Context: Grounded Theory (GT) is a research method that facilitates theory development. Yet, its application in a field with a prevalent positivist view of science, like Software Engineering (SE), often raises concerns among researchers. Objectives: This paper aims to highlight challenges in applying GT in SE and provide practical insights to overcome them. Method: I outline the top five challenges I faced while applying GT to develop a theory of value for value-based feature selection. The theory was developed taking a positivist stand. I also reflect on the lessons that I learned along the way. Results: The top five challenges are: 1) defining research questions, 2) keeping away from the related literature, 3) assuring the trustworthiness of “ground” data, 4) specifying the theory with a proper granularity level, and 5) reporting a GT study in a limited-length research paper. These challenges led to 17 lessons learned. Conclusions: While my experience with GT in SE has been positive, certain aspects of the method need thoughtful consideration, particularly when the research is conducted from a positivist stand. I hope that the experience I share in this paper is valuable for others in the SE community attempting to use GT for their studies.

Pilar Rodríguez is an Associate Professor at the Department of Computer Languages and Systems and Software Engineering, Universidad Politécnica de Madrid (Spain).