An experiment to investigate interacting versus nominal groups in software inspection

  • Authors:

  • Diane Kelly;Terry Shepard

  • Affiliations:

  • Royal Military College of Canada;Royal Military College of Canada

  • Venue:
  • CASCON '03 Proceedings of the 2003 conference of the Centre for Advanced Studies on Collaborative research
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Votta's 1993 question, "Does Every Inspection Need a Meeting?" signaled an examination of an important activity in inspection. Fagan considered meetings to be the primary defect detection activity, but research has generated results that both support and conflict with this. It is generally agreed that a better understanding is needed. Research on group behaviour has motivated Sauer et al to formulate eleven propositions regarding group performance in inspections. These are partially supported by existing research. The research described in this paper addresses the performance of the group versus the performance of the individual in the context of the use of a structured inspection technique. The main conclusion is that most findings eliminated as a result of meetings are subtle findings, while a smaller number added as a result of meetings are findings that are relatively easy to identify.