Abstract

In the statistical literature of the last 20 years, Pls-Path modeling has been a common methodology for the analysis of customer satisfaction and customer loyalty. Since Pls-Path models are systems of causal linear relations between latent variables, they are suitable for modeling constructs as satisfaction, loyalty, quality, etc., which cannot be measured directly and objectively. Nevertheless these models are not so common in the practice of marketing studies. This is due to two main reasons: i) this methodology requires specific, generally expensive, software packages; ii) the lack of a number of explanatory case studies, easy and accessible for the most of marketing analysts. However the first cause is less and less binding: a variety of Pls-Path models can be implemented by the free R-package «plspm» now. Therefore an easy, but interesting case study is presented regarding the satisfaction and loyalty of customers of a supermarket in Bursa, Turkey. The idea of the model is that the Overall Satisfaction of the supermarket customers depends on three main factors: i) quality of goods; ii) prices level; iii) convenience & pleasantness. Moreover, the three factors and the overall satisfaction are assumed to affect the customer loyalty. All these constructs are latent variables indirectly measured by manifest variables. Some interesting findings have emerged: i) the customer loyalty degree depends on the perceived quality of the goods, but not on the overall satisfaction degree; ii) the prices level affects the customer satisfaction but not the customer loyalty. These findings are consistent with the profiles of the customers and the supermarket as well.