REU Site: Leveraging The Learning Sciences & Technologies to Enhance Education and Learning in Secondary Schools

REU 网站：利用学习科学

• 批准号: 1950683
• 负责人: Neil Heffernan
• 金额: $ 32.06万
• 依托单位: Worcester Polytechnic Institute
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-06-01 至 2025-05-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1950683&HistoricalAwards=false
• 关键词: REU; Site; Leveraging; Learning; Sciences

This Research Experiences for Undergraduates (REU) Site at the Worcester Polytechnic Institute will provide a ten-week summer undergraduate research experience in learning sciences and technologies, with a focus on understanding achievement gaps among secondary school students. It will host three cohorts of eight students each summer, engaging 24 unique students. The increased use of educational technology has increased the amount of educational data available to researchers. These data have higher external validity than short-term laboratory studies, supporting the potential for making stronger conclusions about effective instructional practices. These data contain knowledge that could be analyzed to enhance understanding of how students best learn and how best to teach them. This REU will enable young researchers to analyze this data and contribute to understanding high-impact teaching practices, how to group learners for instruction, and what type of feedback or remediation is most applicable to a learner. These research activities will provide new information about the effectiveness of both technological and traditional educational interventions for a variety of learners in secondary school classrooms. REU undergraduates will experiment with deep learning approaches to analyze data from secondary school classrooms that use learning technologies. The goal will be to interpret and understand classroom activities that lead to meaningful student learning. Undergraduate researchers will use data collected by ASSISTments and Graspable Math. These are learning environments for mathematics developed by researchers at Worcester Polytechnic Institute and deployed in several schools throughout the United States. Undergraduates will analyze data using educational data mining techniques, such as student modeling to better estimate intervention impacts and clustering to determine which students are most similar. In addition, they will use E-TRIALS, a platform that simplifies running A/B experiments, to test educational interventions such as new hints or video instruction developed by teachers. To analyze classroom engagement, they will use ACORN, a deep-learning technology that assesses classroom dynamics by automatically monitoring student gaze and learner engagement. This project is supported by the NSF Improving Undergraduate STEM Education Program: Education and Human Resources, which supports research and development projects to improve the effectiveness of STEM education for all students. By funding REU sites, the program supports development of the next generation of STEM professionals.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

伍斯特理工学院的本科生研究经验（REU）网站将提供为期十周的夏季本科生研究经验，学习科学和技术，重点是了解中学生之间的成就差距。每年夏天，它将接待三批8名学生，吸引24名独特的学生。教育技术的使用增加了研究人员可用的教育数据量。这些数据比短期实验室研究具有更高的外部有效性，支持对有效教学实践做出更有力结论的可能性。这些数据包含可以分析的知识，以加强对学生如何最好地学习和如何最好地教他们的理解。该REU将使年轻的研究人员能够分析这些数据，并有助于理解高影响力的教学实践，如何将学习者分组进行教学，以及什么类型的反馈或补救最适用于学习者。这些研究活动将提供新的信息，说明技术和传统教育干预措施对中学课堂上各种学习者的有效性。 REU本科生将尝试使用深度学习方法来分析使用学习技术的中学教室的数据。我们的目标将是解释和理解课堂活动，导致有意义的学生学习。本科生研究人员将使用ASSISTments和Graspable Math. These收集的数据。这是由伍斯特理工学院的研究人员开发的数学学习环境，并部署在美国各地的几所学校。 本科生将使用教育数据挖掘技术分析数据，例如学生建模以更好地估计干预影响和聚类以确定哪些学生最相似。此外，他们将使用E-TRIALS（一个简化A/B实验运行的平台）来测试教育干预措施，如教师开发的新提示或视频教学。 为了分析课堂参与度，他们将使用ACORN，这是一种深度学习技术，通过自动监控学生的目光和学习者的参与度来评估课堂动态。该项目得到了NSF改善本科STEM教育计划的支持：教育和人力资源，该计划支持研究和开发项目，以提高所有学生STEM教育的有效性。通过资助REU网站，该计划支持下一代STEM专业人员的发展。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

How Common are Common Wrong Answers? Crowdsourcing Remediation at Scale

常见错误答案有多常见？

• DOI: 10.1145/3573051.3593390
• 发表时间: 2023
• 期刊: L@S '23: Proceedings of the Tenth ACM Conference on Learning @ Scale
• 作者: Gurung, Ashish;Baral, Sami;Lee, Morgan P.;Sales, Adam C.;Haim, Aaron;Vanacore, Kirk P.;McReynolds, Andrew A.;Kreisberg, Hilary;Heffernan, Cristina;Heffernan, Neil T.
• 通讯作者: Heffernan, Neil T.

Comparing Different Approaches to Generating Mathematics Explanations Using Large Language Models

比较使用大型语言模型生成数学解释的不同方法

• 发表时间: 2023
• 期刊: Doctoral Consortium and Blue Sky. AIED 2023
• 作者: Prihar, Ethan;Lee, Morgan;Hopman, Mia;Kalai, Adam Tauman;Vempala, Sofia;Wang, Allison;Wickline, Gabriel;Murray, Aly;Heffernan, Neil
• 通讯作者: Heffernan, Neil