A Data-Driven Precision Teaching Intervention Mechanism to Improve Secondary School Students' Learning Effectiveness

The continuous development of Educational Data Mining (EDM) and Learning Analytics (LA) technologies has provided more effective technical support for accurate early warning and interventions for student academic performance. However, the existing body of research on EDM and LA needs more empirical studies that provide feedback interventions, and more attention should be paid to primary and secondary school students. This study proposed a data-driven precision teaching intervention mechanism combining EDM and LA technologies. The proposed mechanism aims to assist teachers in predicting students' academic performance and implementing corresponding interventions. This approach enables early warnings and reminders for students in crisis, and offers teaching assistance and support tailored to students at different levels. A quasi-experimental design was employed to examine the impact of the data-driven precision teaching intervention mechanism on secondary school students' learning outcomes. A total of 142 seventh-grade students participated in the intervention experiment, with an experimental group (50) receiving the data-driven precision teaching intervention, control group2 (48) receiving a group intervention stratified by teacher experience, and control group1 (44) receiving a traditional group intervention. Posttest data were collected after three rounds of intervention. Compared to the two control groups, students in the experimental group demonstrated superior academic achievement, intrinsic motivation, self-efficacy, and meta-cognitive awareness. These findings indicate that the data-driven precision teaching intervention approach positively impacted students' academic development, and effectively promoted their personalized learning. The findings provide pedagogical insights into the application of EDM in conjunction with LA prediction and actionable interventions.