Adaptive learning framework for learning computational thinking using educational robotics

Abstract Computational thinking (CT) has been promoted worldwide by educational systems and is an essential skill for technological citizens. Various strategies have been planned and developed to help in introducing, improving, and delivering CT. One of the strategies is by creating and developing the supporting tools for CT learning. In this article, educational robotics (ER) is chosen as the focus tool to support CT learning. Each CT and ER has a massive field of study. There are various available reports determining the suitability of CT subject integrated with ER for students' learning. However, all students do not develop similar style of learning and thinking. There is difference in their personal traits. There is a lack of research that designed CT learning through ER specifically based on student's preferences. Besides, it resulted in a challenge to determine the suitability of CT and ER for different kind of preferences. Therefore, this study aimed to develop an adaptive learning (AL) framework for students to deliver learning of CT through ER. The framework consists of three submodels: domain model, student model, and adaptation model. One case study is defined, which is learning the introductory level of CT through ER (CTER). At the end of the study, it can be observed that the AL framework produced positive results in performance and perception for various student categories. It was noted that students utilizing the AL framework had superior understanding of CTER. Individually or collaboratively, all students who applied or did not apply the AL framework in studying the CTER introduction had positive learning outcomes.