Topics, scaffolding and instrumentation to teach and learn introductory programming through debugging

Informatics education is increasingly required to equip individuals to fully participate in society and to “interface” with modern reality. Accordingly, Informatics education is seeing a surge in widespread adoption across all curricular levels. Adoption alone, however, does not suffice to ensure that learners meaningfully engage with a discipline that is largely perceived and frequently reported as difficult to teach and learn alike. This dissertation addresses this difficulty from the three perspectives of “topics”, “scaffolding” and “instrumentation” in the context of introductory programming teaching and learning in primary schools. After setting a comprehensive research background to our problem domain, we aggregate the results of multiple (experimental and theoretical) studies to analyze and expand upon learning approaches targeted at introductory programming in block-based programming environments. Regarding the “topics” perspective, we find evidence to support the notion that focusing on debugging skills is a support to introductory programming. We argue that tracing, the ability to understand and mentally execute programs, is central to this learning and break down the complex skill of debugging to constituent subskills, in which tracing plays a major role, that may be more manageable in educational contexts. Regarding “scaffolding”, we employ the vast body of research on Cognitive Load Theory to analyze and design scaffolding capable of augmenting learning environments that are already scaffolded to some extent, perhaps in more traditional ways, such as programming microworlds targeted at novices. Regarding “instrumentation”, we discuss the potential of using the fine-grained data collection capabilities of digital environments to build programming trajectories. These trajectories allow analyzing programming and debugging processes on top of products, shedding light on the application of skills, such as debugging, that may disappear when "compressed" in static programming artifacts.