EXP: Linguistic Analysis and a Hybrid Human-Automatic Coach for Improving Math Identity

网络学习和未来学习技术计划为支持展望学习技术的未来并推动我们了解人们如何在技术丰富的环境中学习的努力提供资金。网络学习探索(EXP)项目通过设计和构建新的学习技术，并研究它们促进学习的可能性和有效使用它们的挑战，来探索新类型的学习技术的可行性。这个项目探讨了学生的社会认同对学习的影响，这是数学和科学教育中的一个重要因素。具体地说，它将促进对数学恒等式的科学理解(即通过对100,000多名使用推理思维的不同学生的研究)，推理思维是一种针对K-8数学的混合学习系统，并展示了结果。推理思维通过一种创新的设计来支持数学身份，允许学生通过电子邮件发送一个动画角色(也就是精灵)，并收到人类精心设计的回复。本研究将展示数学同一性在学生使用推理思维的过程中如何表现和变化，以便让软件设计者和课堂教师了解鼓励数学同一性的最佳实践。这将对加强我国为科学技术领域提供训练有素的劳动力的能力产生更广泛的影响。这项研究考察了数学认同的两个组成部分：自我效能感和对数学的兴趣。自我效能感和兴趣都可以通过个性化的课程来增强，以适当地挑战每个学生(教育技术的优势)，但社会刻板印象(即，“女孩不擅长数学”)可能会降低数学认同感，或以其他方式干扰它的发展。这项研究调查了教育技术如何扭转这些趋势。在这项研究的第一阶段，将使用调查方法、自然语言处理(NLP)和教育数据挖掘(EDM)技术来确定在学生与GenieMail互动的语言模式(以及推理思维的其他部分)中，数学认同感是如何糟糕和/或发生变化的。然后，这些发现将被用来增强GenieMail和其他与推理思维系统的教学互动，方法是创建一个人类/人工智能混合系统，目标是提高不同学生群体的数学认同感。

The Cyberlearning and Future Learning Technologies Program funds efforts that support envisioning the future of learning technologies and advance what we know about how people learn in technology-rich environments. Cyberlearning Exploration (EXP) Projects explore the viability of new kinds of learning technologies by designing and building new kinds of learning technologies and studying their possibilities for fostering learning and challenges to using them effectively. This project addresses the effect of students' social identity on learning, an important factor in math and science education. Specifically, it will advance the scientific understanding of math identity (i.e. "I'm (not) a math person") by studying the over 100,000 diverse students who use Reasoning Mind, a blended learning system for K-8 mathematics with demonstrated results. Reasoning Mind supports math identity with an innovative design that allows students to email an animated character (aka the Genie) and receive a human-crafted response. This study will show how math identity manifests and changes during students' use of Reasoning Mind in order to inform software designers and classroom teachers on best practices for encouraging math identity. This will have the broader impact of strengthening our nation's ability to supply science and technology fields with a well-trained workforce.This study examines two components of math identity: self-efficacy and interest in mathematics. Both self-efficacy and interest can be enhanced by curricula that are individualized to appropriately challenge each student (a strength of educational technology), but social stereotypes (i.e. "Girls aren't good in math") may decrease math identity or otherwise interfere with its development. This study investigates how educational technology can reverse these trends. In the first phase of this study, a combination of survey methods, Natural Language Processing (NLP), and Educational Data Mining (EDM) techniques will be used to identify how poor and/or changing math identity emerges in the linguistic patterns of student interaction with GenieMail (as well as in other parts of Reasoning Mind). These findings will then be used to enhance GenieMail and other instructional interactions with the Reasoning Mind system by creating a hybrid human/AI system, with the goal of improving math identity across diverse populations of students.