University of Twente Student Theses


Apollo++: Automated Assessment of Learning Outcomes in Programming Projects

Rump, Arthur (2023) Apollo++: Automated Assessment of Learning Outcomes in Programming Projects.

[img] PDF
Abstract:Learning to program requires feedback, especially when it comes to complex issues like style and structure. At the University of Twente, courses tend to include a project which allows students to learn in practice and be assessed on their practical skills, but this poses a challenge for feedback: since students are working on their own projects with some freedom to make decisions, every project is different. With a large number of students, the only way to scale this feedback practice is with a large number of tutors, but this increases the chances of inconsistent assessments. Another option is to partially automate assessment, which is what we examine in this study. We introduce Apollo++, a tool for automated assessment built specifically for larger programming projects. The key features of Apollo++ are its basis in ILOs and assessment criteria, its support for different program representations and the goal of supporting rather than replacing manual assessment. We investigate the requirements for this tool through stakeholder interviews, propose an architecture, implement the core of this architecture with a prototype and use this prototype to evaluate the design.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science, 81 education, teaching
Programme:Computer Science MSc (60300)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page