Modeling Development Effort in Object-Oriented Systems Using Design Properties

  • Authors:

  • Lionel C. Briand;Jürgen Wüst

  • Affiliations:

  • -;-

  • Venue:
  • IEEE Transactions on Software Engineering - Special section on the seventh international software metrics symposium
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

In the context of software cost estimation, system size is widely taken as a main driver of system development effort. But, other structural design properties, such as coupling, cohesion, and complexity, have been suggested as additional cost factors. In this paper, using effort data from an object-oriented development project, we empirically investigate the relationship between class size and the development effort for a class and what additional impact structural properties such as class coupling have on effort. This paper proposes a practical, repeatable, and accurate analysis procedure to investigate relationships between structural properties and development effort. This is particularly important as it is necessary, as for any empirical study, to be able to replicate the analysis reported here. More specifically, we use Poisson regression and regression trees to build cost prediction models from size and design measures and use these models to predict system development effort. We also investigate a recently suggested technique to combine regression trees with regression analysis which aims at building more accurate models. Results indicate that fairly accurate predictions of class effort can be made based on simple measures of the class interface size alone (mean MREs below 30 percent). Effort predictions at the system level are even more accurate as, using Bootstrapping, the estimated 95 percent confidence interval for MREs is 3 to 23 percent. But, more sophisticated coupling and cohesion measures do not help to improve these predictions to a degree that would be practically significant. However, the use of hybrid models combining Poisson regression and CART regression trees clearly improves the accuracy of the models as compared to using Poisson regression alone.