Conducting Necessary Condition Analysis for Business and Management Students

Jan Dul

• Publisher: SAGE Publications Ltd
• Year: 2019
• ISBN:
  9781526481139
• Edition: 1
• Topic: Business
• Subtopic: Ricerca e sviluppo
• Index: Business

Part of SAGE?s Mastering Business Research Methods series, conceived and edited by Bill Lee, Mark N. K. Saunders and Vadake K. Narayanan and designed to support students by providing in-depth and practical guidance on using a chosen method of data collection or analysis. Necessary Condition Analysis (NCA) is an emerging method of data analysis, based on the idea that research factors can be necessary for an outcome: if the condition is not there, then the result will not occur. These necessary conditions are everywhere, and NCA is an intuitive and straightforward means of finding and testing data, either as a standalone tool or as a complement to other research methods. This book is an invaluable guide to using NCA effectively in business and management dissertations, and offers practical guidance and insight into how to successfully transcribe and analyse data using the NCA approach in research projects. Jan Dul is Professor of Technology and Human Factors at Rotterdam School of Management, Erasmus University, The Netherlands.

1 Introduction

About this book

Necessary conditions are everywhere. Travelling to Amsterdam is a necessary condition for seeing Rembrandt’s painting The Night Watch in person. If you want to drive a car you will need fuel, if you want to graduate you will need to write a dissertation or satisfy the conditions of your programme, and if you want to read this book you will need to open it. Opening the book is a necessary condition because it enables you to read it. This is a very strong condition because not opening the book guarantees that you will not read it. However, opening the book is not a sufficient condition for reading it. After opening it you may decide not to read it; other factors such as motivation and time may play a role as well. Thus, a necessary but not sufficient condition enables the presence of the outcome when present, guarantees the absence of the outcome when absent, but does not guarantee the presence of the outcome when present. Absence of the necessary condition is a bottleneck that perfectly predicts the absence of the outcome. If you do not open the book, you will not read it.

Thank you for opening the book! You may find the motivation and time to read it in its entirety such that you can familiarise yourself with Necessary Condition Analysis (NCA). NCA is a research approach and data analysis method that is based on the logic that factors can be necessary but not sufficient for an outcome to occur. Because necessary conditions are everywhere in real life, NCA can be used in any discipline and profession. In psychology it has been found that intelligence is necessary but not sufficient for creativity: if persons are not intelligent they are not creative, but if they are intelligent they may or may not be creative (Karwowski et al., 2016). In business it is discussed whether senior management commitment is necessary but not sufficient for successful organisational change: if there is no senior management commitment then change will not be successful, but if there is senior management commitment then change may or may not be successful (Knol et al., 2018). In medicine it has been found that meta-cognition – the capacity to identify and then integrate mental experiences – is necessary for motivation of people with schizophrenia for good functioning: if there is not enough meta-cognition then motivation is low, but if meta-cognition is high, motivation may be low or high (Luther et al., 2017).

Why is NCA valuable?

NCA is valuable for several reasons. First, the method is intuitive and straightforward. Any researcher with a basic knowledge of scientific research and research methodology can readily apply the method. Second, the method triggers a new way of theoretical thinking that is based on necessity logic. Therefore, a study with NCA can provide an interesting theoretical contribution. Third, because the necessary condition works in isolation from the rest of the causal structure (that is why it is necessary), the theoretical necessity model can be simple. Often, NCA researchers will employ a theoretical model with only one or a few potential necessity causes. Fourth, the method complements other methodologies that are not based on necessity logic, such as regression analysis. Fifth, the results of NCA can be immediately applied in practice. If a necessary condition is identified, that condition must be in place in virtually every single case to reach the outcome. If that condition is not in place, the outcome will not occur. Thus, it makes sense only to focus on this condition before focusing on other causes to influence the outcome. Practitioners use this necessity logic when designing, managing or controlling factors to influence an outcome. They are aware that in a complex (social) environment it is not possible to design, manage and control all factors, so they focus on crucial factors that must be present to avoid a guaranteed failure, in other words the necessary conditions. Researchers who attended a summer course on NCA have provided their thoughts on why NCA could be useful (Box 1.1).

Box 1.1 Opinions of NCA users about NCA

How to conduct NCA?

NCA can be conducted in four stages (Figure 1.1).

