Codeo - Learning support research: use of automated feedback within undergraduate programming education, a case study

Abstract

This paper describes the research process of a case study in the domain of pedagogical computer science, specifically in the context of how first year undergraduate students learn programming skills, with the purpose of investigating novel mechanisms to aide in engaging undergraduate students when learning to code, as well as add to an existing body of knowledge on computing pedagogy. The methodology employed in this endeavour was a threefold approach; stage one of the studies engaged undergraduate students in a series of data gathering exercises, such as a survey and a semi-structured interview. This allowed for the development of a proof of concept website prototype to engage students in programming exercises, before they would be automatically marked by a series of bespoke unit tests for each question. Finally, the third stage of research was gathering data on specific undergraduate experience with this research tool, where participants were asked to engage with the programming exercises, before then filling in another survey which asked questions prudent to the effectiveness of the research tool. The results of this case study show a generally positive response to interaction with the research tool, with participants highlighting that they believed the tool was useful, that feedback provided by the automated marking process was helpful, and that they would recommend such a learning system to their peers. Conclusions drawn from the case study suggest profound benefits to the use of tools such as the one implemented within this case study, as well as benefits to students seeing their peers as sources of knowledge in a cooperative manner.