Using Structural Equation Modeling to Analyze Impact Factors on Students and AI Agents Interaction Effectiveness in Elementary English Classes

A generative classroom relies on a novel interactive environment designed to achieve predefined educational objectives and generate additional values. The rapid evolution of AI Agent technology has presented immense potential for advancing the optimization and transformation of generative teaching methodologies. Initially, drawing upon collaborative interaction theory, this study established a framework of AI agent-supported human-computer interaction (HCI) learning environment, aiming to thoroughly reveal the mechanism of HCI within elementary English generative classrooms. Subsequently, based on this framework and grounded in dialogic education theory and information ecology theory, the study proposed some hypotheses regarding the impact factors on the students and AI agent interaction effectiveness in such classrooms. To test these hypotheses, empirical research was conducted utilizing structural equation modeling. The findings revealed that content usefulness, content ease of use, perceived enjoyment, teacher guidance, interaction atmosphere, technological facilitation, and interaction willingness are crucial factors in interaction effectiveness between students and AI agents in elementary English generative classrooms. Notably, interaction willingness often served as a mediating variable influencing interaction effectiveness. Based on these insights, this study proposed the following strategies. a). Leverage DeepSeek's performance advantages to drive innovation and upgrade of AI agents. b). Optimize corpus processing mechanisms to enhance the content value of AI agents. c). Build the interaction environment to enhance students' willingness to interact with AI agents. Ultimately, this study aimed to provide theoretical guidance and practical insights for the efficient application of AI agents in elementary English classes.