Mathematics Education Students' Perceptions in Designing AI-Assisted Mathematics Problems: A Case Study of Agricultural Context

Abstract

The rapid advancement of generative artificial intelligence (AI) has opened new opportunities in education, particularly in supporting mathematics problem design. This study investigates the perceptions and intentions of prospective mathematics teacher students regarding the use of generative AI as a tool for designing context-based mathematics problems related to corn cultivation. Employing a mixed-methods approach with an exploratory sequential design, data were collected from mathematics education students enrolled in an Applied Mathematics course through questionnaires and semi-structured interviews. The research procedure implemented a series of hierarchical challenges in problem design, progressing from general agricultural contexts to more specific applications involving corn shelling technology. This design aimed to examine how students adapted their use of AI as the level of contextual complexity increased. Quantitative data were analyzed using descriptive statistics to examine students’ perceptions and intentions, and inferential statistics to identify differences based on demographic characteristics. Qualitative data were analyzed thematically to capture participants’ experiences and reflections. The findings indicate that students demonstrated highly positive perceptions of generative AI, recognizing it as a valuable tool for enhancing digital competence, supporting career development, and facilitating the design of contextual mathematics problems. Participants also reported high levels of comfort and strong intentions to integrate AI into their future teaching practices. However, the frequency of AI use was moderate, reflecting participants’ awareness of the technology’s limitations. No significant differences in perceptions or intentions were found based on gender or prior experience with corn cultivation. Qualitative results revealed that AI functioned as an effective collaborative and iterative tool for idea generation, while challenges such as limited contextual realism, rigid responses, potential logical errors, and concerns about overreliance were noted. The study concludes that while prospective teachers hold positive attitudes toward generative AI, there is a critical need to strengthen AI literacy, critical thinking, and ethical awareness through targeted integration in mathematics teacher education programs