Evaluating primary students’ motivation and computational thinking in scratch-based learning: a confusion matrix analysis

Abstract

This study examined the relationship between student motivation and computational thinking (CT) skills within a Scratch-based learning environment for primary school students. Utilizing a quantitative research design with a pretest-posttest framework, the research involved 28 primary school students engaged in a computational learning program centered on the Jumping Bean concept. A confusion matrix analysis was employed to assess the predictive relationship between motivation levels and improvements in CT skills. The results showed that motivation is a reliable predictor of CT gains, with high precision indicating that highly motivated students are very likely to demonstrate measurable progress. However, the recall score suggests motivation alone is not a conclusive factor, as some motivated students did not achieve the expected CT improvements. This implies that other instructional elements, such as prior knowledge, cognitive differences, teaching methods, and learning design, also significantly impact outcomes. The implications of this research suggest that educators should cultivate motivating learning environments to foster students’ CT skills effectively. Recommendations include integrating gamified elements and personalized feedback to enhance student engagement and motivation in computational learning contexts.