Collaborative Research: RAPID: Empowering Math Teachers with an AI Tool for Auto-Generation of Technology-Enhanced Assessments

人工智能(AI)和强大的大型语言模型(LLM)的激增在K-12教师中引起了兴奋和困惑，他们对AI对教学、评估和学生工作的影响感到兴奋和困惑。对于在以人为中心的教学和学习方面具有专业知识的研究人员来说，分享教师-人工智能团队的经验证据是至关重要的，以提高教师的能力，为所有学生提供更好的学习结果。这个快速项目考察了在高中数学课堂上使用LLM驱动的工具如何使教师能够创建复杂的技术增强评估(TEA)，以形成性地衡量更高层次的思维技能，并促进更深入的学习。编写这类茶的任务需要编程专业知识和广泛的技术技能，因此大多数K-12数学教师无法参与。通过展示教师-人工智能团队的示范模型，该项目解决了一个关键的、及时的需求，即建立一个早期、积极的叙事，将教师置于K-12教育人工智能革命的中心。这项建议是为了回应尊敬的同事信(DCL)：在正式和非正式环境中迅速加快对K-12教育中人工智能的研究(NSF 23-097)，并由学生和教师创新技术体验计划(ITEST)资助，该计划支持建立对实践、计划元素、背景和过程的理解的项目，这些项目有助于提高学生对科学、技术、工程的知识和兴趣。这个时间敏感的项目将通过使用Edfinity软件推动关于在高中数学课堂教学中使用人工智能和最小二乘管理的研究，Edfinity软件使用开源网络格式生成交互式、自动评分和技术增强的形成性评估，以支持学生的学习。在这种情况下，教师将使用LLM工具(Alice)描述数学问题，该工具经过培训，使用自然语言输入，生成用于教学的源代码以及提示和学生反馈。该项目汇集了一个由数学教育工作者、STEM教育和学习科学研究人员、K-12教师教育工作者、人工智能工具开发人员和人工智能专家组成的多学科团队，以审查在印第安纳州和伊利诺伊州34所农村、城市和郊区学校将Alice整合到高中有限数学课程中的情况。该项目将培训高中教师并收集数据，以促进和塑造我们对教师-AI团队和特定领域的LLM提示工程的理解。研究包括从平台收集关于教师ALICS使用和即时工程的日志数据，以及通过调查和访谈收集教师反馈。这些数据将被分析，以了解教师使用LLM工具的经验，对教师对人工智能的态度和信心的影响，以及从教师的角度在生成高质量、交互式、形成性评估方面的成功。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The proliferation of artificial intelligence (AI) and powerful Large Language Models (LLMs) has evoked excitement and confusion among K-12 teachers regarding AI's impact on teaching, assessments, and student work. It is vital for researchers with expertise in human-centered teaching and learning to share empirically grounded proofs-of-concept of teacher-AI teaming to enhance teacher capacities for better learning outcomes for all students. This RAPID project examines how the use of LLM-powered tools in high school math classes empowers teachers to create complex technology-enhanced assessments (TEAs) that formatively measure higher-order thinking skills and facilitate deeper learning. The task of authoring such TEAs has necessitated programming expertise and extensive technical skills and thus excluded most K-12 math teachers from participating. By showcasing an exemplary model of teacher-AI teaming, this project addresses a crucial, timely need to establish an early, positive narrative that places teachers at the center of the AI revolution in K-12 education. This proposal was received in response to the Dear Colleague Letter (DCL): Rapidly Accelerating Research on Artificial Intelligence in K-12 Education in Formal and Informal Settings (NSF 23-097) and funded by the Innovative Technology Experiences for Students and Teachers (ITEST) program, which supports projects that build understandings of practices, program elements, contexts and processes contributing to increasing students' knowledge and interest in science, technology, engineering, and mathematics (STEM) and information and communication technology (ICT) careers.This time-sensitive project will advance research on the use of AI and LLMs for teaching in high school mathematics classrooms through the use of Edfinity software that uses the open-source WeBWorK format to generate interactive, auto-gradable, technology-enhanced formative assessments to support student learning. Within this context, teachers will describe a math problem with an LLM tool (ALICE) that is trained to use natural language inputs, generating source code for TEAs along with hints and student feedback. The project brings together a multidisciplinary team of math educators, STEM education and learning sciences researchers, K-12 teacher educators, AI tool developers, and AI experts to examine the integration of ALICE into high school Finite Mathematics courses across 34 rural, urban, and suburban schools in Indiana and Illinois. The project will train high school teachers and gather data to advance and shape our understanding of teacher-AI teaming and domain-specific LLM prompt engineering. The research involves gathering log data from the platform on teachers' ALICE usage and prompt engineering, as well as teacher feedback through surveys and interviews. These data will be analyzed to understand teachers' experiences in using an LLM tool, the impacts on teacher attitudes toward and confidence in AI, and the success of ALICE from teachers' perspectives for the generation of quality, interactive, formative assessments.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.