Computing Reviews

A Cognitive-Based Mechanism for Constructing Software Inspection Teams
Miller J., Yin Z. IEEE Transactions on Software Engineering30(11):811-825,2004.Type:Article
Date Reviewed: 06/22/05

Software inspection is gaining recognition as an important way to assure software quality, to be used in addition to testing. A drawback of a team approach to inspection is that members tend to identify similar defects. This paper proposes that team members should be as diversified in cognitive styles as possible. It attempts to find a causal relationship between inspection results and the diversity of the personalities of team members. Two experiments were conducted to verify the hypothesis. Myers-Briggs type indicators were used, including extroversion versus introversion, sensing versus intuition, thinking versus feeling, and judging versus perception.

Unfortunately, no conclusive inference could be drawn from the experiments. An initial experiment and an internal replication produced opposite effects. The work is innovative, but further studies still need to be done before a definite conclusion can be formulated.

Previous research projects have analyzed the relationship between inspection results and the attributes of team members, including background, experience, and perspective. These showed more significant results than personality traits. We may need to compare the relative impacts of these factors on the results of inspection.

Furthermore, it may be useful to study not only the relationship between the diversity of personalities and the effectiveness of inspection, but also its relationship with the efficiency of inspection. For example, an inspection team is not simply a collection of individuals. Will a team with diversified personalities identify defects more quickly, or will a less diversified team communicate better, and work out the defects collectively in a more efficient manner?

Reviewer:  T.H. Tse Review #: CR131409 (0512-1327)

Reproduction in whole or in part without permission is prohibited.   Copyright 2024 ComputingReviews.com™
Terms of Use
| Privacy Policy