The Impact of Peer-based Software Reviews

Peer-based software review techniques such as pair programming and peer code review are increasingly used within information systems development teams. How do these techniques influence team performance? While previous research has treated the decision to apply peer-based review techniques as a trade-off between increased effort and improved code quality, this study provides a deeper theoretical understanding of how these techniques affect team performance. Based on extant literature on team feedback and transactive memory systems, a theoretical model is developed which links pair programming and peer code review to different forms of feedback. Depending on task complexity, different feedback mechanisms are suggested to influence team performance directly through improved code quality, but also indirectly by strengthening teams’ transactive memory systems. Findings from ongoing extensive empirical research are expected to contribute to existing literature on peer-based software review techniques, which mostly examined effects on the performance of individuals in experimental settings.