A Dynamic Coupling for Reusable and Efficient Software System (2007)

Author(s): Misook Choi and Jong Suk Lee
Venue: International Conference on Software Engineering Research, Management and Applications
Date: 2007

Type of Experiement: Case Study


M. Choi and J. S. Lee, “A dynamic coupling for reusable and efficient software system,”
in Fifth Internation Conference on Software Engineering Research , Management and
Applications, pp. 720–726, 2007.

Applied proposed new dynamic coupling metric on an unnamed online ordering system.
They then compared the proposed metric to other conventional metrics such as: Coupling
Between Objects (CBO), Export Object Coupling (EOC), Import Object Coupling (IOC),
total number of dynamic messages. The new metric is then put into a table with comparisons
with results from the other metrics.

Choi and Lee propose a new dynamic metric that can measure the coupling accurately within
the object level considering structural relationships, method call types and the number of
method calls between classes. They develop a mathematical equation to illustrate how
they came about formulating their new dynamic metric. They separate each measurement
into its specific module and in the end result they use these modules in such a fashion to
compute their DCCOL (Dynamic Coupling between Classes in Object Level). They prove
the theoretical soundness of their metric by using the five axioms proposed by Briand et al.

They compare their metrics towards the CBO, EOC, IOC, and total number of dynamic
messages metrics. They show that only theirs and Arisholm’s total number of dynamic
messages can measure the dynamic coupling that exists in a special case. They then compare
their metric in a table versus the others. They analyze the data and conclude that their
measure measures a new and more accurate relationship between classes in object level.
However they do not show any correlation of their metric with any external quality attribute;
only that their metric is sound and can accurately measure coupling.