Generative artificial intelligence (genAI) is currently being used in varied ways within educational contexts to support teacher and student learning. In this study, we examine approaches to evaluate the technical and ethical aspects of one genAI digital teaching simulation designed to help teachers learn how to elicit student thinking: (a) the accuracy of the genAI student’s responses, (b) variability in participants’ ability to identify the genAI student’s procedural and conceptual understanding, and (c) participants’ perceptions of the genAI simulation’s authenticity and usability. A convenience sample of 35 adult participants across the United States participated in this study during the fall of 2024. Findings suggest that the genAI student’s responses were generally appropriate, relevant, and aligned with the intended student profile, with minor inconsistencies observed. Most participants recognized one or more aspects of the genAI student’s strategy and indicated that the genAI student’s responses were realistic, grade-level appropriate, on-topic, timely, and coherent. We argue that attending to technical and ethical aspects of AIED systems is critical to ensuring that genAI tools provide meaningful learning opportunities within teacher education.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Evaluating the Use of Generative Artificial Intelligence to Support Learning Opportunities for Teachers to Practice Engaging in Key Instructional Skills

  • Jamie N. Mikeska,
  • Beata Beigman Klebanov,
  • Aakanksha Bhatia,
  • Shreyashi Halder,
  • Michael Suhan

摘要

Generative artificial intelligence (genAI) is currently being used in varied ways within educational contexts to support teacher and student learning. In this study, we examine approaches to evaluate the technical and ethical aspects of one genAI digital teaching simulation designed to help teachers learn how to elicit student thinking: (a) the accuracy of the genAI student’s responses, (b) variability in participants’ ability to identify the genAI student’s procedural and conceptual understanding, and (c) participants’ perceptions of the genAI simulation’s authenticity and usability. A convenience sample of 35 adult participants across the United States participated in this study during the fall of 2024. Findings suggest that the genAI student’s responses were generally appropriate, relevant, and aligned with the intended student profile, with minor inconsistencies observed. Most participants recognized one or more aspects of the genAI student’s strategy and indicated that the genAI student’s responses were realistic, grade-level appropriate, on-topic, timely, and coherent. We argue that attending to technical and ethical aspects of AIED systems is critical to ensuring that genAI tools provide meaningful learning opportunities within teacher education.