5.1 Customer satisfaction index (CSI) models

Since 1970s, researchers of consumer behavior and marketing in developed countries have begun to make comprehensive studies on customer satisfaction. In 1989, Fornell and his colleagues in Michigan University helped Sweden build the first nation-level measurement system of customer satisfaction - Swedish Customer Satisfaction Barometer (SCSB) (Fornell, 1992). The SCSB became a uniform, cross company, cross-industry national measurement instrument of customer satisfaction and evaluations of quality of products and services. The basic model for estimating these indices is a structural equation model which links customer satisfaction to its determinants namely perceived quality, customer expectations and perceived value, and, in turn, to its consequences, namely customer loyalty and customer complaints.

Later in 1994, American Customer Satisfaction Index (ACSI) was launched (Fornell, 1996). In ACSI model customer expectation, perception of quality, perceived value were introduced as the antecedents of customer satisfaction and customer loyalty and customer complaint as consequences.

In the middle of 1990s, CSI was gradually recognized by governments and companies worldwide as a good instrument to gauge a nation's or company's output quality. Till now, nation-level CSIs have Swedish Customer Satisfaction Barometer (SCSB), American Customer Satisfaction Index (ACSI), European Customer Satisfaction Index (ECSI), Norwegian Customer Satisfaction Barometer (NCSB), German Barometer, Swiss Index of Customer satisfaction (SWICS), Korean Customer Satisfaction Index (KCSI), Malaysian Customer Satisfaction Index (MCSI). In addition, Brazil, Argentina, Mexico, Canada, Australia, Hong Kong and some regions like Taiwan, are striving to build their own CSI systems.

Although these CSIs are fundamentally similar in measurement model (i.e. causal model), they have some obvious distinctions in model's structure and variable's selection so that their results cannot be compared with each other. On the other hand, for nations who are attempting to construct their own CSIs, only to take full advantages of other nations' CSI experiences can they establish the CSIs which are suited for their nation's characters. Therefore, a comparison and analysis of the differences among these existed CSIs seems to be indispensable and valuable.

Four influential national customer satisfaction index models are evaluated from three perspectives;

1. The latent and manifest variables in CSI models
   • A manifest variable (or indicator) is an observable variable - i.e. a variable that can be measured directly. A manifest variable can be continuous or categorical.
   • A latent variable describes an unobservable construct and cannot be observed or measured directly. Latent variables are essential elements of latent variable models. A latent variable can be categorical or continuous.
2. The exogenous and endogenous variables in CSI models
   • Exogenous variables in causal modeling are the variables with no causal links (arrows) leading to them from other variables in the model. In other words, exogenous variables have no explicit causes within the model. The concept of exogenous variable is fundamental in path analysis and structural equation modeling. The relationships among latent variables.
   • Endogenous variables in causal modeling are the variables with causal links (arrows) leading to them from other variables in the model. In other words, endogenous variables have explicit causes within the model.
3. The relationships among variables.

The selected national CSI models are known as the most sophisticated and influential CSI models;

1. Swedish Customer Loyalty Barometer (SCSB) 1989
2. American Customer Loyalty Index (ASCI) 1994
3. Norwegian Customer Loyalty Barometer (NCSB) 1996
4. European Customer Loyalty Index (ECSI) 2000