Transparent Assessment in Online Learning: Making the User a Partner in Complex Inference Processes

在线学习中的透明评估：让用户成为复杂推理过程的合作伙伴

• 批准号: 0537322
• 负责人: Steven Tanimoto
• 金额: $ 36万
• 依托单位: University of Washington
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2006
• 资助国家: 美国
• 起止时间: 2006-02-01 至 2010-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0537322&HistoricalAwards=false
• 关键词: Transparent; Assessment; Online; Learning; Making

Computer-based learning environments continue to grow in complexity as technology for education keeps improving. One of their functions is behind-the-scenes assessment -- determining how well the student is doing and identifying any learning difficulties the student is facing. A key issue is how to make parts of that assessment activity visible to the student. Good feedback can enhance the efficiency of learning. However, full and immediate disclosure of assessment activity is not always consistent with pedagogy. This project is developing and studying a variety of policies and methods to make learning environment more transparent. Full transparency is not achieved simply by revealing all the system's assessment data; the data and processes must be presented in understandable forms. The primary target student population for the project's experiments consists of University of Washington freshmen and sophomores who enroll in a course entitled "Pixels, Numbers and Programs." The research seeks a clearer understanding of the potential costs and benefits of revealing different kinds and amounts of assessment information to students as they work in computer-based learning environments. Most online learning environments that perform educational assessment have been rigid, not only in enforcing a linear interaction style (e.g., sequence of questions posed by the system, with little opportunity for student initiative), but also in terms of assumptions about student disposition to trust the software and engage willingly in the session. Transparency has the potential to engender trust by a wider set of students and thus broaden the range of students who can benefit from online learning environments. The project works toward an improved understanding of openness of assessment in computerbased learning environments. This, in turn, can assist designers of future systems in dealing with this increasingly important issue.

随着教育技术的不断改进，基于计算机的学习环境的复杂性不断增加。他们的功能之一是幕后评估-确定学生做得如何，并确定学生面临的任何学习困难。一个关键问题是如何让学生看到评估活动的一部分。良好的反馈可以提高学习效率。然而，充分和立即披露评估活动并不总是符合教学法。该项目正在制定和研究各种政策和方法，使学习环境更加透明。完全透明不能仅仅通过披露系统的所有评估数据来实现;数据和过程必须以可理解的形式呈现。该项目实验的主要目标学生群体包括华盛顿大学的大一新生和大二学生，他们注册了一门名为“像素、数字和程序”的课程。“这项研究旨在更清楚地了解在计算机学习环境中向学生提供不同种类和数量的评估信息的潜在成本和效益。大多数执行教育评估的在线学习环境都是严格的，不仅在执行线性交互风格方面（例如，系统提出的一系列问题，学生主动性的机会很少），而且还取决于对学生信任软件并愿意参与课程的倾向的假设。透明度有可能使更多的学生产生信任，从而扩大可以从在线学习环境中受益的学生的范围。该项目致力于提高对基于计算机的学习环境中评估的开放性的理解。这反过来又可以帮助未来系统的设计者处理这个日益重要的问题。