Modeling complex interactions of switching barriers - a latent profile approach

Existing research on drivers of customer loyalty such as satisfaction and switching barriers often does not account for possible nonlinear interaction effects (e.g., Jones et al. 2000). General and generalized linear models (e.g., linear regression, logistic regression, or structural equation modeling), however, typically do not fully account for the possible nonlinear interactive relationships that emerge when an entire profile of variables is considered in its holism. In consequence, the understanding of interaction effects is somewhat limited and subject to misinterpretations. This paper examines moderating effects of switching barriers on the link of satisfaction on loyalty intention. The methodical challenge is set by the need to capture possible nonlinear interactions among many moderating factors. Therefore, this article aims (a) to provide a demonstration with a particular focus on the use of finite mixture models to capture these interactive effects and (b) to compare these results to those of a traditional binomial logit regression.