Use of Scrum in a Virtual Environment to Enhance Collaboration and Systemic Reasoning of Engineering Students

Abstract Engineering students are often requested to work collaboratively in the solution of complex problems, however, accurately measuring the performance of individual team members can be difficult due to the conjunction of contributions into a single deliverable. This paper shows the impact of applying the Scrum methodology to manage and evaluate collaborative projects. Scrum is a widely used methodology for agile projects, and has been shown that it allows continuous creation of value to stakeholders. Moreover, Scrum application in academic projects has demonstrated that students can achieve results faster, and that their communication skills are improved. Nevertheless, literature has shown that the environment in which Scrum is implemented greatly affects the team outcomes, requiring then a physical infrastructure that allows proper communication and encourages teamwork. In contrast to current literature, this paper explores use of Scrum in a virtual collaborative environment, using multimedia platforms to enhance asynchronous collaboration between students, and allowing the teacher real-time review of individual students' contributions and performance. In order to collect data, thirty students were divided in random teams of 5 people and assigned roles according to Scrum methodology. A real-life inspired problem was then assigned to the teams, providing them the context of the problem, the main objective of the project, and a collection of User Stories to be solved during a fixed time interval of 5 weeks. Sprints were constrained to a one-week period, allowing the teacher to continuously observe the status of the project, providing feedback to individuals, and focusing on the Sprint Retrospective ceremony. Within this approach, the study is focused in two key measurements: students' perception of teamwork performance, and development of systemic reasoning skill. The first measurement is approached by a pre-post study design based in a Likert scale questionnaire. Results show improvement in the self-perceived performance of the students and in the quality of collaboration of their peers. The second measure is approached by the observation of three specific behaviors of the student among their contributions: identification of key variables, comprehension of variables interactions, and comprehension of problem limitations.