MetaDash: A Teacher Dashboard Informed by Real-Time Multichannel Self-Regulated Learning Data

该项目由教育和人力资源核心研究计划支持，该计划支持STEM学习和学习环境中的基础研究。重要的是要找到新的方法来支持学生构建、分析、批判和使用STEM现象的模型的能力。鉴于教师是任何教育创新的主要调解人，支持STEM教师有效地让学生学习批判性思维技能是当务之急。这个项目涉及MetaDash的研究和开发，这是一个教师仪表板，提供关于学生在STEM教学过程中的认知、情感、元认知和动机自我调节学习过程的信息。仪表板由多个通道提供信息，这些通道综合了学生的面部表情、眼睛凝视行为、皮肤电活动和言语等信息。MetaDash将通过提供实时学生数据(包括学生个人和聚合数据)来改善教学决策，从而影响当前的教师培训。研究方法论围绕Metadash的设计和测试展开，Metadash是一种智能、多通道的数据可视化工具，可实时显示学生学习过程和知识构建的关键方面。该研究方法研究了(1)如何以及何时基于人机交互设计原则和教师可用性研究来呈现多通道输入；(2)如何优化统计方法来处理来自多个来源的非结构化数据；(3)如何使用多通道测量(如眼球跟踪和面部表情)为自我调节、动机和挫折感等概念创建行为特征。MetaDash的最终目标是通过支持教师和学生对CAMM SRL过程的监控来促进STEM学习。因此，MetaDash将通过为教师提供：(1)多渠道STEM学习和从学生那里收集的认知、情感、元认知和动机(CAMM)自我调节学习(SRL)数据；以及(2)学生个人和聚合数据，以加速教师决策，从而推动当前的仪表板。

The project is supported by the Education and Human Resource Core Research program, which supports fundamental research in STEM learning and learning environments. It is important to find new ways to support students' ability to construct, analyze, critique, and use models of STEM phenomena. Given that teachers are the main mediator of any educational innovation, it is imperative to support STEM teachers to effectively engage students in critical thinking skills. This project involves the research and development of MetaDash, a teacher dashboard that provides information regarding students' cognitive, affective, metacognitive, and motivational self-regulatory learning processes during STEM instruction. The dashboard is informed by multi-modal channels that synthesize information such as student facial expressions, eye gaze behavior, electrodermal activity, and verbalizations. MetaDash will impact current teacher training by providing real-time student data (both individual student and aggregated) to enhance instructional decision-making.Research methodology centers around the design and testing of Metadash as an intelligent, multichannel data visualization tool that displays key aspects of students' learning processes and knowledge construction in real time. The research approach investigates (1) how and when to present the multi-channel input based on human-computer-interaction design principles and informed by teacher usability studies; (2) how to optimize statistical approaches to handle unstructured data from multiple sources; and (3) how to create behavioral signatures for constructs such as self-regulation, motivation and frustration using multi-modal measures such as eye-tracking and facial expression. The ultimate goal of MetaDash is to foster STEM learning by supporting teachers' and students' monitoring and control of CAMM SRL processes. As such, MetaDash will advance current dashboards by providing teachers with: (1) multichannel STEM learning and cognitive, affective, metacognitive, and motivational (CAMM) self-regulated learning (SRL) data collected from students; and (2) individual student and aggregate data to accelerate teachers' decision-making.