An Intelligent Assistant for Programming Coursework Assessment and Feedback

Country: United Kingdom;

Deadline: 31.07.2015

Web-Site: findaphd.com

 

Recent National Student Surveys revealed that many UK university students are dissatisfied with the timeliness and usefulness of the feedback received from their tutors. Ensuring timeliness in marking often results in a reduction in the quality of feedback. For example, automatic coursework assessment systems can give instant feedback to students based on the results of verifying program correctness, but they are unable to provide feedback that will help students to learn from their mistakes. In Computer Science where learning relies on practising and learning from mistakes, feedback that pin-points errors and explains means of improvement is important to achieve a good student learning experience. Though suitable use of Information and Communication Technology should alleviate this problem, existing Virtual Learning Environments and e-Assessment applications such as Blackboard, BOSS, MarkTool and GradeMark are inadequate to support a coursework assessment process that promotes timeliness and usefulness of feedback while maintaining consistency in marking involving multiple tutors.

eCAF is a novel Internet application for facilitating an efficient and transparent coursework assessment and feedback process. The eCAF system supports detailed marking scheme editing and enables tutors to use such schemes to pin-point errors in students’ work so as to provide helpful feedback efficiently. Tutors can also highlight areas in a submitted work and associate helpful feedback that clearly links to the identified mistakes and the respective marking criteria. Early experience indicates that an efficiency gain of 20% could be achieved when using eCAF to assess and give detailed feedback to a class size of 70. However, as eCAF is designed to support human marking, its efficiency is bound by the efficiency of the tutors.

This project aims at investigating means to promote timeliness in coursework assessment while maintaining a high quality of feedback through a semi-automatic process of coursework assessment and feedback.
Experience shows that, on average, giving detailed and helpful feedback to a submission of a non-trivial programming coursework by a human tutor takes about 30 minutes. This process is often longer at the start as the tutor familiarises himself/herself with the marking scheme and the potential issues in the submitted work. As the range of errors are often limited, the marking process typically speeds up after marking 10-20 pieces. Marking coursework electronically makes it possible to record all mistakes identified and feedback in a database as the marking proceeds. This information can potentially help making the marking process more efficient while enabling the use of tutor’s style of assessment and feedback, rather than being solely dependent on assessing program correctness. This project investigates means to utilise the recorded coursework assessment and feedback data to support a semi-automatic assessment and feedback process. For example, the system could detect mistakes that are similar to previously recorded mistakes and offer the tutor to apply the same feedback as a starting point. Such a feature will depend on a non-trivial analysis of the submitted programs. The analysis is expected to be on the level of both program structure and natural language semantics.

This studentship is combined with a teaching assistant role. The successful candidate will be required to carry out up to seven hours per week of teaching support for a distance learning programme under the direction of the project supervisor; therefore the student must be capable of teaching on a specific module. Details of teaching responsibilities and a list of taught modules can be found on the Aston University Website by searching for reference R140307.
 


Funding Notes:

This studentship includes a fee bursary to cover the home/EU fees rate plus a maintenance allowance of £13,863 in 2015/16. Applicants from outside the EU may apply for this studentship but will need to pay the difference between the ‘Home/EU’ and the ‘Overseas’ tuition fees, currently this is £11,448 in 2015/16.