Title: Prediction of testability using the design metrics for object-oriented software

Authors: Yogesh Singh; Anju Saha

Addresses: University School of Information Technology, GGSIP University, Dwarka, Delhi, India ' University School of Information Technology, GGSIP University, Dwarka, Delhi, India

Abstract: One of the cost effective methods to monitor the testing effort is to assess the testability of the software from the early phases of the software development. Software metrics based testability assessment can enable testers to focus more on the testing of the less testable classes. This paper presents a study based on the experimental analysis that uses fourteen design metrics as the independent variables and two JUnit based test metrics as the dependent variables. The results of the study indicate a number of promising effects of design metrics on testability of a class in object oriented software.

Keywords: design metrics; object-oriented software; software testability; linear regression; least median squares; regression trees; empirical study; software development; software metrics.

DOI: 10.1504/IJCAT.2012.048204

International Journal of Computer Applications in Technology, 2012 Vol.44 No.1, pp.12 - 22

Published online: 26 Jul 2012 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article