Given the inadequacy of assessed outcomes (e.g., final exam) and the importance of evaluating the learning process in STEM education, we use deep learning to develop the STEM learning behavior analysis system (SLBAS) to assess the behavior of learners in STEM education. We map learner behavior to the ICAP (interactive, constructive, active, passive) framework, helping instructors to better understand the learning process of learners. The results show that SLBAS exhibits high accuracy. Moreover, Cohen’s kappa coefficient between expert coding and SLBAS is high enough to support replacing expert coding in the observation method with SLBAS to recognize the learning process of learners during STEM activities. Finally, statistical analysis establishes a correlation between the learning process and learning effectiveness. The results of this study are in line with most previous studies, demonstrating that STEM education differs from traditional teacher-centered courses in that it helps learners to improve the process of knowledge construction with practice and hands-on opportunities rather than simply receiving knowledge passively.

Background In the realm of Science, Technology, Engineering, and Mathematic (STEM) education, computer programming stands as a vital discipline, amalgamating cross-disciplinary knowledge and fostering the capacity to solve real-world problems via fundamental concepts and logical methodologies inherent to computer science. Recognizing the important of computer programming, numerous countries have mandated it as a compulsory course to augment the competitiveness of K-12 learners. Nevertheless, the inherent complexity of computer programming for K-12 learners often goes unacknowledged. Constraints imposed by the course format, coupled with a low instructor–learner ratio, frequently inhibit learners’ ability to resolve course-related issues promptly, thereby creating difficulties in the affective domain. While precision education tools do exist to ascertain learners’ needs, they are largely research-oriented, thereby constraining their suitability for deployment in pragmatic educational settings. Addressing this issue, our study introduces the precision education-based timely intervention system (PETIS), an innovative tool conceived to enhance both programming skills and affective learning in K-12 learners. Our research investigates the influence of PETIS on learners’ performance and evaluate its efficacy in facilitating computer programming education in K-12 environments. Results Quantitative results demonstrate that the application of the precision education-based timely intervention system (PETIS) proposed by this research significantly improves programming skills and affective-domain learning objectives for K-12 learners. Similarly, qualitative results indicate that PETIS is beneficial for both teaching and learning in K-12 computer programming courses. Conclusions These results not only confirm that timely intervention and feedback improve K-12 learners’ programming skills and affective-domain learning objectives in computer programming courses, but also yield implications as to the feasibility of applying precision education in real-world STEM scenarios.

Maker education that incorporates computational thinking streamlines learning and helps familiarize learners with recent advances in science and technology. Computational thinking (CT) is a vital core capability that anyone can learn. CT can be learned through programming, in particular, via visual programming languages. The conclusions of most studies were based on quantitative or system-based results, whereas we automatically assessed CT learning progress using the Scratch visual programming language as a CT teaching tool and an integrated learning tracking system. The study shows that Scratch helped teachers to diagnose students’ individual weaknesses and provide timely intervention. Our results demonstrate that learners could complete tasks and solve problems using the core CT steps. After accomplishing numerous tasks, learners became familiar with the core CT concepts. The study also shows that despite increased learning anxiety when solving problems, all learners were confident and interested in learning, and completed each task step by step.

No abstract

No abstract