Summary and personal reflection

Summary of the book

This book has introduced Necessary Condition Analysis (NCA) as a methodological approach that adds a distinct necessity perspective to the toolbox of empirical research and practical decision-making. NCA starts from a simple idea: some factors are not just helpful contributors to an outcome, they are indispensable. If a necessary condition is not present (at a required level), a desired outcome cannot occur, and no other factor can compensate for this absence. Conversely, for undesirable outcomes, certain stop factors can be kept below a threshold to make the outcome impossible.

Part I of the book set out the principles of NCA. It began in Chapter 2 with causality, clarifying how necessity logic (if not \(X\), then not \(Y\)) differs from sufficiency logic (if \(X\), then \(Y\)) and from configurational, probabilistic, or typicality-based perspectives. NCA extends classical binary necessity to a continuous view in which different levels of a condition enable different levels of an outcome, while still maintaining a clear zone in which cases are impossible, above a ceiling line in the \(XY\)-plot. The theory chapter (Chapter 3) then showed how necessity relationships can be formulated as parsimonious necessity theories, discussed directions of necessity, double necessity, and clarified what “necessary but not sufficient” means in a causal framework.

The mathematical foundations of NCA (Chapter 4) characterize necessity in geometric terms: bounded variables, ceiling lines, degrees of freedom, effect size, and several model fit measures such as ceiling accuracy, sharpness, and spread. Together, these measures describe how well the ceiling line separates the “impossible” empty corner from the feasible area, and how sharply the necessity pattern appears in the data.

The statistics chapter (Chapter 5) then explained how NCA uses permutation tests, simulations, and power analyses to test whether an observed necessity effect could have arisen by chance. It also showed how regression-based concepts such as mediation, moderation, omitted variable bias, and confounding are different or do not apply from the perspective of necessity logic and NCA.

The credibility chapter (Chapter 6) synthesized these foundations into a set of criteria for deciding whether a necessary condition claim is believable: theoretical support (a well-developed necessity hypothesis), practical relevance (a substantively meaningful effect size), statistical significance, robustness checks, and model fit. Together they provide a structured way to evaluate whether an observed ceiling pattern truly supports a necessity claim rather than reflecting a statistical artifact or a trivial result.

Part II translated these principles into concrete steps for conducting NCA studies. First, the hypothesis chapter (Chapter 7) described how to develop formal necessity hypotheses by drawing on explicit and tacit knowledge, building a preliminary necessity theory, performing thought experiments, and adding a causal explanation.

The data chapter (Chapter 8) discussed study designs (observational, longitudinal, case-based, experimental), sampling, case selection, and measurement for quantitative, qualitative, and set-membership data, as well as the use of archival data and data transformations.

The data analysis chapter (Chapter 9) presented NCA’s core analytic approach, which involves specifying the model bounding box and the ceiling line, visually inspecting the \(XY\)-plot, estimating the ceiling line, computing effect size and \(p\)-values, identifying and handling outliers, and assessing model fit. It then distinguished between necessity-in-kind (whether a condition is necessary at all) and necessity-in-degree (how much of a condition is required for a specific target outcome), operationalized via bottleneck tables. These bottleneck tables show which levels of conditions constrain particular target levels of the outcome, thus directly connecting NCA results to potential action. Robustness checks (varying ceiling functions, scopes, thresholds, outlier decisions, and target outcomes) help to assess the stability of necessity claims.

The reporting chapter (Chapter 10) explained how to write up NCA studies for different purposes (empirical, theoretical, methodological introduction, and practical publications) and introduced the SCoRe checklist for high-quality reporting, covering theory, methods, data, analysis, results, and discussion.

The multimethod chapter (Chapter 11) positioned NCA within a broader landscape of methods and explained how NCA can be combined with regression-based approaches (MLR, SEM, PLS) and with QCA. Rather than treating methods as competing, the book advocates a causal-pluralist view where necessity, sufficiency, and probabilistic perspectives complement one another: they are first analyzed separately and then interpreted collectively using tools like NERT, NEST and BIPMA. Each method answers a different causal question and combining them allows a richer understanding of the same phenomenon.

