Evaluation of work product defects during corrective & enhancive software evolution: a field study comparison

  • Authors:

  • David P. Hale;Joanne E. Hale;Randy K. Smith

  • Affiliations:

  • University of Alabama, Tuscaloosa, AL, USA;University of Alabama, Tuscaloosa, AL, USA;University of Alabama, Tuscaloosa, AL, USA

  • Venue:
  • ACM SIGMIS Database
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Information systems portfolio management assumes that software will evolve to maintain alignment with operational needs, a goal that must be met through effective ongoing maintenance. Thus, a primary goal of software maintainers is to ensure that production code is updated without the introduction of defects. However, there is a dearth of research that examines the work product defects that occur as these applications evolve. The goal of this study is to characterize software evolution lifecycle work product defects and factors that may increase or reduce their occurrence. The study takes place within a global consulting organization conducting ongoing software maintenance for a Fortune 100 telecommunications firm by a project team assessed at Capability Maturity Model Integration (CMMI) Level 3. This study reports on 991 work product reviews conducted across the evolution activities of the ISO/IEC 12207 Software Development Life Cycle Processes. After controlling for team and expertise differences, the study's major finding is that corrective evolution projects inject a greater number of work product defects than enhancive evolution projects. This result does not arise from the schedule compression often associated with corrective evolution. Rather, it is concluded that the increase in work product defects is associated with the increased complexity of analysis-stage problem diagnosis found in corrective evolution projects. The analysis is augmented by additional covariates including the number of work product reviewers, preparation time of reviewers, and size of the project.