Enabling automated assessment and feedback for 1st-year Informatics UG programming assignments

Project Overview

Teaching programming--a core element of any Computer Science education--requires constant, timely and high-quality feedback. This becomes more difficult as student numbers increase. Informatics have taken on 340 new UG students in 2015/6 and are projected to take on another 280 new students for 2016/7. It is, therefore, necessary to provide better support for the teaching of programming in Informatics UG programmes at King’s. Automated assessment and feedback technology offers an opportunity to improve the quality of feedback provided while increasing the quantity and continuity of such feedback without increasing the resource demand per student.

We are currently developing a flexible platform for automated assessment of programming tasks. This will be brought to production readiness in June, but will require instantiation for specific modules. In this project, we will develop a key component for this platform which will enable the automated assessment of Java-based programming tasks using automated tests. Java is a core programming language in the King’s Informatics curriculum and a top-ranking programming language in the software industry. Support for automated assessment of Java-based programming assignments, thus, has the potential to impact a wide range of Informatics modules across the entirety of our UG programmes and year groups, ensuring impact on a large group of NMS UG students. The completed system will also be made available to other groups across NMS and the College to support programming education in other Departments. Moreover, the system developed through this project will be a key enabler towards more innovative approaches towards automated assessment and feedback.

Project Progress

29/09/2016

Nexus is now up and running and ready to accept submissions from King's first-year students.

31/05/2016

The project has now been accepted for funding and we're looking for a student to start working on this.

Publications

[-] 2018

Steffen Zschaler, Sam White, Kyle Hodgetts, and Martin Chapman: Modularity for Automated Assessment: A Design-Space Exploration. Workshop Software Engineering für E-Learning-Systeme (SEELS), GI-Tagung "Software Engineering 2018", 2018.
[pdf] [http] Refereed Workshop Paper,  research areas: Modularity, Teaching
Research contribution:  Writing contribution:

[+] 2017