The final chapter (Chapter 12) translated NCA into practice. Here, necessary conditions become levers for change. NCA helps to identify must-have factors for desirable outcomes and stop factors for preventing risks. Its threshold logic and bottleneck areas make it possible to design interventions that target those cases that cannot reach a target outcome unless a specific condition is improved, or that will remain protected from an undesired outcome as long as a risk factor stays below its threshold. The chapter introduced concepts such as intervention effort, effectiveness, waste and efficiency, and illustrated how NCA-based interventions can be more targeted and cost-effective than interventions guided by average-effect logic.

Taken together, the book shows that NCA is not a replacement for existing methods, but a complementary approach that adds a missing causal lens: necessity. It invites analysts to move beyond the question “What works on average?” and to also ask “What is indispensable?” For researchers, NCA encourages the development of necessity theories and provides tools to test them rigorously. For practitioners, NCA offers a way to detect and remove bottlenecks in real systems, whether in education, business, healthcare, public policy, or any other field. With its conceptual framework, analytic tools, reporting standards, software support, and practical guidance, the book aims to enable thoughtful, high-quality use of NCA in both academic and applied work, and to stimulate further development of necessity-based thinking in the years to come.

Reflections on a decade of NCA

The publication of this book coincides with the tenth anniversary of Necessary Condition Analysis. The first formal exposition of NCA appeared in the journal Organizational Research Methods in 2016, but the idea itself required another decade of thinking, experimenting, and refining before it was ready to enter the scientific world. My own path toward NCA was far from linear. Trained first in mechanical engineering, then in biomedical engineering, I worked in applied research and consultancy in the field of human factors and ergonomics, and then moved into academia in business and management in the social sciences. I did not begin my career in the fields where NCA would ultimately take root.

I held managerial and administrative positions. These experiences gave me a deep appreciation for practical decision-making: understanding constraints, identifying what is truly essential, and distinguishing what merely helps from what must be present. With my strong connections to engineering and medical sciences, I was accustomed to environments where replication, empirical validation, and evidence-based practice were indispensable. Predictions had to be precise and actionable.

Entering the social science field late and as an outsider was, in some ways, a disadvantage. I lacked the traditional background and networks. But it was also an advantage: I came without disciplinary blinders, free to be curious, to ask naive questions, and to challenge established assumptions. When I moved into the social sciences, I found a landscape overwhelmed by theories and concepts, often complex, difficult to interpret, and hard to translate into action. I saw a surprising tolerance for drawing strong conclusions from single-shot studies. The lack of replication, limited availability of high-quality data, and the growing complexity of models left me wondering how decisions could be made with confidence.

At the same time, it became clear to me that this situation was not due to negligence. The social sciences are young compared to engineering and medicine, and the scale of funding available for research is simply not comparable. Despite these constraints, the social sciences have achieved great advancements. Real-life problems involve people, decisions, organizations, and institutions, and these problems cannot be understood and solved without the insights of the social sciences. Addressing real-world challenges requires the combined strengths of sciences working together.

In this context, I sensed that something essential (literally, something necessary) was missing. The deterministic logic of necessity, which can perfectly predict the absence of an outcome and is foundational in how real systems function, offered a way out. Although the idea emerged during my encounter with the social sciences, it applies equally in engineering and medicine. Indeed, it is now taking hold in those fields as well. I often wondered how it was possible that the necessity perspective of NCA had not been developed earlier.

Before 2016, I also took time to think carefully about dissemination. I reflected on how new ideas spread and where resistance may lie: among gatekeepers, reviewers, editors, and entrenched methodological traditions. I knew that for NCA to have a chance, it should not be perceived as a competitor to established methods but as a complementary lens that adds something genuinely new. I also knew that it would need openness, transparency, and accessibility. From the start, my principle was that NCA should be freely available: free software, free materials, free support, and freely shared ideas. Dissemination meant planting seeds among doctoral students, researchers, and practitioners, and letting those seeds grow where they found fertile ground.

