VOLUME II. Fuzzy-set Qualitative Comparative Analysis (fsQCA) and its Relationship with Innovation: Discussion and Interpretation of Results

The present document is a continuation of the work "Crisp-set Qualitative Comparative Analysis (csQCA) and its relationship with Innovation: Discussion and Interpretation of Results." Its objective is to introduce researchers to the concepts and applications involved in Qualitative Comparative Analysis (QCA) with fuzzy-set data (fsQCA) through demonstrations of its use with various examples in innovation management.

Since the late 1980s, its precursor, Qualitative Comparative Analysis (QCA), has been at the forefront of social science methodology because it is based on a set of established relationships and objectives for discovering sufficient and necessary conditions.

It should be noted that the original Boolean version of QCA is commonly referred to as csQCA, where cs (crisp-sets) denotes a crisp data set. The version that allows the use of multiple-category conditions is known as mvQCA, where mv (multi-value) describes the multi-value, and for the version fsQCA, where fs (fuzzy-set) describes the fuzzy data set. The objective of QCA analysis, in general, is to account for a particular outcome, as opposed to regression-based analyses, which generally aim to be foundational tools for explaining the effects of causes (Wagemann and Schneider, 2010).

Regarding the state of social sciences as a general basis for innovation management, Sartori (1970) stated:

"...there is a deplorable state of science... oscillating between two flimsy extremes: unconscious thinking, which the overwhelming majority does, and overly conscious thinking, done by a small minority..."

Calling on scholars to acquire training in primary logic:

"...to guide a course between crude logical mismanagement on one side, and logical perfectionism (and paralysis) on the other..."

It was in the late 1980s when Dr. Charles Ragin brought Boolean algebra and set theory to the social sciences with his groundbreaking book "The Comparative Method" (Ragin, 1987), which describes in depth all components of QCA. Even so, the real stimulus in attention began a few years later, with the book "Fuzzy-sets Social Science" (Ragin, 2000). Now, there are already academics using configurational comparative methods, given the possibility of formalizing case-oriented analysis and thus providing tools to improve comparative research. These methods are particularly suitable for identifying the minimally necessary and/or minimally sufficient (combinations of) conditions that produce an outcome of interest (i.e., evaluating the causes of effects), with great potential for quantitative applications such as in engineering (Mendel and Korjani, 2010; Marks, et al., 2018).

Given the advantages offered, in recent discussions on configurational comparative methods, scholars argue that quantitative approaches based on regressions vs. QCA are best applied together (Ragin, 2008; Schneider and Wagemann 2010; Rihoux, 2006). However, there is a warning for enthusiasts that academics should not become monomaniacal about QCA (Ragin and Rihoux 2004, p. 6).

On the other hand, there are early efforts in the application of fsQCA in the field of entrepreneurship and innovation, such as the work of Kraus (et al., 2017), which compiles 77 articles published from 2005 to 2016 with the keywords: fsQCA, business management, entrepreneurship, and innovation, revealing a gradual increase in these fields for the use of fsQCA. Thus, innovation-oriented management sciences have the opportunity to leverage what has been achieved in social sciences through the significant contributions of Dr. Ragin. Therefore, the present work is composed of eleven chapters, which we briefly describe:

CHAPTER 4. Fuzzy-set Qualitative Comparative Analysis (fsQCA). This chapter introduces the reader to a very special type of data, the fuzzy type, using the fsQCA software. It reveals basic concepts of its use through the knowledge of its nature, the possibility of using it by levels and continuously. Operations of the fuzzy data set are presented, such as Boolean negation, conjunction (logical AND), and union (logical OR). It includes models and methods of calibration, analysis of necessary and sufficient conditions, and consistency, how to create truth tables and analyze the corners of vector space. Fourteen exercises demonstrate the use of the fsQCA software, confirming the use of the main concepts.

CHAPTER 5. Evaluation of the fsQCA data set. This chapter describes the importance of what is known as INUS and SUIN conditions, as well as consistency and coverage in general. The manual development of five cases confirms the use of the main concepts.

CHAPTER 6. Calibration and its importance in fsQCA. This chapter opens a debate on the implications of calibration before starting measurements in scope and context, the pros and cons of using indicators by quantitative research, the relation of SEM vs. fsQCA, qualitative research, and the need for calibration, direct and indirect methods of calibration. The manual development of four cases confirms the use of the main concepts.

CHAPTER 7. Configurational thinking in fsQCA. This chapter demonstrates the importance of configurational thinking from both qualitative and quantitative perspectives. It discusses the evaluation of the degree of membership in a configuration as well as the comparison of causal pathways.

CHAPTER 8. Net effects in fsQCA. The chapter describes what should be understood by net effects, the problems associated with it, proposes a shift in focus to types of cases, presents a comparison of fuzzy data cases, and the analysis of configurations.

CHAPTER 9. Net effects vs. Configurations in fsQCA. Through two hypothetical cases, the study is conducted, and each of the concepts is developed, explaining their results by comparison, in order to understand the nature of net effects versus configurational effects by fsQCA.

CHAPTER 10. Guide to csQCA analysis. With crisp data, from an innovation case, a quick and visual guide is offered to the reader through screenshots and 8 exercises, explaining how to solve a case by performing: analysis of necessary conditions, representing data in the truth table, identifying contradictory and logical remainders, sufficiency analysis, truth table minimization for occurrence and non-occurrence of the phenomenon, and interpretation of results.

CHAPTER 11. Guide to fsQCA analysis. With fuzzy data, from an innovation case, a quick and visual guide is offered to the reader through screenshots and 9 exercises, explaining how to solve a case by performing: access to the working file, analysis of necessary conditions with the occurrence of the phenomenon and its reports, interpretation of coverage, generation of the truth table, application of the Standard Analyses option, interpretation of generated solutions, generation of the truth table with non-occurrence of the phenomenon, analysis of main implications, and case interpretation.

GLOSSARY. A glossary of the most commonly used terms in the subject is presented.