Using learning styles of software professionals to improve their inspection team performance

Inspections of software artifacts during early software development AIDS managers to detect early faults that may be hard to find and fix later. While inspections are effective, evidence suggests that inspection abilities of individuals vary widely which affect overall inspection effectiveness. Cognitive psychologists have used Learning Styles (LS) to measure an individual's characteristic strength and ability to acquire and process information. This concept of LS is being utilized in software engineering domain as a means to improve inspection performance. This paper presents the results from an industrial empirical study, wherein the LS's of individual inspectors were manipulated to measure its impact on the fault detection effectiveness of inspection teams. Using inspection data from nineteen professional developers, we developed virtual teams with varying LS's of individual inspectors and analyzed the team performance. The results from the current study show that, teams of inspectors with diverse LS's are significantly more effective at detecting faults as compared to teams of inspectors with similar LS's. Therefore, LS's can aid software managers to create high performance inspection team(s) and manage software quality.