Figure 1.1 The four stages of conducting Necessary Condition Analysis

In Stage 1 the researcher formulates the necessary condition hypothesis: ‘X is necessary for Y’. It is possible to formulate more than one hypothesis. In Stage 2 the researcher collects the data that are needed to perform NCA. This stage includes the selecting or sampling of cases, and the measurement of X and Y for each case, resulting in a dataset. It is also possible to use an existing dataset. Stage 3 is the core of NCA where the dataset is analysed regarding necessity and a conclusion is drawn about the necessary condition hypothesis. In Stage 4 the researcher reports the results of the NCA study.

Where is NCA applied?

NCA can be valuable to researchers who want to conduct a new research project that is based on necessity logic or extend an existing research project to add novelty and offer additional insights. NCA can provide results that are academically rigorous and practically relevant. Such a project could be a PhD thesis, or a study for publication in a journal, or most meaningfully for this book, a Master’s dissertation. Since the availability of NCA’s core article in 2016 (Dul, 2016a) and the related free NCA software (Dul and Buijs, 2015), the method has been applied in many disciplines, including a variety of business and management fields. Examples include Strategy, Finance, Operations, Innovation, Information Management, Human Resource Management, Organisational Behaviour, Entrepreneurship, and Transportation (Box 1.2). Business and management students have also used NCA for their Master’s theses with success (Box 1.3).

Box 1.2 Examples of NCA applications in business and management

• Corporate Social Performance: a Necessary Condition Analysis (Van der Laan and Dul, 2016).
• Is Brokerage Necessary for Innovative Performance? A Necessary Condition Analysis (Breet et al., 2018).
• Firm Capabilities and Performance: a Necessary Condition Analysis (Tho, 2018).
• Success and Failure of Nascent Stock Markets (Albuquerque de Sousa et al., 2016).
• Implementing Lean Practices in Manufacturing SMEs: Testing critical success factors using Necessary Condition Analysis (Knol et al., 2018).
• A Methodological Demonstration of Set-theoretical Approach to Social Media Maturity Models using Necessary Condition Analysis (Lasrado et al., 2016).
• When are Contracts and Trust Necessary for Innovation in Buyer-supplier Relationships? A Necessary Condition Analysis (Van der Valk et al., 2016).
• Is High Performance Without High Performance Work Practices Possible? A Necessary Condition Analysis (Hauff et al., 2017).
• Is Creativity Without Intelligence Possible? A Necessary Condition Analysis (Karwowski et al., 2016).
• No Particular Action Needed? A Necessary Condition Analysis of gestation activities and firm emergence (Arenius et al., 2017).
• Determinants of Safe and Productive Truck Driving: Empirical evidence from long-haul cargo transport (De Vries et al., 2017).

Box 1.3 Examples of Master’s theses with NCA applications in business and management

• Necessary Conditions for Maintaining Physical Activity Interventions (Guiking, 2009).
• Critical success factors of firms that cooperate in innovation (Sarrucco, 2011).
• Necessary conditional hypotheses building and occupational safety in Dutch warehouses (Bakker, 2011).
• Critical success factors for IT project success (Verheul, 2013).
• Explaining employee satisfaction with the headquarter-subsidiary relationship (Van Dalen, 2014).
• Critical success factors of new product development in the medical industry (Meijer, 2014).
• Critical success factors for information system success within the empty container positioning process (Helwig, 2014).
• Customer orientation and business performance: a content analysis of Dutch SMEs’ websites (Van ’t Hul, 2015).
• The effect of capital structure and corporate governance on stock liquidity (Kuipers, 2016).
• Necessary conditions for new ventures’ positive performances (Ferrari, 2016).
• Software-based platform ecosystems: relationship between vertical openness and performance (Overschie, 2016).
• Testing the necessary conditions of technology acceptance by potential organisational users of a mandatory IT in the pre-implementation phase (Verhoeve, 2017).
• The necessary conditions for entrepreneurial behaviour by middle management (Smits, 2018).
• The role of organisational factors in the pursuit of exploratory innovation across business units: a Necessary Condition Analysis (Thieule, 2018).

