Debugging in Novice Programmers:
A Multi-institutional, distributed research group

• LEADER: Sue Fitzgerald, Metropolitan State University
• Gary Lewandowski, Xavier University
• Renee McCauley, College of Charleston
• Laurie Murphy, Pacific Lutheran University
• Beth Simon, University of California, San Diego
• Lynda Thomas, Aberystwyth University
• Carol Zander, University of Washington, Bothell

• Ray Lister, University of Technology, Sydney
• Jan Erik Mostrom, Umea University

Background

Debugging can be difficult and frustrating for novice programmers. As educators, we have found that students' difficulties in developing debugging skills can be a significant barrier to success in beginning programming courses.

As computing education researchers, we began, in fall 2005, to conduct background literature review and analysis to see what research had been done which might be persuasive to us in changing the way we teach debugging or beginning programming. After more than a year of weekly meetings and review of more than 50 papers, we find ourselves remarkably unconvinced that current research results are directly applicable to what we want to know as educators.

Current Activities

As such, we are currently engaged in a number of projects to push forward in this area:

• Literature Review: We are in the process of completing a relevant literature review. This review will target work on novice debugging assessments and research. This review will be written from the point of view of the beginning programming educator.
• An ITICSE 2007 Working Group. We have proposed a working group to develop web-based videos that will document and expose the process of debugging for beginning programming students. The issue of debugging as a process (and that processes pose difficulties for teaching) has arisen as a central issue in our discussions.
• Novice Debugging Experiment and National Science Foundation Mini-Workshop. We are currently conducting a distributed experiment of novice debugging skills. This experiment, in some ways, replicates the work reported in Katz and Anderson (Katz, I. and Anderson, J. 1987. Debugging: An analysis of bug location strategies. Human- Comput. Interaction 3, 4, 351-399.). We seek to identify novice students' bug location strategies. We also wish to determine if locating bugs is the main source of student difficulty, or if they also have difficulty in fixing bugs once they have been located. Data collection will take place at seven colleges and universities during spring 2007. With support from the NSF, we will gather for a mini-workshop in early June to begin analyzing this data.