Empirical Studies on UML Sequence Diagrams in Education: A Systematic Literature Review

Abstract

The Unified Modelling Language (UML) is a standard notation for object-oriented modelling and plays a central role in the design of software systems. Both static and dynamic UML models are essential for representing complex system behaviour; however, prior studies indicate that novice learners often struggle to understand the relationships between these models. Students frequently construct sequence diagrams (SDs) that are inconsistent with corresponding class diagrams. Although formative feedback and tool support are well established in programming education, comparable support for UML modelling remains limited. This paper presents a systematic literature review of empirical studies on UML SDs in computing education, with a focus on notation, consistency, completeness, and model quality. The review identified 26 relevant studies from five major electronic databases. These studies were classified into three main categories: comprehension, model quality, and methods and tools. The results show that research on comprehension represents the primary focus of literature, followed by studies addressing model quality, while relatively few studies focus on methods and tools. The review highlights gaps, including limited support for novices and a lack of immediate formative feedback, pointing to opportunities for tools that provide ongoing diagnoses and guidance to help novices create valid SDs, while supporting consistency, completeness, and qualitative aspects.