《中国考试》
CSSCI
北大核心
2021年第5期72-80,共9页
journal of China Examinations
基金
教育部人文社会科学研究规划2019年度基金项目“中小学核心素养测评的模型建构与实证研究”(19YJA880012)。
关键词
多维项目反应理论
数学素养测量
题目质量分析
多维认知诊断
multidimensional item response theory
mathematical literacy measurement
item quality analysis
multidimensional cognitive diagnosis
