Theory building for empirical software engineering in qualitative research: Operationalization

Jorge Pérez,Jessica Díaz,Ángel González-Prieto,Sergio Gil-Borrás

2024-12-03

Abstract:Context: This work is part of a research project whose ultimate goal is to systematize theory building in qualitative research in the field of software engineering. The proposed methodology involves four phases: conceptualization, operationalization, testing, and application. In previous work, we performed the conceptualization of a theory that investigates the structure of IT departments and teams when software-intensive organizations adopt a culture called DevOps. Objective: This paper presents a set of procedures to systematize the operationalization phase in theory building and their application in the context of DevOps team structures. Method: We operationalize the concepts and propositions that make up our theory to generate constructs and empirically testable hypotheses. Instead of using causal relations to operationalize the propositions, we adopt logical implication, which avoids the problems associated with causal reasoning. Strategies are proposed to ensure that the resulting theory aligns with the criterion of parsimony. Results: The operationalization phase is described from three perspectives: specification, implementation, and practical application. First, the operationalization process is formally defined. Second, a set of procedures for operating both concepts and propositions is described. Finally, the usefulness of the proposed procedures is demonstrated in a case study. Conclusions: This paper is a pioneering contribution in offering comprehensive guidelines for theory operationalization using logical implication. By following established procedures and using concrete examples, researchers can better ensure the success of their theory-building efforts through careful operationalization.

Software Engineering,Logic in Computer Science

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the operationalization stage of theory building in systematic qualitative research, especially in the field of software engineering. Specifically, the goal of the paper is to provide a set of procedures to systematize the operationalization stage, ensuring that theories can be transformed into testable hypotheses and that these hypotheses can be verified by empirical data. The following is a more detailed interpretation:

### Background of the Paper and Problem Description

1. **Four Stages of Theory Building**
- **Conceptualization**: Relevant concepts and their relationships are formed through inductive and abductive processes.
- **Operationalization**: Concepts are transformed into measurable constructs, and propositions are transformed into testable hypotheses.
- **Testing**: The validity of the theory is examined through empirical research.
- **Application**: The theory is applied and understood in practical situations.

2. **Focus of Current Research**
- This paper focuses on the operationalization stage, that is, how to transform the concepts and propositions in the theory into measurable constructs and testable hypotheses.
- Key terms involved in the operationalization process include: concept, empirical indicator, construct, proposition, and hypothesis.

3. **Existing Problems**
- **Disconnection between Concept and Operationalization**: Many studies overlook the operationalization stage, making it difficult to apply theories in practice.
- **Complexity and Methodological Tradition**: The operationalization process itself is complex, and some research paradigms may not attach importance to this stage.
- **Lack of Guidance**: Especially in disciplines such as software engineering, the qualitative research tradition is still developing and lacks clear operationalization guidelines.

### Solutions and Methods

1. **Logical Implication Instead of Causal Relationship**
- The author proposes using logical implication instead of causal relationship for operationalization. This is because causal relationships are difficult to fully prove in empirical research, especially in open systems.

2. **Parsimony Principle**
- In order to reduce the number of generated hypotheses, the author suggests choosing strategic propositions, that is, those propositions that describe the change of system state, to avoid unnecessary redundancy.

3. **Specific Operational Steps**
- **Define Constructs**: Transform abstract concepts into specific variables and indicators.
- **Generate Hypotheses**: Transform propositions into testable hypotheses.
- **Simplify Hypotheses**: Reduce the number of hypotheses through the parsimony principle to improve the simplicity and operability of the theory.

### Results and Contributions

- This paper proposes a comprehensive operationalization guide to help researchers better transform theories into testable hypotheses.
- Through specific case studies, it shows the practical application of these operationalization steps, ensuring the success of theory building.

### Summary

The main contribution of this paper lies in providing a set of systematic operationalization methods, especially in qualitative research, to help researchers transform abstract theories into measurable constructs and testable hypotheses, thereby enhancing the practicality and reliability of theories. Through this method, researchers can better ensure the success of their theory - building efforts.