Presentation Schedule
Predicting STEM Learning Outcomes Using Ensemble Machine Learning: Evidence from E-Microlearning Intervention (110149)
Session Chair: Cristina Pita Yáñez
Sunday, 12 July 2026 13:45
Session: Session 3
Room: UCL Torrington, B08 (Basement Floor)
Presentation Type:Oral Presentation
This study aims to predict student outcomes following an e-microlearning intervention in STEM among learners in disadvantaged areas of Sabah, Malaysia. An ensemble decision tree–based machine learning approach was employed to predict student scores and the probability of achieving a passing mark (≥70%). Primary data from 366 eighth-grade students were analysed to identify the most significant predictors of STEM performance. The results indicate that the random forest model outperformed gradient boosting, achieving a root mean square error (RMSE) of 1.964 for regression and an F1 score of 0.62 for classification after hyperparameter tuning. Key predictors of performance include students’ awareness of interdisciplinary connections between science and non-science subjects, as well as parental support. Additionally, investigative skills were identified as a key psychological factor influencing learning outcomes. These findings provide evidence-based insights for educators and policymakers to design targeted interventions aimed at improving STEM achievement among students in disadvantaged contexts, thereby contributing to broader national and global education goals.
Authors:
Sarimah Surianshah, Universiti Malaysia Sabah, Malaysia
Wong Sing Yun, Universiti Malaysia Sabah, Malaysia
Sarah Bridges, University of Nottingham, United Kingdom
Aniah Izzah Atirah Ahmad, Universiti Malaysia Sabah, Malaysia
About the Presenter(s)
Sarimah Surianshah is currently a Senior Lecturer at Universiti Malaysia Sabah (UMS), Malaysia.
See this presentation on the full schedule – Sunday Schedule
Conference Comments & Feedback
Place a comment using your LinkedIn profileShare this Presentation
Comments
Powered by WP LinkPress