Teaching novices to debug

Download files
Access & Terms of Use
open access
Copyright: White, Andrew
Altmetric
Abstract
This thesis considers how to teach novices to debug computer programs. The investigation is specifically aimed at students in first year of computer science at university. The goal is to find an effective method of improving the debugging ability of novice subjects. Novices are less effective than experts. Although weaker across the board, novices display a critical lack of strategic skills in the debugging domain. Novices become lost and discouraged easily. Novices do not make good use of program structure, do not use test cases intelligently, and do not track their progress. Even a slight improvement in these aspects has the potential to significantly improve performance. Expert debugging behaviour was examined. This provided the basis for a model of debugging to be used in instructing novices. Teaching materials were developed to instruct and support students in following the model. The model and training were then tested in both laboratory and classroom situations. Results were mixed, with a generally positive trend. The model by itself was not overly effective. When the model was accompanied by training materials containing examples and scaffolding, students who were willing to spend time examining the training usually showed some improvement. In a few cases, the student's debugging style was dramatically improved. The improved performance seems to come partially from improved skill and partially from motivation. The model and materials show potential, and are likely to be more effective when used in an environment that allows interaction, rather than the current paper-based, non-interactive tutorial.
Persistent link to this record
Link to Publisher Version
Link to Open Access Version
Additional Link
Author(s)
White, Andrew
Supervisor(s)
Quinn, Clark
Shepherd, John
Creator(s)
Editor(s)
Translator(s)
Curator(s)
Designer(s)
Arranger(s)
Composer(s)
Recordist(s)
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
2009
Resource Type
Thesis
Degree Type
PhD Doctorate
UNSW Faculty
Files
download whole.pdf 670.2 KB Adobe Portable Document Format
Related dataset(s)