How Methodological Features Affect Effect Sizes in Education
How Methodological Features Affect Effect Sizes in Education
复制标题
DOI：
10.3102/0013189x16656615
复制
发表时间：
2016-06-01
影响因子：
8.2
通讯作者：
Slavin, Robert E.
中科院分区：
文献类型：
--
作者：
Cheung, Alan C. K.;Slavin, Robert E.
随着证据在教育政策中变得越来越重要，理解研究设计如何在评估教育计划的实验中有助于报告的效果大小是至关重要的。共有645项研究来自最近对学前教育、阅读、数学和科学项目评估的12篇综述。与未发表的文献、大规模研究和独立措施相比，已发表的文章、小规模试验和实验者制定的措施的效果大小分别是未发表的文献、大规模研究和独立措施的两倍。准实验的效应大小显著高于随机实验。不包括辅导研究，初等和中等/高等研究之间的效果大小没有显著差异。回归分析发现，在控制了所有其他因素后，所有因素的影响保持不变。讨论了方法学特征对效果大小的影响的解释，以及对循证政策的影响。
As evidence becomes increasingly important in educational policy, it is essential to understand how research design might contribute to reported effect sizes in experiments evaluating educational programs. A total of 645 studies from 12 recent reviews of evaluations of preschool, reading, mathematics, and science programs were studied. Effect sizes were roughly twice as large for published articles, small-scale trials, and experimenter-made measures, compared to unpublished documents, large-scale studies, and independent measures, respectively. Effect sizes were significantly higher in quasi-experiments than in randomized experiments. Excluding tutoring studies, there were no significant differences in effect sizes between elementary and middle/high studies. Regression analyses found that effects of all factors maintained after controlling for all other factors. Explanations for the effects of methodological features on effect sizes are discussed, as are implications for evidence-based policy.