A Pilot Case Study on Innovative Behaviour: Lessons Learned and Directions for Future Work

TL;DR

This paper emphasizes the importance of pilot case studies in empirical software engineering for refining research variables and case study design, demonstrated through an industrial study on innovative behaviour.

Contribution

It presents a detailed account of how a pilot case study can improve research design and variable identification in empirical software engineering research.

Findings

Identified new research variables influencing innovative behaviour.

Supported a more sophisticated case study design.

Highlighted the value of pilot studies in refining research approaches.

Abstract

Context: A case study is a powerful research strategy for investigating complex social-technical and managerial phenomena in real life settings. However, when the phenomenon has not been fully discovered or understood, pilot case studies are important to refine the research problem, the research variables, and the case study design before launching a full-scale investigation. The role of pilot case studies has not been fully addressed in empirical software engineering research literature. Objective: To explore the use of pilot case studies in the design of full-scale case studies, and to report the main lessons learned from an industrial pilot study. Method: We designed and conducted an exploratory case study to identify new relevant research variables that influence the innovative behaviour of software engineers in the industrial setting and to refine the full-scale case study design for the next phase of our research. Results: The use of a pilot case study identified several important research variables that were missing in the initial framework. The pilot study also supported a more sophisticated case study design, which was used to guide a full-scale study. Conclusions: When a research topic has not been fully discovered or understood, it is difficult to create a case study design that covers the relevant research variables and their potential relationships. Conducting a full-scale case study using an untested case design can lead to waste of resources and time if the design has to be reworked during the study. In these situations, the use of pilot case studies can significantly improve the case study design.