Association and Aggregation Class Relationships: is there a Difference in Terms of Implementation?

Abstract

In this work we revisit the known problem of the lack of well-defined semantics, at the implementation level, for association and aggregation relationships in class diagrams to hypothesize that, in the context of software development, there is a better understanding of the associations between classes when omitting aggregation. We conducted an experimental study in which a questionnaire was applied to 100 subjects, 50 in the test group and 50 in the control group. The obtained responses were analyzed with statistical methods. From our results we conclude that it is not useful, from the point of view of software design, to differentiate between association and aggregation class relationships when a model will be implemented.