Enhancing Programming Education: How Flowchart and AI Drive Motivation, Self‐Efficacy and Computational Thinking

ABSTRACT Background Programming education remains challenging for beginners due to the abstract nature of code and logic. Traditional instructional methods often fall short of engaging students or supporting those with low prior knowledge. Visual tools and AI have shown promise but are rarely integrated systematically into pedagogical frameworks. Objectives This study examined the effects of integrating Problem‐Based Learning (PBL) and Self‐Directed Learning (SDL) with a flowchart‐based tool (fChart) and a generative AI assistant (ChatGPT) on students' programming‐related affective and cognitive constructs (Learning Motivation (LM), Programming Self‐Efficacy (PSE), Computational Thinking (CT) and Course Satisfaction (CS)), as well as objective programming performance (test scores). Methods Using a quasi‐experimental design, 80 undergraduates were divided into high and low prior knowledge groups. Over 12 weeks, students learned Python programming through a blended PBL‐SDL approach, using a flowchart‐based tool (fChart) in the first phase and a generative AI assistant (ChatGPT) in the second. Data were collected at three time points (pre‐, mid‐ and post‐test) and analysed using repeated‐measures ANOVA, mixed‐design ANOVA and paired‐sample t ‐tests, as appropriate. Results and Conclusions Learning Achievement (LA), LM, PSE and CT improved across the instructional phases; whereas SDL increased, CS showed no significant change. Prior programming knowledge influenced overall performance levels, yet most constructs showed similar patterns of change across groups. These findings suggest that the sequential instructional design correlated with positive learning‐related changes.