Partial Least Squares Structural Equation Modeling (PLS-SEM) and the Necessary Condition Analysis (NCA) in IB Research
Masterclass Leads: Nicole Richter (University of Southern Denmark), Christian Ringle (Hamburg University of Technology)
Time and Date: 5 July, 2-5pm CET
This workshop introduces and encourages the combined use of partial least squares structural equation modeling (PLS-SEM) and the necessary condition analysis (NCA) that enables researchers to explore and validate hypotheses following a sufficiency logic, as well as hypotheses drawing on a necessity logic.
PLS-SEM belongs to a family of regression-based methods for estimating models with latent variables developed by the Swedish econometrician Herman Wold (1985). Since the 2000s, PLS-SEM has gained widespread popularity in a variety of disciplines among them (international) marketing and management research. The method estimates theoretically established causal-predictive relationships between latent variables (i.e. constructs measured by observed indicators). The results can empirically substantiate the determinants (X) that lead to an outcome (Y). Authors who interpret their PLS-SEM findings often use expressions such as “X increases Y” or “a higher X leads to a higher Y”. The interpretation, therewith, follows a sufficiency logic. Understanding relationships in terms of sufficiency logic is extremely relevant. Researchers, for instance, aim to understand the factors that lead to a stronger intention to use certain technology by applying different theories of technology acceptance; or they aim to understand the factors that contribute to a higher loyalty of their customers.
In contrast, the NCA is a relatively novel research methodology that has attracted much attention in the academic community in recent years. The NCA follows a necessity logic (“X is necessary for Y”) and can identify necessary conditions in data sets. A necessary condition is a critical factor for an outcome: if the necessary cause is not in place the outcome will not materialize. Hence, the necessary condition can be a bottleneck, critical factor, constraint, disqualifier, etc. The right level of a necessary condition must be put and kept in place to avoid guaranteed failure. By adding a different logic and data analysis approach, an NCA adds both rigor and relevance to theory, data analysis, and publications.
Against this background, with a combined use of PLS-SEM and NCA, we can determine the factors that produce the best possible outcome (i.e. the should-have factors; sufficiency logic) and those that are critical for an outcome (i.e. the must-have factors; necessity logic). Importantly, the should-have factors can only increase an outcome after the must-have factors have been taken care of. If necessary conditions are ignored or neglected in a field where we theoretically assume they exist, the result will be incomplete findings and recommendations. PLS-SEM is an approach to identify the determinants that can increase an outcome. NCA identifies the necessary level of a determinant that is needed to enable the outcome (Richter et al., 2020).
In this workshop, we will, therefore, introduce sufficiency and necessity logic as well as the foundations of a combined PLS-SEM and NCA use. For a case study illustration we use the SmartPLS 4 software. We provide insights into the logic, assessment, challenges and benefits of a combined use of PLS-SEM and NCA.