The Determinants of Consumer Satisfaction: the Moderating Role of Ambiguity

Youjae Yi, University of Michigan
ABSTRACT - This study has investigated the moderating role of ambiguity in the process of consumer satisfaction formation. It is found that product ambiguity moderates the way that expectation, performance, and disconfirmation determine consumer satisfaction. When the product is ambiguous, consumer expectations have direct effects on consumer satisfaction as well as indirect effects through disconfirmation. On the other hand, when the product is unambiguous or easy to evaluate, product performance has direct effects on consumer satisfaction as well as indirect effects via disconfirmation. The theoretical and practical implications of these findings are discussed as well.
[ to cite ]:
Youjae Yi (1993) ,"The Determinants of Consumer Satisfaction: the Moderating Role of Ambiguity", in NA - Advances in Consumer Research Volume 20, eds. Leigh McAlister and Michael L. Rothschild, Provo, UT : Association for Consumer Research, Pages: 502-506.

Advances in Consumer Research Volume 20, 1993      Pages 502-506

THE DETERMINANTS OF CONSUMER SATISFACTION: THE MODERATING ROLE OF AMBIGUITY

Youjae Yi, University of Michigan

ABSTRACT -

This study has investigated the moderating role of ambiguity in the process of consumer satisfaction formation. It is found that product ambiguity moderates the way that expectation, performance, and disconfirmation determine consumer satisfaction. When the product is ambiguous, consumer expectations have direct effects on consumer satisfaction as well as indirect effects through disconfirmation. On the other hand, when the product is unambiguous or easy to evaluate, product performance has direct effects on consumer satisfaction as well as indirect effects via disconfirmation. The theoretical and practical implications of these findings are discussed as well.

Consumer satisfaction (CS) is a central concept in modern marketing thought and practice. The marketing concept emphasizes delivering satisfaction to consumers and obtaining profits in return. As a result, overall quality of life is expected to be enhanced. Thus, consumer satisfaction is crucial to meeting various needs of consumers, business, and society. The realization of this importance has led to a proliferation of research on consumer satisfaction over the past two decades. Attempts to make significant contributions toward understanding this important area have been made, including numerous studies and annual conferences on consumer satisfaction/dissatisfaction and complaining behavior (e.g., Hunt and Day 1982, 1985; Oliver 1980). See Yi (1990) for a recent review.

Out of this empirical research has come the confirmation/ disconfirmation paradigm whereby consumer satisfaction is hypothesized to result from a process of comparison. Theoretical support for this model comes from the adaptation level theory positing that one perceives stimuli only in relation to an adapted standard (Helson 1964). The standard is a function of perceptions of the stimulus, the context, and the organism. Once created, the adaptation level serves to guide subsequent evaluations in that positive and negative deviations will remain in the general vicinity of one's original position. Oliver (1980) applied this theory to the study of consumer satisfaction by arguing that expectations about product performance can be seen as an adaptation level. He suggested that expectations create a frame of reference for comparative judgments.

According to the model, consumers judge satisfaction with a product in comparison with their expectations about the product performance. If the performance is above the expectations, an increase in satisfaction is expected. If the performance is below expectations, a decrease in satisfaction is expected. Disconfirmation is thus expected to affect consumer satisfaction. Oliver (1980) found that disconfirmation was positively related to consumer satisfaction. Positive disconfirmation (perceived performance above the expectation) increased consumer satisfaction, while negative disconfirmation (perceived performance below the expectation) decreased consumer satisfaction. Thus, consumer satisfaction is hypothesized primarily as a function of disconfirmation.

Although many studies accept this paradigm, the exact nature of CS processes is not so straightforward. For example, one might ask a question: What is the role of perceived performance in the CS model? One might view perceived performance merely as a standard for comparison to assess the confirmation or disconfirmation. In this view, all influences of perceived performance on CS are expected to be captured by disconfirmation. Another view is that perceived performance has its own influence on CS as another predictor. That is, perceived performance is expected to have a direct effect on CS, in addition to the indirect effect through disconfirmation.

The findings as to this issue are mixed. Some studies have found the direct effect of perceived performance on CS (e.g., Churchill and Surprenant 1982; Oliver and DeSarbo 1988; Tse and Wilton 1988). Oliver and DeSarbo (1988), for example, have found that perceived performance has its own effect on CS; in fact, the importance of perceived performance was second only to that of disconfirmation. Tse and Wilton (1988) also examined the role of perceived performance in consumer satisfaction formation with a tape recorder. The model with perceived performance outperformed other single predictor models with expectations or disconfirmation, and two-variable models with expectations and disconfirmation. In addition, perceived performance had indirect effects on consumer satisfaction through its influence on perceived disconfirmation. Thus, perceived performance seemed to have both direct and indirect effects (through its effect on disconfirmation) on satisfaction. Consumer satisfaction could be increased not only by minimizing disconfirmation, but also by increasing performance. On the other hand, the direct effect of perceived performance has not been examined or significant in other studies (e.g., Cadotte, Woodruff, and Jenkins 1987; Oliver 1980).

