Do SQA Programs Work - CMM Works. A Meta Analysis

  • Authors:
  • Daniel Galin;Moti Avrahami

  • Affiliations:
  • Ruppin Academic Center;Lipman Electronic Engineering Ltd.

  • Venue:
  • SWSTE '05 Proceedings of the IEEE International Conference on Software - Science, Technology & Engineering
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Many software development professionals and managers of software development organizations are not fully convinced in the profitability of investments for the advancement of SQA systems. The results included in each of the articles we found, , cannot lead to general conclusions on the impact of investments in upgrading an SQA system. Our meta analysis was based on CMM level transition (CMMLT) analysis of available publications and was for the seven most common performance metric. The CMMLT analysis is applicable for combined analysis of empirical data from many sources. Each record in our meta analysis database is calculated as "after-before ratio", which is nearly free of the studied organization's characteristics. Because the CMM guidelines and SQA requirement are similar, we claim that the results for CMM programs are also applicable to investments in SQA systems. The extensive database of over 1,800 projects from a variety of 19 information sources leading to the meta analysis results - proved that investments in CMM programs and similarly in SQA systems contribute to software development performance.