Agile methods and approaches such as eXtreme programming (XP) have become the norm for successful organizations not only in the software industry but also for businesses seeking to improve internal software processes. Pair programming in some form is touted as a major functionality and productivity improvement. However, numerous studies show that simply placing two programmers side by side in front of a single computer screen is not enough. We must look at other factors such as programmer expertise, project preparation, and perceived solution quality to understand pair programming’s promises and pitfalls. In our study, we apply tailored programming challenges to a multifaceted group of first-year through senior Information Systems (IS) and non-IS majors to analyze how participant attitudes and perceived benefits of pair programming change from pre- to post-study, as well as determine whether the quality and functionality of the solutions differ across education levels and disciplines. Our findings show a strong interaction effect of gender and major composition (CIS vs. non-CIS majors) in all four dimensions of the ATMI attitude scale. Findings also suggest that experience in problem solving and solution formation are more important than prior specific domain knowledge. Finally, participants’ perceived ability, sense of accomplishment, and completion of the assigned work, regardless of background or demographic, determined their performance outcome on the pair-programming tasks, which suggests that not all forms of attitude and perceived benefits contribute to the performance outcome.
Chen, Kuanchin and Rea, Alan
"Do Pair Programming Approaches Transcend Coding? Measuring Agile Attitudes in Diverse Information Systems Courses,"
Journal of Information Systems Education: Vol. 29
Available at: https://aisel.aisnet.org/jise/vol29/iss2/2