Validating the defect detection performance advantage of group designs for software reviews: report of a replicated experiment

It is widely accepted that software development technical reviews (SDTRs) are a useful technique for finding defects in software products. The normative SDTR literature assumes that group reviews are better than individual reviews. Recent debates centre around the need for review meetings. This paper presents the findings of a replicated experiment that was conducted to investigate whether group review meetings are needed and why. We found that an interacting group is the preferred choice over the average individual and artificial (nominal) groups. The source of performance advantage of interacting groups is not synergy as was previously thought, but rather in discriminating between true defects and false positives identified by individual reviewers. As a practical implication, nominal groups may be an alternative review design in situations where individuals exhibit a low level of false positives.