Adaptive learning model for learning computational thinking through educational robotic

Abstract

Computational thinking (CT) has been promoted worldwide by educational systems and is an essential skill for technological citizens. In delivering CT, various kinds of educational tools were developed by researchers to support the learning. One of the attractive tools in providing the CT is educational robotic (ER). However, delivering CT to students through ER has many challenges. There is a lack of studies presenting the general view on the integration of ER and CT as both subjects have big scope in terms of teaching and learning. Thus, this study designed a conceptual data model to represent the relationship between CT and ER. In addition to the complexity in determining the suitability of both subjects for students’ learning, students also have differences in their personal traits, resulting in different learning styles and thinking styles. Therefore, this study aimed to enhance an adaptive learning (AL) model for students, which is based on the students’ learning style and knowledge level. The enhanced AL model comprised three sub-models: domain model, student model, and adaptation model. Two case studies were selected, which are learning advance of CT and the introductory of computational thinking through educational robotic (CTER). At the end of the study, it can be observed that the enhanced AL model produced positive results in performance and perception for various student categories. In learning advanced CT, both groups of students exhibited a positive perception of using the AL model. Nevertheless, the group of students who applied the enhanced AL model outperformed the other group in term of performance. Additionally, in learning CTER, it can be observed that students had a good perception in using enhanced AL model, while the group of students who either applied AL model or did not in learning CTER introduction had a good result towards the learning performance. In conclusion, this study showed that the enhanced AL model could improve learning performance, especially for learning advanced CT and can be used for learning CTER.