This is the authors' abstract. We don't add key points for this paper.
This article explores the use of adaptive learning platform (ALP) data to conduct early identification and provide support to students who have a low‐performance outcome (C or lower) in a numerical methods engineering course. The data from assigned ALP lessons for two semesters was used to create decision‐tree models to identify students who would benefit from advising and tutoring support. In the following two semesters, low‐performing students were identified early in the semester and provided with support, and their performance was compared to their peers. The best‐performing prediction model achieved an accuracy of 85% in predicting low‐performing students in the third week of the course. The support included weekly one‐on‐one tutoring and advising sessions. Although only 23% of the identified students accepted support, they scored one‐third a letter grade better than those who did not. Additionally, students who received support were invited to participate in a focus group at the end of the semester. Positive outcomes reported included improved understanding of course material, higher academic performance, advice on learning strategies, and guidance on non‐course‐related topics like internships and employment. Although most students valued receiving personalized invitations, a few felt singled out as low‐performing. Students acknowledged the significance of individualized support, gave advice on how to word the invitation emails, and made helpful suggestions for improving help sessions, particularly in terms of personalization and recognizing their heavy academic workload.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Kaw et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: