Franklin Publishing Company

Phone: 920-424-3152

Fax: 920-424-7312

AN ANALYSIS OF EXPECTATIONS, PERCEIVED PERFORMANCE,

AND DISCONFIRMATION FOR E-SATISFACTION

ABSTRACT

Though consumers obtain more information and entertainment from new technological devices and do more online shopping and purchasing, their e-satisfaction has been relatively low. Thus, the main purpose of this study is to investigate the relationship between online satisfaction and its main constructs: expectations, perceived performance, and disconfirmation. Empirically, this study finds that all three constructs have a positive relationship with satisfaction and that all three constructs have different levels of explanatory power for e-satisfaction. Perceived performance and expectations explain e-satisfaction better than disconfirmation. This finding suggests that if e-retailers want to improve their clients’ satisfaction, they should pay attention to their performance rather than to disconfirmation. It also suggests that future study should investigate whether e-shopping is a high involved or low involved event because the level of involvement may dictate the key determinant of e-satisfaction.

INTRODUCTION

As technology continues to improve, the prices of electronic devices such as computers, cell phones, internet access, and others have dropped dramatically. Thus, more consumers have enjoyed these items at low costs and have obtained more information and entertainment from these media. At the same time, online shopping and purchasing have increased. However, Overby and Lee (2006) argue that a majority of online shoppers exit at the checkout point without making any purchase even after filling their electronic shopping carts. The Internet shopping conversion rate, the number of visitors who come to a particular retail site divided by the number of actual buyers, is very low. One of the main reasons for this low conversion rate is low online customer satisfaction.

Though many theories have been developed to explain consumer satisfaction, the disconfirmation theory became the dominant theoretical paradigm in satisfaction studies (Oliver, 1997; Yi, 1990) and provides strong evidence on how consumers form their satisfaction. Expectancy disconfirmation theory indicates that the level of satisfaction depends on the difference between the level of perceived performance of the product and the consumer's expectations of the product (Folkes, 1984; Ilgen, 1971; Oliver, 1980; Weaver and Brickman, 1974; and Woodruff, Cadotte and Jenkins, 1983). Under this theory, consumers form expectations of product performance prior to purchase. These expectations are derived from past experience with the product itself, past experience with similar products, other marketing stimuli, and existing attitudes and confidence felt of the consumer. These expectations are compared to the perceived performance. A product that is worse than expected creates negative disconfirmation, and one that is better than expected creates positive disconfirmation. Consumers obviously prefer positive disconfirmation. Thus, the main purpose of this study is to investigate the relationship between online satisfaction and its main constructs: expectations, perceived performance, and disconfirmation. The paper has three sections: (1) hypotheses based on the previous studies, (2) research methodology to test these hypotheses, and (3) findings and implications.

HYPOTHESES

The following three hypotheses are developed to identify the relationship between satisfaction and three constructs: expectations, perceived performance, and disconfirmation.

H1: Satisfaction has a positive relationship with expectations.

H2: Satisfaction has a positive relationship with perceived performance.

H3: Satisfaction has a positive relationship with disconfirmation.

These hypotheses are based on the most commonly used and supported consumer satisfaction model:

Satisfaction = f (Disconfirmation)

= f (Expectations - Perceived Performance)

This model indicates that consumer satisfaction is influenced by the difference between the perceived actual performance of a product and the expectations of the product. This relationship is based on theories of expectancy disconfirmation (Ilgen, 1971; Weaver and Brickman, 1974; Oliver 1980) and on norms and attributes (Woodruff, et al., 1983; Folkes, 1984). As the literature review section shows, this model also implies that satisfaction is either directly influenced by expectations and/or perceived performance or indirectly influenced by the two variables through disconfirmation.

The first hypothesis relates to consumer expectations. These expectations are derived from experiences with the product itself or similar products, from experiences of relevant others with the product, and/or from the firm's marketing communications. In general, expectations are formed before consumers purchase and use the product. The mathematical formula shows that the lower expectations are, the lower satisfaction will be. However, this mathematical relationship is valid only when expectations are compared to perceived performance. When expectations are analyzed alone without other variables, they have a positive relationship with satisfaction. The rationale is that without some positive expectations no one will consider online purchasing a product or service.

The second hypothesis is about perceived performance. The model shows that when other variables are constant, perceived performance has a positive relationship with satisfaction. The last hypothesis addresses disconfirmation, the gap between expectations and perceived performance. After purchase, consumers compare their expectations to their perception of the online shopping experience (perceived performance), and decide if the experience is satisfactory or dissatisfactory. A shopping experience that is better than expected generates positive disconfirmation; conversely, an experience that is poorer than expected generates negative disconfirmation (Oliver, 1997). Thus, disconfirmation has a positive relationship with satisfaction. In other words, the higher the perceived performance is compared to expectations, the higher consumer satisfaction is.

