The teaching of novice computer programmers: Bringing the scholarly-research approach to Australia

BRACElet is a multi-institutional multi-national research study of how novice programmers comprehend and write computer programs. This paper reviews the first action research cycle of the BRACElet project and, in the process, charts a path for the upcoming second cycle. The project remains close to educational practice, with much of the data being either data collected directly from exams sat by novices, or data from think-out-loud protocols where the task undertaken by a novice or an expert is modelled on an exam question. The first action research cycle analysed data in terms of the SOLO taxonomy. From think-aloud responses, the authors found that educators tended to manifest a SOLO relational response on small reading problems, whereas students tended to manifest a multistructural response. Furthermore, those students who manifested a relational response tended to do better overall in the exam than students who manifested a multistructural response. The second action research cycle will explore the relationship between the ability to read code and the ability to write code. Apart from reporting on the BRACElet project itself, this paper serves as an invitation for institutions and individuals to join the second action research cycle of the BRACElet project.