A Conceptual Model to Support Teaching of Software Engineering Controlled (Quasi-)Experiments

Throughout controlled experimentation, it is possible to provide evidence of the software being developed. In the academic environment, Experimentation in Software Engineering (ESE) is essential to understanding cause-effect relations, enabling a vision of the development process, and taking action on actual events in the software industry. As much as the experimentation processes have been used in industry and academia, there is a lack of formalization of the principles of ESE teaching and artifacts that can be useful to support it in higher education. One of the means to contribute to such a topic would be the design of a conceptual model, which is widely discussed in the literature, thus applying empirical methods for a better understanding of the context and representation of ESE teaching. Thus, in this paper, we developed a conceptual model to support the teaching of controlled experiments and quasi-experiments. To design the conceptual model, we carried out an analysis of metadata from controlled experiments and quasi-experiments in the literature and conducted a survey to collect data from instructors who teach ESE. We evaluated the model with the Technology Acceptance Model (TAM). Results consist of a feasible conceptual model aiming to standardize the basic concepts of ESE and further support the production and reuse of ESE materials.