The years following the first publication were a period of extremes. On one side were strong negative reactions, sometimes surprisingly intense. There were science trolls (yes, they exist) who attacked not only the method but also its intentions. There were editors who feared that using NCA was “naive”, reviewers who issued strong opinions without knowing the method, and peers who advised others not to engage with the method. Some speculated that “if applied in medicine, people will die.” Others insisted that NCA should first prove itself in the top-tier econometrics journal before it could be taken seriously. And there were cynical remarks about miraculous empty spaces or trivial data transformations. These moments were not easy, but in areas where NCA has been accepted, they are largely behind us now. As Schopenhauer observed, new ideas first encounter ridicule, then opposition, and finally acceptance as self-evident.

Yet there were equally strong and uplifting experiences. Brave editors such as James LeBreton were willing to give the method a fair chance. Scholars like Herman Aguinis immediately saw the potential of necessity thinking, and many colleagues contributed enthusiastically (as recognized in the Acknowledgments). Even some early critics who once dismissed the method later became among the strongest supporters once they understood what NCA actually offers. A few even speculated wildly that the invention merited a Nobel Prize or could be a panacea for all methodological challenges (it is not!).

Alongside these positive developments, I also encountered weaker applications: rushed implementations, superficial readings, and a kind of “science populism” when people present NCA with great certainty while missing its underlying logic, sometimes even giving incorrect answers. Others tried to monetize the method, which only strengthened my commitment to keep NCA free and openly available.

Even with this turbulence, the growth of NCA has been remarkable. More scholars have used the method than I ever imagined, and many have helped push it forward. I am proud of what has been achieved: the development of concepts, terminology, mathematical foundations, statistical procedures, and tools that previously did not exist. Much had to be invented from scratch like names for ceiling lines, new concepts like the bottleneck table, or new measures of fit, and many other components that now form the method’s core.

Why has NCA not yet been more widely adopted in practice, despite its potential? One reason is academic incentives: researchers are encouraged to move quickly to the next publication, leaving limited time for sustained engagement with practical application. Many who use NCA strive for both academic rigor and practical relevance, but the system rarely rewards that combination. Nevertheless, I remain optimistic. With growing experience, more applications in practice, and continued methodological development, NCA is becoming increasingly mature, and its practical value is gradually being recognized.

Looking forward, I see three important directions:

• Broader uptake in academia, including in fields such as the medical sciences, where NCA can help prevent undesired outcomes by identifying and eliminating indispensable single conditions, as well as in disciplines that still focus primarily on explaining why outcomes occur on average.

• Further deepening of the method. Once a method is established and accepted, the following decades allow for refinement, extension, and specialization.

• More real-world applications demonstrating how NCA can guide action, decision-making, and practical interventions. Practitioners who currently rely solely on average-effect logic can benefit from necessity logic as well.

At the same time, there are no shortcuts in NCA. Applying NCA requires methodological competence, respecting the fundamentals of NCA: understanding its deterministic necessity logic (with possible exceptions), its mathematical and statistical characteristics, and its assumptions, without slipping into probabilistic or conventional statistical thinking. Founders and early methodologists of other methodological approaches such as SEM, QCA, and grounded theory have warned in their own traditions that users should refrain from mechanical application, neglected assumptions, and overclaiming results. I have no illusion that NCA is immune to these tendencies, or that this book can fully prevent them. NCA, too, will be (and from the very beginning already has been) misinterpreted, routinized, overclaimed, and at times misused.

I therefore call upon NCA users to respect its principles: build strong theoretical support with explicit necessity hypotheses; avoid tweaking analyses to “find” necessity (since non-necessity can be equally informative); and report approaches, results and conclusions transparently. These principles are operationalized in the SCoRe checklist for analysts, authors, readers, peers, editors, reviewers and others to assess and enhance the quality of NCA applications.

Finally, on a personal note, with thankfulness, many people have joined me along the way, contributing to the development and growth of NCA. But only one person has witnessed every effort, doubt, frustration, and joy, and has supported me unconditionally throughout the entire journey. For that, Renata, I am deeply grateful.