An Empirical Study of Slice-Based Cohesion and Coupling Metrics

Author(s): TIMOTHY M. MEYERS and DAVID BINKLEY
Venue: ACM Transactions on Software Engineering and Methodology
Date: 2007

Type of Experiement: Case Study
Sample Size: 63

Quality
5

This article provides empirical support for the usefulness of coupling and cohesion measurements, specifically "slice-based" metrics, when identifying where to re-engineer software and measuring re-engineering work. The study includes statistical analyses showing that slice-based cohesion metrics quantify software rot and demonstrate precisely where the most improvement effort is required.

Via a thorough analysis of the metrics in question and two real-world longitudinal studies, the paper convincingly demonstrates that the slice-based metrics do not simply reflect the age and size of a program, but are actually and quantitatively indicative of relevant information. The consideration of 63 subject programs and the high accuracy and precision of the metric being measured mean that the study is not very vulnerable to threats of validity.

0