METHODOLOGY

Regression analysis is used to test the hypotheses on e-satisfaction. This analysis can show how much of the variance is explained by expectations, perceived performance, and disconfirmation. In addition, each hypothesis is reviewed with the sign, direction, and level of significance. The three independent variables of expectations, perceived performance, and disconfirmation with items for e-satisfaction are reduced to indices, with each of the items correlated with the appropriate index. The index for each variable is based on at least .50 correlation with each item. The indices are refined by deleting the items correlating less than the cut-off, recalculating the indices, and rerunning the correlations. The coefficient Alphas for perceived performance and disconfirmation also are measured.

RESULTS

This study adopts two steps to test the three hypotheses. First, it develops three indices for expectations, perceived performance and disconfirmation based on the 15 items relating to e-satisfaction. Here disconfirmation is defined as the difference between expectations and perceived performance. Table 1 shows the correlation coefficients of significant items for the three constructs. The expectations and perceived performance indices are based on 10 significant items, and the disconfirmation index is based on four items. This analysis indicates that the disconfirmation index is different than those of expectations and perceived performance. All three indices receive high reliability scores, and the perceived performance index has the highest reliability score, .874.

Second, these indices are used for a simple regression analysis. In this model, satisfaction is a function of one of the three indices, which is converted from the 15 items after treating all of them as a single variable. The main purpose of this analysis is to determine the sign of the beta coefficient and find the explanatory power of each model. Table 2 contains the standardized regression coefficients of the three models and other summary statistics. It reports that all three models have a positive sign, and these signs were significant at the 0.01 level. These results support all three hypotheses. The expectations and perceived performance equations achieve adjusted R squares of .226 and .276, respectively. However, the disconfirmation equation receives the lowest adjusted R square of the three models. Thus, in the case of e-shopping, perceived performance and expectations have more explanatory power than disconfirmation.

SUMMARY

To improve the understanding of satisfaction for e-shopping, this study examines the relationship between satisfaction and the known constructs: expectations, perceived performance, and disconfirmation. Empirically, this study focuses on the models based on these constructs. It finds that all three constructs have a positive relationship with satisfaction, and that all three constructs have different levels of explanatory power for e-satisfaction. Perceived performance and expectations explain e-satisfaction better than disconfirmation. This finding suggests that if e-retailers want to improve their clients’ satisfaction, they should pay more attention to their clients’ expectations and their performance rather than to disconfirmation. This finding is different than the common belief on satisfaction, that the disconfirmation theory should be the best means for consumer satisfaction. One possible speculation is that e-shopping is a high involved event rather than a low involved event because the level of involvement may dictate the key determinant of e-satisfaction. However, this study suggests that future research should identify whether e-shopping is a high involved or a low involved event and then investigate the determinants of e-satisfaction at each level. Because this study contains limitations, it should be considered exploratory, which raises issues for further investigation. One limitation of this study is that its sample selection is on a convenience basis. Another limitation is that the study is conducted in only one region in the midwestern United States, though each region has its own unique lifestyle including online shopping.

REFERENCES

Folkes, Valerie S., 1984. "Consumer Reactions to Product Failure: An Attributional Approach." Journal of Consumer Research, 10 (March), 398-409.

Ilgen, Daniel R., 1971. "Satisfaction with Performance as a Function of the Initial Level of Expected Performance and the Deviation from Expectations." Organizational Behavior and Human Performance, 6 (January), 345-356.

Overby, Jeffrey W. and Eun-Ju lee, 2006. "The Effects of Utilitarian and Hedonic Online Shopping Value on Consumer Preference and Intentions," Journal of Business Research, 59, 1160-1166.

Parasuraman, A., Valarie A. Zeithaml, and Leonard L. Berry, 1988. "SERVQUAL: A Multiple Item Scale For Measuring Consumer Perceptions of Service Quality," Journal of Retailing, 64(Spring), 12-37

Weaver, Donald and Philip Brickman, 1974. "Expectancy, Feedback, and Disconfirmation as Independent Factors in Outcome Satisfaction," Journal of Personality and Social Psychology, 30(March), 420-428.

Woodruff, Robert., Ernest R. Cadotte, and Roger L. Jenkins, 1983. "Modeling Consumer Satisfaction Processes Using Experience-Based Norms," Journal of Marketing Research, 20(August), 296-304.

Zaichkowsky, Judith Lynne, 1985. "Measuring the Involvement Construct," Journal of Consumer Research, 12(December), 341-352.