华东师范大学学报(教育科学版) ›› 2020, Vol. 38 ›› Issue (6): 70-85.doi: 10.16382/j.cnki.1000-5560.2020.06.005

• STEM教育 • 上一篇    下一篇

寻找“最佳证据”:如何运用元分析进行文献综述——以STEM教育对学生成绩的影响研究为例

曾昭炳, 姚继军   

  1. 南京师范大学教育科学学院,南京 210097
  • 发布日期:2020-06-23
  • 基金资助:
    2018年度教育部人文社会科学研究规划基金项目“中小学生STEM学习效果的影响因素及其作用机制研究”(18YJA880108);“江苏高校优势学科建设工程资助项目”(PAPD)

Identifying the “Best Evidence”: How to Use Meta-analysis to Conduct a Literature Review—A Case of STEM Education’s Effect on Students’ Academic Achievement

Zeng Zhaobing, Yao Jijun   

  1. School of Education Science, Nanjing Normal University, Nanjing 210097, China
  • Published:2020-06-23

摘要: 元分析相对于传统的主观性文献综述方法而言,能够更加有效、客观和规范地从既有实证研究文献中梳理出一般性、规律性的结论,因此成为教育循证改革过程中寻找"最佳证据"的重要方法。与国外相比,我国运用元分析方法进行实证文献综述的研究还比较少。针对于此,本研究以STEM教育对学生学业成绩的影响为例,展示了运用元分析方法进行实证研究综述的过程。研究发现STEM教育有利于提高学生的学业成绩(d=0.410),STEM教育方法、受教育阶段、地区和样本量等因素均会显著影响到STEM教育的效果。这样的研究,不仅提供了关于STEM教育效果的一般性证据,而且在方法层面探讨了如何通过文献综述获得可靠证据以支持教育改革。

关键词: 教育循证改革, STEM教育, 学业成绩, 元分析

Abstract: Compared with the method of traditional subjective literature review, meta-analysis is more objective, normative, and can effectively sort out general conclusion from existing empirical researches. Therefore, it has become an important method to identify the “best evidence” in the process of evidence-based educational reform. So far there are few educational studies in China using the meta-analysis method to review the empirical literatures. In view of this, this research analyzes the impact of STEM education on students’ academic achievement as an example of how to use meta-analysis to synthesise existing empirical research. We found: 1) STEM education is conductive to students academic achievements (d=0.410); 2) factors like educational methods, students’ educational stage, location and sample size can affect the effects. The study provides the general evidence of effects of STEM education and explores how to obtain reliable evidence through a literature review to support educational reform.

Key words: evidence-based educational reform, STEM education, academic achievement, meta-analysis