Theory Inspires, but Examples Engage: A Mixed-Methods Analysis of Worked Examples From CoderBot in Programming Education

Programming has become increasingly important in our society. However, the learning process presents significant challenges, particularly for novice students of introductory courses. From the students’ perspective, programming concepts are often perceived as complex and challenging to understand. Chatbots have emerged as promising and effective pedagogical agents, offering continuous support and personalized feedback throughout the programming learning process. In this paper, we present CoderBot, a pedagogical agent grounded in Example-Based Learning designed to assist novice students in comprehending programming concepts using correct and erroneous practical examples. To evaluate the self-efficacy and acceptance of CoderBot in the classroom, we conducted an exploratory study involving 103 undergraduate students from several regions of our country, all of whom were enrolled in introductory programming courses. The quantitative findings highlight the ease of use associated with CoderBot, along with noticeable improvements in students’ understanding of programming concepts and increased levels of motivation and self-confidence. Moreover, the qualitative results indicate that CoderBot holds the potential to be an effective pedagogical agent for supporting programming instruction, particularly in terms of clarity, accessibility, and ongoing assistance. However, the findings also suggest the need for further expansion of the available examples and improvements in the clarity of responses to realize the tool’s educational potential fully. These results offer valuable insights into integrating chatbots within academic environments, underscoring the role such tools can play in enhancing the learning experience for programming students.