TY - CHAP
T1 - Response-Based Segmentation in PLS Path Modeling
T2 - Application Of Fimix-Pls to American Customer Satisfaction Index Data
AU - Rigdon, Edward E.
AU - Gudergan, Siegfried P.
AU - Ringle, Christian M.
AU - Sarstedt, Marko
N1 - Publisher Copyright:
© 2015, Academy of Marketing Science.
PY - 2015
Y1 - 2015
N2 - The role of customer satisfaction indices is important in understanding business success and is evidenced through established effects on customer retention and profitability. Today, Fornell et al.’s (1996) American Customer Satisfaction Index (ACSI) ranks among the most salient models in marketing. Several studies have provided strong evidence for the measure’s reliability and validity and provide a basis for comparing the effects of antecedent constructs on overall satisfaction and loyalty. However, by using aggregated data for their validation analyses, these studies assume that the data stem from a single homogenous population—a single model represents all observations. This assumption of homogeneity is unrealistic as customers are likely to differ in their perceptions and evaluations of firms’ characteristics, as well as in their familiarity with a given firm’s offerings (Rigdon et al. 2010). Moreover, observable characteristics are often inadequate in capturing the apparent heterogeneity in the data (Sarstedt and Ringle 2010). Even though prior research has found substantial unobserved customer heterogeneity within a given product or service class (Wu and DeSarbo 2005), it usually remains unaddressed which leads to biased parameter estimates and, consequently, flawed conclusions.
AB - The role of customer satisfaction indices is important in understanding business success and is evidenced through established effects on customer retention and profitability. Today, Fornell et al.’s (1996) American Customer Satisfaction Index (ACSI) ranks among the most salient models in marketing. Several studies have provided strong evidence for the measure’s reliability and validity and provide a basis for comparing the effects of antecedent constructs on overall satisfaction and loyalty. However, by using aggregated data for their validation analyses, these studies assume that the data stem from a single homogenous population—a single model represents all observations. This assumption of homogeneity is unrealistic as customers are likely to differ in their perceptions and evaluations of firms’ characteristics, as well as in their familiarity with a given firm’s offerings (Rigdon et al. 2010). Moreover, observable characteristics are often inadequate in capturing the apparent heterogeneity in the data (Sarstedt and Ringle 2010). Even though prior research has found substantial unobserved customer heterogeneity within a given product or service class (Wu and DeSarbo 2005), it usually remains unaddressed which leads to biased parameter estimates and, consequently, flawed conclusions.
KW - Customer Retention
KW - Finite Mixture
KW - Loyal Customer
KW - Salient Model
KW - Service Class
UR - http://www.scopus.com/inward/record.url?scp=85145080480&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-11797-3_86
DO - 10.1007/978-3-319-11797-3_86
M3 - Chapter
AN - SCOPUS:85145080480
T3 - Developments in Marketing Science: Proceedings of the Academy of Marketing Science
SP - 145
BT - Developments in Marketing Science
PB - Springer
ER -