Six Sigma for Software Application of Hypothesis Tests to Software Data

  • Authors:
  • Krishna Arul;Harsh Kohli

  • Affiliations:
  • Global Software Group, Motorola Technology Center, Alba Campus, Livingston, Edinburgh EH54 7EG, UK krishna.arul@motorola.com;Global Software Group, Motorola Inc., 21440 West Lake Cook Road, Deer Park, IL 60010, USA h.kohli@motorola.com

  • Venue:
  • Software Quality Control
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

The article discusses statistical analysis on Release data from a mature Global Software Group (GSG) with three business domains with an aim to ascertain if any correlation can be established between the data collected after release and the benefits to the centre from the outcome of the analysis. Thus re-evaluating the process involved in creating and delivering software products to our customers. This article will strive to explain some of the Six Sigma methodologies and tools used in comparing groups (ANOVA method) and categorical data analysis techniques on software data. The paper will also discuss in detail normal distribution, tests for normality and comparative methods along with Contingency Table Analysis (Mosaic plots) and Correspondence Analysis. The application of statistics for SOFTWARE process improvement is increasing in the industry; we present how these techniques can be applied in practice.