The Shape of the Function Linking Expectancy-disconfirmation to Customer Satisfaction:Predictions of Prospect Theory

Zou Deqiang,Zhao Ping

DOI: https://doi.org/10.3969/j.issn.1008-3448.2008.06.013

2008-01-01

Abstract:Predictions of prospect theory on(1) reference depen-dence;(2) diminishing sensitivity;and(3) loss aversion of value functions are increasingly being applied in the field of consumer perception and judgment.Given the intrinsic association between reference dependence and the expectancy-disconfirmation para-digm,which is the dominant perspective in understanding customer satisfaction,the features of diminishing sensitivity(non-linearity) and loss aversion(asymmetry) are drawn upon in certain empirical studies exploring the complex curvilinear nature of the S-shaped function linking expectancy-disconfirmation to customer satisfac-tion.However,most of them improperly equate negative(positive) performance to negative(positive) disconfirmation,and thus fail to strictly examine the effect of reference dependence.In addition,the diminishing sensitivity and loss aversion characteristics of prospect theory are seldom verified simultaneously.To fill these voids,a cubic polynomial function is proposed which incorporates both the non-linearity and asymmetry of the relationship.Expectancy-dis-confirmation is explicitly operationalized as the difference between customers' ratings on their performance expectation and perceived performance.Using nationwide survey data in four different con-texts,i.e.,cell phones(n= 2518),laptop computers(n= 1932),desk-top computers(n= 1954),and TV sets(n= 2204),the above men-tioned predictions are tested empirically.Specifying the relationship between expectancy-disconfirmation and customer satisfaction with the cubic polynomial function proposed provides the predictions of prospect theory with empirical support.This study clarifies the subtle but significant distinction between negative(positive) perfor-mance and negative(positive) disconfirmation,and thus solidifies the underlying principles justifying the predictive validity of pros-pect theory in the customer satisfaction domain.Methodologically,the cubic polynomial function incorporates the asymmetric nature of an S-shaped curve effectively,with improved credibility and less constraints on parameter estimation in comparison to alternative specifications.Managerially,findings in this study shed some light on the efficient allocation of corporate resources in promoting cus-tomer relationships.