Cognitive engagement as a predictor of learning gain in Python programming

Abstract The relationship between cognitive engagement and learning gains in computer programming has not been well-studied. This study examined the relationship between students’ cognitive engagement and learning gains in the context of Python programming. Cognitive engagement, defined by the Interactive, Constructive, Active, Passive (ICAP) framework, refers to the level and depth of mental effort and involvement a learner invests in an educational activity. In this paper, we provide details about how students’ actions were classified into three levels of cognitive engagement. We studied these actions’ frequency and duration differences and performed regression analysis. The results revealed significant student diversity regarding frequency and time allocation to these engagement categories and highlight the complex interplay between students’ cognitive activities and their corresponding time investments. Further, the regression analysis results showed that the constructive and passive levels of engagement were significant predictors of students’ learning gains in the case of Python programming.These findings offer actionable insights into why some students may have lower learning gains. By examining the specific levels of cognitive engagement that lead to better learning outcomes, this study hopes to inform the development of more effective learning environments that support student engagement and improve programming education.