Dynamic coupling metrics for object oriented software systems: a survey

Coupling measures are considered important due to their inherent relationships with external software quality attributes. The ongoing research on object oriented software quality has been focusing on devising dynamic metrics for software quality assessment because of their ability to capture the runtime behavior of software. A number of dynamic coupling metrics have been proposed in recent times to measure software quality. This paper is an attempt to fill in the gap between existing work and researchers by providing them essential knowledge about the dynamic coupling metrics, methodologies and techniques proposed in this area. This paper presents a survey of existing research related to dynamic coupling metrics for object oriented software systems. Metrics have been categorized according to some of their major characteristics such as dynamic analysis techniques, validation of metrics, relation with static metrics etc. The capability of these metrics in quantifying external software quality attributes is also studied. The survey found that very few dynamic coupling metrics have been proposed and most of the proposed metrics lack empirical validation limiting their applicability in software quality assessment.