摘要
指标权重确定方法在教育测评中起到关键作用,将直接影响测评结果。层次分析法(Analytic Hierarchy Process,AHP)和熵值法是教育测评中最为常用的权重确定方法,但分别存在主观性较强和样本依赖性强的问题。此外,常用的主客观综合赋权法忽视了专家水平对指标权重的影响。而贝叶斯修正的AHP-熵值法的权重计算方法则通过利用专家自身权重对AHP法所得权重进行贝叶斯修正,然后与熵值法结合以得到组合权重。研究在综合分析贝叶斯修正的AHP-熵值法的优势及计算方法基础上,以内地西藏班教育质量测评为例,通过对多种方法进行比较,发现该方法计算出的结果可行、可操作且有效。
Index weight determination method plays a key role in education evaluation,which will directly affect the evaluation results.Analytic hierarchy process(AHP)and entropy method are the most commonly used methods,but they have the prominent problems of strong subjectivity and high sample dependence.In addition,the commonly used subjective and objective comprehensive weighting method ignores the influence of expert level on index weight.The weight calculation method of Bayesian modified AHP-entropy firstly modifies the weight obtained by AHP method by using the expert’s own weight,and then calculates the combined weight with entropy method.Based on the comprehensive analysis of the advantages and calculation methods of Bayesian modified AHP-entropy,taking the education quality evaluation of Tibetan classes in Chinese hinterland as an example,through the comparison of various methods,it is found that the results calculated by this method are feasible,operable and effective.
作者
沈光辉
叶彤
孙小坚
范涌峰
Shen Guanghui;Ye Tong;Sun Xiaojian;Fan Yongfeng(School of Teacher Education,Jiangsu University,Zhenjiang 212013;The Evidence-based Research Center for Educational Assessment,Jiangsu University,Zhenjiang 212013;School of Mathematics and Statistics,Southwest University,Chongqing 400715;Faculty of Education,Southwest University,Chongqing 400715)
出处
《心理学探新》
CSSCI
北大核心
2023年第3期269-277,共9页
Psychological Exploration
基金
国家社会科学基金重大项目“国家基础教育质量监测与评价体系研究”(19ZDA359)
2022年度教育部人文社会科学研究青年项目“中小学复杂问题解决能力及其影响机制的追踪研究”(22YJC880065)
2023年度教育部人文社会科学研究西部和边疆地区项目“中小学教师评价素养发展指数构建及应用研究”(23XJC880004)
2020年度重庆市博士后特别资助项目“STEAM教育的学习动机测评和CAT系统的实现”。
关键词
贝叶斯
层次分析法
熵值法
教育测评
组合赋权
Bayes
analytic hierarchy process
entropy method
education evaluation
combination weights
