Abstract:

Working in a team for a limited period of time and to deliver a high quality product are some of the skills that students should gain during their studies at the university. Empirical studies in software engineering play an important and significant role in the evaluation of tools, techniques, methods, and technologies before they are dynamically validated in industrial setup. After the presentation of the projects, students have a meeting with teaching staff to discuss and analyze the strengths and weaknesses of their work. The analysis should be performed by keeping in view the documentation, the project plan and the work experience the team has gathered from the start to the end of the project completion. To learn from the experience is a key to improvement. To perform empirical studies in real contexts is very difficult due to various obstacles, so, we intend to create a stable environment that allows us to perform reliable empirical studies with students. In many studies, students are used instead of professional software developers, although the objective is to draw conclusions valid for professional software developers. The differences may be considered only minor, and it is concluded that software engineering students may be used instead of professional software developers under certain conditions.
This paper presents a framework to classify empirical studies performed in an educational context. We will describe the relevance and objectives of the framework. Then, we present the schema (architecture) of the framework and explain in some detail its different components. The main objective of the framework is to provide a classification scheme for understanding and evaluating empirical studies in software engineering area in educational context. With this framework we can analyze most of the experimental work that has been performed in our software engineering courses. We involve students from BSc, MSc and PhD degrees in Computing from our university in developing a software project required by real clients. The educational approach is mainly based on (PBL) Project-Based Learning principles. Over the past four years we had five hundred and twenty students involved in the four educational syllabuses. In each academic year, we had approximately 100 students in (PMS) Software Process and Methodologies and (DAI) Development of Software Applications courses and 30 students in (ACSI) Analysis and Design of Information Systems and (GPSI) Project Management of Information Systems courses. Normally, the PMS/DAI and ACSI/GPSI courses has the same students. From the academic year 2010/2011 to 2017/2018, the 200 students enrolled in the ACSI and GPSI courses developed 150 work projects in different SWEBOK (Software Engineering Body of Knowledge) KAs (Knowledge Areas). We classify these works projects to explain the usefulness and applicability of the using the proposed framework.

Keywords:

Empirical studies, experimental design, Project-Based Learning, software engineering management, software engineering process.