Improving learning outcomes in advanced mathematics for underprepared university students through AI-driven educational tools

Abstract

This study investigates the impact of artificial intelligence (AI) tools on the academic performance of university students with insufficient mathematical preparation enrolled in higher mathematics courses. Utilizing a quasi-experimental design with pre- and post-testing, the research explores the effectiveness of AI-based platforms such as SOWISO, ChatGPT, and WolframAlpha in supporting underprepared learners. The study incorporates both quantitative and qualitative analyses and evaluates the mediating roles of academic motivation and attitudes toward AI, as well as the moderating effect of digital literacy and instructional support. Results indicate that AI interventions significantly improve learning outcomes, especially when integrated with structured pedagogical guidance and platforms that offer step-by-step feedback. The findings highlight the value of interactive AI environments in reducing academic anxiety, promoting learner autonomy, and fostering equitable access to advanced mathematical instruction. Implications for inclusive pedagogy and the role of AI in transforming higher education are discussed, along with recommendations for future research and practice.

Keywords: AI in education; higher mathematics; underprepared students; academic performance; digital literacy; instructional scaffolding; educational technology; motivation; inclusive pedagogy; cognitive support.

This article is published under the terms of the Creative Commons Attribution License 4.0