摘要
近年来我国中学阶段的班级规模整体缩小,部分省市还开展了"小班化教育"试点,但班级规模对学生成绩的影响尚不明确。本文利用2015年"国际学生评估项目(PISA)"大陆地区科学测评数据,通过泛精确匹配基础上的多层线性回归模型,估计了班级规模与学生成绩之间的关系。研究发现,个人和家庭特征与学生的班级选择显著相关,优势家庭更排斥"大班",但是"小班"的成绩优势并不明显。初中阶段的班级规模与学生成绩呈现非线性关系,"大班"的学生显著更低。高中阶段的班级规模与学生成绩显著正相关,"大班"成绩更好。分样本回归发现,上述关系在城市和乡镇中具有异质性,乡镇的"小班"成绩更差。"小班"优势不明显可能与教师教学行为有关,初步分析表明只有城市中学中的"小班"教师采用了与班级规模相应的教学方式。
To identify the effect of attending smaller class on student s achievement,we use the PISA 2015 data of China's Mainland area(B-S-J-G),to analyze whether a small-size class can improve a students academic performance.We categorize classes into three groups using common Chinese practices.The HLM model is applied to our baseline analysis after constituting a quasi-experimental design by CEM.The model shows a heterogeneous impact of smaller class size,condition on personal,peer,and school characteristics.Among junior high school participants,the class size effects are non-linear.At the senior high school level,students performance is positively correlated with class size.The associations between class size and test score are more complicated when we divide the sample into rural,township,and urban groups.The poorer achievement in rural and towns confirm that these smaller classes illustrate are a negative reality.From our further investigation,we find that only teachers in city high schools take advantage of smaller class size by changing their pedagogical practice accordingly.
作者
郑琦
杨钋
ZHENG QI;YANG Po
出处
《北京大学教育评论》
CSSCI
北大核心
2018年第4期105-127,186,187,共25页
Peking University Education Review
基金
北京大学教育财政科学研究所"数字化时代的教育财政策略"课题(CZS201903)
