Collaborative Research: Learning Linkages: Integrating Data Streams of Multiple Modalities and Timescales

这项关于教育和学习的研究(REAL)项目源于2014年10月关于改善教与学的数据密集型研究的IDEAS实验室。这项工作的目的是：(1)将各学科的研究人员聚集在一起，促进新的、变革性的、多学科的方法，利用与教育相关的大型数据集中的数据，创造可操作的知识，以便在中期改善STEM的教学和学习环境；(2)从长远来看，使学习发生革命性变化。在这个项目中，来自卡内基-梅隆大学、韦斯特大学、亚利桑那州立大学和西北大学的研究人员将合作，加强对学习影响的理解，并改善高中和中学STEM课堂的教与学。为了实现这一目标，他们将利用最新的数据处理工具和许多不同的数据流，这些数据流可以在技术丰富的教室中收集，以(1)确定影响学习的课堂因素，(2)探索如何使用这些数据来自动跟踪学生随着时间的推移的理解和能力的发展。有两股力量正准备改变关于学习的研究。首先，越来越多的学生工作在计算机和在线上进行，产生了大量与学习有关的数据。与此同时，计算、数据挖掘和学习分析的进步正在为收集、分析和表示这些数据提供新的工具。同时，可用的数据和分析工具使智能和响应系统能够为个别学习者提供个性化的学习体验。PI的目标是收集高度丰富的数据，远远超出典型的计算机数据捕获，利用最新的数据处理工具来生成关于STEM教学和学习的新见解。为了最大限度地发挥潜力，同时降低自动化数据收集和分析的风险，他们将：(1)从两个不同的计算机支持的学习环境(一个用于中学数学，一个用于高中科学)的八个不同的两周设定中收集和整合不同的数据来源，包括日志文件、视频和书面人工制品；以及(2)将日志文件数据的分析与综合数据集的分析进行比较，以了解使用日志文件数据评估学生学习和熟练程度的可能性和局限性。合作者预计，他们的发现将为适用于广泛学科和环境的理论和实践建议提供参考。

This Research on Education and Learning (REAL) project arises from an October 2014 Ideas Lab on Data-intensive Research to Improve Teaching and Learning. The intentions of that effort were to (1) bring together researchers from across disciplines to foster novel, transformative, multidisciplinary approaches to using the data in large education-related data sets to create actionable knowledge for improving STEM teaching and learning environments in the medium term; and (2) revolutionize learning in the longer term. In this project, researchers from Carnegie-Mellon University, Wested, Arizona State University, and Northwestern University will collaborate to enhance understanding of influences on learning, and improve teaching and learning in high school and middle school STEM classes. To accomplish this, they will leverage the latest tools for data processing and many different streams of data that can be collected in technology-rich classrooms to (1) identify classroom factors that affect learning and (2) explore how to use that data to automatically track development of students' understanding and capabilities over time. Two forces are poised to transform research on learning. First, more and more student work is conducted on computers and online, producing vast amounts of learning-related data. At the same time, advances in computing, data mining, and learning analytics are providing new tools for the collection, analysis, and representation of these data. Together, the available data and analytical tools enable smart and responsive systems that personalize learning experiences for individual learners. The PIs aim to collect highly enriched data that go far beyond typical computer data capture, leveraging the latest tools for data processing to generate new insights about STEM teaching and learning. Working to maximize the potential while mitigating the risks of automated data collection and analysis, they will: (1) collect and integrate diverse sources of data including log files, videos, and written artifacts from across eight different two-week enactments of two different computer supported learning environments (one used in middle school math and one in high school science); and (2) compare analyses of log-file data with analyses of integrated datasets to understand the possibilities and limitations in using log-file data for assessment of student learning and proficiency. The collaborators expect their findings will inform both theories and practical recommendations applicable across a wide range of disciplines and settings.