aied2003.pdf (802.23 kB)
Download fileTowards a debugging tutor for object-oriented environments
chapter
posted on 2023-06-07, 13:22 authored by Benedict du Boulay, Pablo Romero, Richard Cox, Rudi LutzProgramming has provided a rich domain for Artificial Intelligence in Education and many systems have been developed to advise students about the bugs in their programs, either during program development or post-hoc. Surprisingly few systems have been developed specifically to teach debugging. Learning environment builders have assumed that either the student will be taught these elsewhere or thatthey will be learnt piecemeal without explicit advice.This paper reports on two experiments on Java debugging strategy by novice programmers and discusses their implications for the design of a debugging tutor for Java that pays particular attention to how students use the variety of program representations available. The experimental results are in agreement with research in the area that suggests that good debugging performance is associated with a balanced use ofthe available representations and a sophisticated use of the debugging step facility which enables programmers to detect and obtain information from critical momentsin the execution of the program. A balanced use of the available representations seemsto be fostered by providing representations with a higher degree of dynamic linkingas well as by explicit instruction about the representation formalism employed in the program visualisations.
History
Publication status
- Published
Publisher
University of SydneyPage range
399-407Pages
9.0Book title
Supplementary Proceedings of Artificial Intelligence in Education Conference (AIED2003), Sydney, AustraliaPlace of publication
Sydney, AustraliaISBN
9781864875720Department affiliated with
- Informatics Publications
Full text available
- Yes
Peer reviewed?
- Yes