Abstract:

This study investigates the predictive roles of computational thinking and task persistence on students’ achievement in a project-based Arduino robotics programming course. This study addresses a gap in understanding how cognitive and non-cognitive factors jointly influence learning outcomes in hands-on STEM environments. The study was grounded in Papert’s constructionism and employed a correlational survey design involving 73 second-year undergraduate students enrolled in Robotics Programming II across two universities in Southeast Nigeria. Data were collected using three validated instruments, namely: the Robotics Achievement Test (RAT), Computational Thinking Scale (CTS; α = 0.88), and Task Persistence Scale (TPS; α = 0.70). Multiple regression analysis was used to determine the predictive relationships among the variables. The results showed that both computational thinking and task persistence significantly predict students’ achievement. Computational thinking demonstrated a moderate positive relationship with achievement (R = 0.421, R² = 0.177, F(1,71) = 15.285, p < .001), while task persistence similarly showed a significant predictive effect (R = 0.419, R² = 0.175, F(1,71) = 15.076, p < .001). These results show that each variable independently explains approximately 17% of the variance in academic performance. The study advances current knowledge by empirically validating the dual importance of cognitive (computational thinking) and non-cognitive (task persistence) factors within project-based robotics education. The findings suggest that effective instructional design in Arduino-based learning environments should integrate explicit computational thinking scaffolds alongside strategies that enhance persistence, such as iterative problem-solving and resilience-building practices. The study concludes that optimising both dimensions can enhance students’ achievement in programming courses and recommends further research incorporating additional predictors (e.g., self-efficacy and prior knowledge) to develop more comprehensive models of student success in STEM education.

Keywords:

Computational thinking; task persistence; robotics programming; Arduino; project-based learning; student achievement.