NCA has been applied to many types of research questions, both in qualitative research and in quantitative research. In qualitative research usually a small number of cases are studied, normally less than 20. I call this a ‘small N study’, where N stands for the number of cases. Most of the studies until 2016 that are presented in Box 1.3 are examples of small N studies. In quantitative research a large number of cases are studied with usually more than 20 cases. This is called a ‘large N study’. All the studies in Box 1.2 are examples of large N studies. The study with the largest N is a study with 12,255 persons showing that intelligence is necessary for creativity (Karwowski et al., 2016). In both qualitative and quantitative research, the research question often deals with characteristics, efforts, or steps that can be managed, designed or controlled for reaching or preventing an outcome that is of interest. That outcome can be something desirable, e.g. performance, innovation, sustainability, financial results, change, creativity, well-being, or health. In the selected business and management studies of Box 1.2 and Box 1.3 most of the outcomes are business outcomes like financial performance, innovation performance or social performance. When the desirable outcome is formulated as ‘success’, the necessary condition is sometimes called a ‘critical success factor’, or ‘key success factor’, as several examples in Box 1.3 show. Hence, ‘critical’ means that the factor must be present for success and that there will be a guaranteed failure if the factor is absent. The outcome can also be something undesirable, e.g. stress, sickness, risk, disease, or failure. Thus, the absence of the necessary condition ensures the absence of the undesired outcome. Without the tubercle bacteria a person does not have tuberculosis.

Gary Goertz, one of the modern thinkers about necessity logic and research, states that ‘for any research area one can find important necessary condition hypotheses’ (Goertz and Starr, 2003: 65–66). I support this statement wholeheartedly for any social science area including business and management. My discussions with researchers and students in all kinds of research areas show that it is always possible to quickly formulate that X, which is something that can be influenced in practice, such as enough funding for a change project, is necessary for Y, which is some outcome that is of interest in practice, such as successful change. Most likely, necessity logic also applies to your topic of research, and you could formulate a research question that can be answered with NCA. In Appendix 1, I give recommendations for how research questions and necessary condition hypotheses can be formulated for any research topic.

A brief history of NCA

Although NCA is a new research methodology, its logic goes back to David Hume’s philosophy of science (1777), and even to Aristotle (350 BC). Necessity statements are common in any research discipline and practice. However, necessary condition analysis in terms of developing or testing necessity statements with data has been absent for a long time. The simple reason is that no such analysis method was available. Since Francis Galton’s (1886) discovery of correlation and regression, the research focus has been on regression analysis and its underlying additive average effect logic: predicting the outcome (on average) from one or several predictors. Such analyses, however, cannot assess the necessity of single predictors.

Recently, the methodological interest in necessity analysis was revived. In 1987, Charles Ragin introduced a methodology called Qualitative Comparative Analysis (QCA) that includes the analysis of binary necessity statements, in effect ‘presence/absence of X is necessary for presence/absence of Y’. However, in the years that followed QCA primarily focused on sufficiency analysis, i.e. identifying alternative combinations of conditions that are sufficient for the outcome. Currently, QCA’s necessity analysis is usually recommended to precede the sufficiency analysis, but QCA applications in business and management in particular regularly lack a necessity analysis. This can be observed, for example, in the Journal of Business Research that has published a large number of QCA studies in business and management. For a discussion on the differences between NCA and QCA see Dul (2016b) and Vis and Dul (2018).

Around the turn of the century, several researchers in political science stressed the importance of necessary conditions in their field of research (Dion, 1998; Braumoeller and Goertz, 2000; Ragin, 2000). In 2003, building on these developments, Gary Goertz and Harvey Starr published Necessary Conditions: Theory, Methodology, and Applications, in which they discussed and integrated a broad range of topics related to necessity logic (Goertz and Starr, 2003). Their book identifies potential directions for the methodological development of necessity analysis and its potential application in political science and sociology. Later, in 2008, Tony Hak and I integrated necessity analysis in our book Case Study Methodology in Business Research (Dul and Hak, 2008). We suggested using necessity logic and analysis in business and management research, and showed how it can be applied not only in small N case studies but also in large N studies. Afterwards we combined forces and developed necessity analysis beyond binary necessity analysis (Dul, et al. 2010; Goertz, Hak, and Dul, 2013). Our goal was to perform a necessity analysis not just with variables that have only two levels (e.g. absent/present), the necessary conditions in kind, but also with variables that have more than two levels. This allows us to make a more precise in degree necessity statement in...