Simulating families of studies to build confidence in defect hypotheses

  • Authors:

  • Forrest Shull;Daniela Cruzes;Victor Basili;Manoel Mendonça

  • Affiliations:

  • Fraunhofer Center-Maryland, 4321 Hartwick Road, Suite 500, College Park, MD 20740, USA;Department of Computer Science, University of Maryland, College Park, MD 20742, USA and Computer Networks Research Group (NUPERC), Salvador University (UNIFACS), Rua Ponciano de Oliveira, 126 Salv ...;Department of Computer Science, University of Maryland, College Park, MD 20742, USA;Computer Networks Research Group (NUPERC), Salvador University (UNIFACS), Rua Ponciano de Oliveira, 126 Salvador, BA 41950-275, Brazil

  • Venue:
  • Information and Software Technology
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

While it is clear that there are many sources of variation from one development context to another, it is not clear a priori what specific variables will influence the effectiveness of a process in a given context. For this reason, we argue that knowledge about software process must be built from families of studies, in which related studies are run within similar contexts as well as very different ones. Previous papers have discussed how to design related studies so as to document as precisely as possible the values of likely context variables and be able to compare with those observed in new studies. While such a planned approach is important, we argue that an opportunistic approach is also practical. The approach would combine results from multiple individual studies after the fact, enabling recommendations to be made about process effectiveness in context. In this paper, we describe two processes with which we have been working to build empirical knowledge about software development processes: one is a manual and informal approach, which relies on identifying common beliefs or 'folklore' to identify useful hypotheses and a manual analysis of the information in papers to investigate whether there is support for those hypotheses; the other is a formal approach based around encoding the information in papers into a structured hypothesis base that can then be searched to organize hypotheses and their associated support. We test these processes by applying them to build knowledge in the area of defect folklore (i.e. commonly accepted heuristics about software defects and their behavior). We show that the formal methodology can produce useful and feasible results, especially when it is compared to the results output from the more manual, expert-based approach. The formalized approach, by relying on a reusable hypothesis base, is repeatable and also capable of producing a more thorough basis of support for hypotheses, including results from papers or articles that may have been overlooked or not considered by the experts.