Another question might arise: What is the role of expectation in the CS model? That is, does it influence CS indirectly only through disconfirmation? Or does it affect CS directly as well? The findings are again mixed. On the one hand, some studies have found the direct effect of expectations on CS, in addition to the indirect effect mediated by disconfirmation (Bearden and Teel 1983; Churchill and Surprenant 1982, plant; Oliver and Linda 1981; Swan and Trawick 1981; Tse and Wilton 1988; Westbrook and Reilly 1983). On the other hand, the direct path from expectation to CS was not examined or significant in other studies (e.g., Cadotte, Woodruff and Jenkins 1987; Churchill and Surprenant 1982, video disc player; Oliver and Bearden 1983). For example, Oliver and Bearden (1983) found that expectations did not have any significant effects on consumer satisfaction, although the effect of disconfirmation was found.

In summary, there are mixed findings as to the antecedents of consumer satisfaction. Consumer satisfaction was found to be directly affected by expectations in some studies. On the other hand, some studies showed that expectation had little effect on consumer satisfaction, while perceived performance had a significant effect on consumer satisfaction. See Figure 1 for an overview.

These findings suggest that the effects of expectation and performance on consumer satisfaction may be more complex than hypothesized by the original expectation-disconfirmation model. How can we resolve the issue? The present study attempts to shed new light on this issue by taking a different perspective. Rather than asking whether or not there is a direct effect of a certain variable (e.g., expectation) on satisfaction, one can ask the following question: When does a certain variable have a direct effect on satisfaction? In other words, a shift is proposed from the "Is" question to the "When" question. The purpose of this study is therefore to identify conditions under which the effects of expectations, disconfirmation, and performance on satisfaction would be strong or weak. Specifically, product ambiguity is proposed as a moderator of the CS formation process.

FIGURE 1

EXPECTATION DISCONFIRMATION MODEL OF CS

THE ROLE OF AMBIGUITY IN CS PROCESSES

The above review suggests that the views vary across studies as to how expectation, disconfirmation, and product performance influence CS. One factor in particular that has been overlooked is the degree to which ambiguity associated evaluation of the product affects CS. In other words, how does product ambiguity moderates the formation of CS?

Product experience can often be ambiguous, such as when the quality of a product is difficult to evaluate. The difficulty may arise when a product (e.g., clothes or insurance) cannot be judged based on objective criteria, or when a product (e.g., diamond) may have many credence qualities or subjective attributes (Darby and Karni 1973; Holbrook 1978). Alternatively, there might be a potential for multiple interpretations of product quality (e.g., Hoch and Deighton 1989). Sometimes it is difficult to determine what is acceptable, desired, or valued from the product. If product experience is difficult to evaluate or ambiguous, consumers are likely to lack confidence about their performance ratings.

The concept of ambiguity can be illustrated by looking at compact disc players. When compact disc players first came on the market, many consumers did not know how to judge what made a good compact disc player. Even after a number of years on the market, it is still difficult to evaluate compact disc players as they offer many different features, come in different sizes and power capacities, produce very different sound reproduction levels, and vary greatly in price. This example suggests that product evaluations can often be ambiguous.

It is proposed in this study that ambiguity will affect the relative importance of predictor variables on satisfaction. Human judgements in general pose a contest between expectations and evidence, prior expectations and situational information, or theory and data (Alloy and Tabachnik 1984; Bobrow and Norman 1975). In the context of consumer satisfaction, expectations are consumers' a priori theories about the product, whereas product performance represent the influence of the data or evidence provided in a given situation. When products are ambiguous, consumers' satisfaction is hypothesized to be determined largely by their prior expectations. If there is no objective way of judging a product, consumers may subjectively judge the product based on prior expectations. That is, consumers use a top-down, theory-driven process to judge satisfaction. Such an assimilative processing has been found when the evidence is ambiguous (Herr, Sherman and Fazio 1982; Hoch and Deighton 1989). Ambiguous information is easy to assimilate because consumers believe that ambiguous evidence is not actually ambiguous (Hoch and Ha 1986). Self-perception theory also suggests that consumers will reply on internal cues such as prior expectations when confronted with ambiguous information.

