The number line (NL) is an important tool in mathematics education. Students with mathematical difficulties (MD) tend to have difficulties in NL tasks, as indicated by qualitative analyses of eye-tracking data. However, these qualitative analyses are laborious especially for large amounts of data. Our paper uses an innovate approach to facilitate the analysis of student strategies: AI is used to support the human researchers. We use student gaze heatmaps in combination with AI, in particular a clustering algorithm, to identify strategies of 140 fifth-grade students on the NL. Through AI-enhanced analysis, we found, first, a set of student NL strategies different from previous research. Second, we found that students with and without MD—at certain numbers—differed in strategy use, which was not found in this way before.
This project has received funding by the Federal Ministry of Education and Research as a part of the program KI-ALF (01NV2123).