ABSTRACT

In the present article, we present a systematical review of previous empirical studies that conducted formative assessment interventions to improve student learning. Previous meta-analysis research on the overall effects of formative assessment on student learning has been conclusive, but little has been studied on important features of formative assessment interventions and their differential impacts on student learning in the United States’ K-12 education system. Analysis of the identified 126 effect sizes from the selected 33 studies representing 25 research projects that met the inclusion criteria (e.g., included a control condition) revealed an overall small-sized positive effect of formative assessment on student learning (d = .29) with benefits for mathematics (d = .34), literacy (d = .33), and arts (d = .29). Further investigation with meta-regression analyses indicated that supporting student-initiated self-assessment (d = .61) and providing formal formative assessment evidence (e.g., written feedback on quizzes; d = .40) via a medium-cycle length (within or between instructional units; d = .52) were found to enhance the effectiveness of formative assessments.

摘要

在本文中，我们对以前的经验研究进行了系统的综述，这些经验研究进行了形成性评估干预措施以改善学生的学习。以前关于形成性评估对学生学习的整体影响的荟萃分析研究是结论性的，但是在美国的K-12教育体系中，关于形成性评估干预措施的重要特征及其对学生学习的不同影响的研究很少。对从代表25个满足纳入标准（例如，包括控制条件）的研究项目的33项研究中确定的126种效应大小进行的分析显示，形成性评估总体上对学生学习的小规模积极影响（d = 0.29）数学利益（d= 0.34），识字率（d = 0.33）和艺术（d = 0.29）。对荟萃回归分析的进一步调查表明，通过中等周期的长度（在或以内），支持学生发起的自我评估（d = 0.61）并提供正式的形成性评估证据（例如，对测验的书面反馈；d = .40）。在教学单位之间；d = .52）被发现可以增强形成性评估的有效性。