Application of the Cognitive Dissonance Theory to the Service Industry

Abstract

While numerous studies used the theory of cognitive dissonance to explain consumer behavior, there is no study as of today that has applied the theory to the service industry to investigate the role of cognitive dissonance in influencing important marketing-related variables such as perceived value and trust. This study examines the relationship between cognitive dissonance and relationship variables (trust and value) and proposes a model to understand how consumers process information facing a negative word-of-mouth (WOM) message that is incongruent with their existing belief, along with how cognitive dissonance affects their behaviors. The proposed model is tested using data collected via mail survey on customers of hotel–motel organizations. The study finding suggests that the theory of cognitive dissonance can be extended beyond a postdecision situation to a service context to explain how customers process information facing a WOM message that is incongruent with their existing belief. It is shown that cognitive dissonance is a predominant predictor of repatronage behavior, even for satisfied customers.

KEYWORDS:

• cognitive dissonance
• repurchase behavior
• service
• trust
• value

Notes

*p ≤ .05. **p ≤ .01.

Note. Two-tailed tests. Standardized coefficients are reported. Purpose of trip was measured as a nominal variable with 0 coded for business travel and 1 coded for leisure travel. Exp. var. = explained variance shows the relative importance of predictor variables in the form of a percentage of total explained variance. Total explained variance for each equation was obtained by summing the squared betas (standardized coefficients). Explained variance attributable to each predictor variable was then calculated by dividing the squared beta by the total explained variance. Values may not sum exactly to 100% due to rounding. 2SLS = Two-stage least squares.

^a Denotes the predicted value of the variable.

*p ≤ .1. **p ≤ .05. ***p ≤ .01.