慕课同伴互评模型设计研究

Design of Peer Assessment in xMOOCs

2011 年以来，慕课热潮席卷全球，成为高等教育领域的变革力量。随着慕课教学理念和实践的流 行和深入发展，如何对慕课学习者进行学习测评成为大规模、开放式、在线教学活动环境下教师、研究人员和慕 课学习者关注重点之一。同伴互评作为学习测评的一种方式被认为能够有效应对慕课环境下大规模学习者学 习测评需求。基于此，本文在文献研究的基础上详述了同伴互评的研究发展，并结合本人在美国访学期间参与 佛罗里达州立大学慕课教学实践分析了其优点和不足，从教学法和教育技术应用的角度提出慕课学习环境下五 种不同的同伴互评模型，即同伴评分模型、专家扮演同伴互评模型、社交网络同伴互评模型、跨文化同伴互评模 型和批判性同伴互评模型，指出慕课环境下进行同伴互评模型设计应关注的设计原则、标准和未来研究方向。

Because of rapid development and adoption of ICT in global education， MOOCs have been applauded as the disruptive technology and practice in higher education. Since 2011， xMOOCs has become increasingly accepted and practiced in higher education， and 2012 is called “the Year of MOOCs. 冶However， the learning outcomes evalu ation in xMOOCs remains as one of the biggest challenges to teachers， researchers and learners. In order to overcome such a challenge， peer assessment has been viewed as one of such potential solutions. Aligned with this assumption， the author conducted a literature review focusing on peer assessment. The purpose of the literature review is to track peer assessment research and development， and to identify different learning measure ment strategies. The literature were mainly collected from ERIC， a database on educational reports， evaluations， and research from the Educational Resources Information Center， consisting of Resources in Education Index， and Current Index to Journals in Education. Additionally， some relevant articles from Google Scholar were also collected. In the literature review， the author also employed his personal experience and observation when participating in the Small On line Open Course offered by the College of Educationat Florida State University as a visiting scholar. The author proposed five design models of peer assessment to measure xMOOC learners on a large scale. These models include: peer grading model， expert -playing model， social network model of peer evaluation， intercultural model of peer assessment， and critical peer assessment model， respectively. Each model was analyzed with their imple mentation， merits and demerits in actual xMOOC learning environments. Additionally the authors discussed design is sues and criteria for peer assessment and future research directions. At the end， the author recommended to use five de sign models as assessment tools in xMOOCs and as exemplary models to guide future design in assessing massive online learners.