Purpose: This study aims to improve the competence of teachers at SMA Muhammadiyah Kota Sukabumi in designing and implementing Science, Technology, Engineering, and Mathematics (STEM)-based learning to facilitate deep learning. Research Methodology: This community service used in-service training involving theoretical presentation, modeling, simulation, and feedback for approximately 10 teachers. Data from pre- and post-tests, observations, and product assessments were analyzed descriptively using quantitative and qualitative techniques.
Results: The results showed a significant improvement in teachers’ understanding of STEM and deep learning concepts, indicated by an increase in the average score from 58 to 84. Teachers were also able to develop STEM-based learning tools, such as teaching modules and student worksheets, categorized as good to very good. In addition, participant engagement was very high, with full attendance and active involvement during the training activities.
Conclusions: The STEM-based training using the in-service training model is effective in enhancing teachers’ conceptual understanding and practical skills in implementing deep learning-oriented instruction.
Limitations: The study was limited by a short training duration and the small number of participants, which may affect the generalizability of the results.
Contributions: This study contributes to the field of STEM education and teacher professional development by providing an effective training model to support deep learning implementation. It is particularly beneficial for teachers, schools, and educational institutions aiming to improve 21st-century learning practices.