ModelGame: A Quality Model for Gamified Software Modeling Learning

Abstract

Gamification has been adopted in software development tasks in recent years. This adoption seeks, for example, to improve the engagement of developers while creating UML models or writing code. Empirical studies report that UML models suffer from incompleteness and inconsistency problems. This study conjectures that gamification mechanics can improve learner engagement while learning software modeling, mitigating such problems concerning UML models. The current literature lacks studies that explore gamification and UML model quality in the context of software modeling learning. This article, therefore, proposes ModelGame, which is a quality model to support software modeling learning in a gamified way. It serves as a reference framework so that instructors can obtain a parameterized way to evaluate UML models created by learners. The quality of UML models can be improved by applying gamified activities and providing guidelines aware of quality issues. A qualitative questionnaire was answered by 19 instructors who teach software modeling at higher education institutions. The results show that (1) 94.7% recognize that the proposed model can improve the quality of UML models, indicating that they would adopt the ModelGame in their learning practices; and (2) 47.4% do not use any gamification mechanics in their classes. The results are encouraging, showing the potential for applying and improving the teaching and learning of software modeling.