基于准实验的AI驱动警务专业个性化教学实证研究

A Quasi-Experimental Empirical Study of AI-Driven Personalized Instruction in Policing Education

摘要: 在人工智能快速发展的背景下，如何通过个性化教学改革提升人才培养质量已成为高校教育的重要议题。本文以高校警务专业为研究对象，系统设计了基于AI的个性化教学方案，涵盖课程内容智能推荐、多元化教学模式、智能评测与反馈以及虚拟场景模拟等环节，并在实际教学中开展应用与验证。研究方法采用准实验设计，以J校警务专业2021~2024学年学生为样本，设置实验组与对照组，开展多维度实证分析。结果显示，AI教学显著提升了学生语言能力(t = 7.85, p < 0.001, d = 1.03)、课堂互动(IRR = 1.42, p < 0.001)与学习公平性(差距缩小23.5%)，但理论课程成绩提升有限(d = 0.38)。研究指出，AI驱动的方案在提升学生成绩、强化语言能力、促进课堂互动与协作、缩小学习差距方面均取得显著成效，但在逻辑建构型课程中支持略显不足。

Abstract: Against the backdrop of rapid advancements in artificial intelligence, enhancing talent cultivation quality through personalized teaching reforms has become a critical issue in higher education. This study focuses on university police science programs, systematically designing an AI-based personalized teaching plan encompassing intelligent course content recommendations, diversified teaching modes, smart assessment and feedback, and virtual scenario simulations. The plan was implemented and validated in actual teaching settings. Employing a quasi-experimental design, the research analyzed students from J University’s police science program during the 2021~2024 academic years, establishing an experimental group and a control group for multidimensional empirical analysis. Results indicate that AI-driven instruction significantly enhances students’ language proficiency (t = 7.85, p < 0.001, d = 1.03), classroom engagement (IRR = 1.42, p < 0.001), and learning equity (23.5% reduction in achievement gaps), though improvements in theoretical course grades remain limited (d = 0.38). The study indicates that AI-driven solutions achieved significant effectiveness in enhancing student performance, strengthening language proficiency, promoting classroom interaction and collaboration, and narrowing learning gaps. However, their support proved somewhat inadequate in logic-construction-based courses.