Information

What is a necessary condition?

A necessary condition is a critical determinant of an outcome: if the condition is not in place the outcome will not occur. A student will not be admitted to a PhD program when the GMAT score is too low, creativity will not exist without intelligence, and organizational change will not occur without management commitment. Such single conditions can be a bottleneck for the outcome. If the necessary condition is not in place there is guaranteed failure, and this cannot be compensated by other determinants. But when the condition is in place there is no guaranteed success. The condition is necessary but not sufficient. To prevent failure, each single necessary condition must be in place. Necessary Condition Analysis (NCA) provides the logic and a methodological tool for identifying necessary conditions in different datasets.

Why is NCA important?

Necessary (but not sufficient) conditions are widespread in real life and therefore relevant to various research areas, such as management, business research, sociology, and psychology. But until recently no technique was available to identify necessary conditions in datasets. Traditional (regression based) data analysis techniques fail to do so, even though they are frequently applied for exactly that matter. Necessary Condition Analysis (NCA) is a new technique that can do the job (Dul 2016).

Whom is NCA for?

NCA is applicable to any discipline, and can provide strong results even when other analyses such as regression analysis show no or weak effects. By adding a different logic and data analysis approach, NCA adds both rigor and relevance to theory, data analysis, and publications. NCA is a straightforward and user-friendly method that requires no advanced statistical or methodological knowledge beforehand. It can be used in both quantitative research as well as in qualitative research.

Where has NCA been used?

NCA is a rapidly emerging empirical research method that is being used in a variety in fields such as (industrial) psychology, international business, operations management, information sciences, finance, strategy, entrepreneurship, organizational behavior, and human resource management. For example, Karwowski et al. (2016) found that intelligence is necessary for creativity. Van der Valk et al. (2016) identified trust and contracts are necessary for successful collaboration between companies for innovation. Arenius, Engel, and Klyver (2017) tested whether particular gestation activities for establishing a new firm are necessary for profit two years after the firm’s start. Furthermore, de Vries et al. (2017) found that safety consciousness is necessary for driving performance, and Knol et al. (2018) identified several necessary conditions for successful implementation of lean production.  NCA has also been applied outside the social sciences, for example in medicine (Luther et al. 2017).

In general NCA is particularly useful when several factors contribute to an outcome (multi-causality), and it is not known which factors are essential for that outcome. Research questions may deal with characteristics, efforts and steps of people, teams, and organizations that are crucial for successful outcomes (e.g., high performance).

What are the main advantages of applying NCA:

• It provides new substantive insights in social science phenomena (expressed as necessity rather than average trend); using NCA provides an alternative perspective, may enhance existing research, or falsify theories;
• It has great practical meaning because identified necessary condition must always be put and kept in place
• Journal editors and reviewers appreciate that authors use a new solid methodology that provides new insights and contributes to academic rigor and practical relevance.

What editors said about NCA?

• “From my perspective, it is the most interesting paper I have handled at this journal, insofar as it really represents a new way to think about data analyses"
• “I just added the paper to the required reading list for my doctoral-level seminar on research methods”
• “I believe the paper holds the potential to be widely cited and to change how organizational researchers approach testing for cause and effect relationships”

What participants of this course said about NCA?

• “Simple technique that requires no ‘preparation’, data transformation/manipulation/ correction. A perfect plug-and-play method that can give output in under 10 minutes”
• “It’s a new way of thinking and therefore it may lead to many interesting insights, just reanalyzing old dataset”
• “Analysis provides insights that cannot be obtained with another method”
• “Insights are very relevant for practice”
• “I do believe that is has great exploratory value that is congruent with recent emphasis on big data”

Who can attend the course?

The course is primarily open to PhD candidates and junior faculty, but also other researchers (including master students and senior faculty) who are interested in this novel approach are welcome. It is expected that each participant has at least some experience with traditional regression analysis to understand the differences between NCA and regression, and to appreciate how NCA complements regression. Also researchers with a background in Qualitative Comparative Analysis (QCA) can benefit from the course to appreciate how NCA complements QCA. No special methodological or substantive knowledge is required for this course. Examples will be drawn from many different (business) fields. Each participant is expected to bring a laptop computer with a Windows (Microsoft) or IOS (Apple) operating system. The exercises with the NCA software (in R) can be done without prior programming knowledge or experience.

What is the content of the course?

The course consists of three parts:

1. Individual preparations to be done before the course (approximately 16 hours)

• Participant reads literature about necessary conditions (reader)
• Participant installs the NCA software on own laptop (detailed instructions will be provided)

2. Morning Class (approximately 4 hours)

NCA as logic and data analysis approach: How it works?

• Format: lecture with class discussions

3. Afternoon Class (approximately 4 hours)

Applying NCA to own dataset

• Format: Participants works on own laptop with assistance by lecturer, and class discussions
• Content: ceiling line, effect size, accuracy, bottleneck table, interpretation of results, handling of problematic or unusual cases, how to publish findings