When products are unambiguous or easy to evaluate, on the other hand, consumers are likely to follow a bottom-up, data-driven process to evaluate a consumption experience. Unambiguous evidence is difficult to assimilate, because the evidence is relatively obvious and clear to the consumers. In such cases, perceptions of product performance are formed with conviction and they should be highly diagnostic in the formation of CS. That is, for products which can be objectively judged, consumers are certain of the performance levels of the products. Thus, when products are unambiguous, consumers' satisfaction judgements would be determined primarily by product performance.

Based on the above arguments, the following hypotheses are proposed.

H1. When product experience is ambiguous, consumer expectation is highly likely to have direct effects on CS.

H2. When product experience is unambiguous, product performance is highly likely to have direct effects on CS.

METHOD

A pretest was conducted to determine which products are easy or difficult to evaluate. Forty one students at the University of Michigan participated in the pretest. They rated the difficulty of evaluating product quality for twenty categories on a ten-point scale. From these twenty products two were to be selected and used in the main study. Ten of the products were low-involvement, frequently purchased products, whereas the other ten products were high-involvement, infrequently purchased products. These products were considered familiar to the potential subjects of the study. Comparison of these products were also made with regard to the following: the amount of information seeking done before the purchase decision and the relative importance of the product.

TABLE 1

SUMMARY OF RESULTS

Laundry detergent, insurance, aspirin, camera, microwave oven, and computer were rated among the difficult to evaluate, whereas soft drinks, cereal, jeans, and ball-point pen were rated as easy to evaluate. Cereal and laundry detergent were selected as low and high ambiguity products, respectively, based on the mean ratings on difficulty of evaluation (Ms=2.75 vs. 6.27, p<.001). These products were relatively similar in other aspects such as information seeking and importance.

A main study was then conducted to compare the CS processes across low and high ambiguity products. Subjects consisted of 117 business students at the University of Michigan who were asked to respond to a consumer satisfaction survey. They were first asked to state the brand of laundry detergent that they had most recently purchased. The most recently purchased brand was chosen because the preferred brand may not have been purchased most recently, and the image brought forth by the preferred brand may present biases in consumers' judgments of the products. They were then asked to evaluate the most recently purchased brand. Next, they were asked similar questions about the cereal. Thus, each subjected responded to the surveys for both laundry detergent and cereal.

Multiple measures were used to assess key constructs. For example, two measures of perceived performance were taken. One measure consisted of the following question: "How do you evaluate the quality of of the detergent (cereal) after having experience with it?" A 7-point scale was used to record responses (1=very low, 7=very high). Another measure was obtained by asking how good each respondent found the detergent (cereal) to be. Similarly, each of the other key constructs (expectation, disconfirmation, and consumer satisfaction) were measured with two indicators.

Three measures were obtained to ascertain that product ambiguity indeed varied between the two products as expected. First, the absolute difficulty of evaluation was assessed by asking respondents "Overall, how difficult was it to evaluate this product?" The responses were given on the 7-point scale (1=very easy, 7=very difficult). The mean ratings showed that there were expected variations in ambiguity (Ms=4.72 vs. 6.07, p<.01). Second, the relative measure of evaluation difficulty was obtained with a question "How difficult was it to evaluate this product, compared to an ink pen?" The responses ranged from "much easier" (1) to "much more difficult." The results indicated that cereal was considered easier than laundry detergent (Ms=3.28 vs. 5.14, p<.01). Third, subjects' confidence in evaluation was measured by asking them how confident they were in their judgments of the product quality. As expected, subjects had less confidence in the evaluation of laundry detergent than that of cereal (Ms=4.20 vs. 5.83, p<.01). These manipulation checks showed that the two products were indeed different in ambiguity.

In addition, the importance of the product was also measured and compared. The results showed that there was no significant difference in the importance of the product to respondents. This check gave a rough indication that there was no confounding of ambiguity and involvement in the study.

RESULTS

The model in Figure 1 was estimated for the laundry detergent (high ambiguity) and cereal (low ambiguity) products, respectively. Since multiple measures were available for all constructs, the data were analyzed via LISREL. Table 1 provides a summary of results.

When the model was fit to the cereal data, it gave the following results: c2 (14)=27.71, p=.02, GFI=.95. The direct path from performance to CS was .67, which is statistically significant at the 0.5 level. On the other hand, the direct effect of expectation on CS was .32, which is not significant. These results supported the hypothesis that product performance is likely to have direct effects for low ambiguity products.

FIGURE 2

CS PROCESSES UNDER LOW AND HIGH AMBIGUITY

The model was also fit to the laundry detergent data, and the following results were obtained: c2 (14)=34.39, p=.002, GFI=.93. The direct effect of performance on CS was not significant (.19 with the t-value of .74), whereas the direct effect of expectation on CS was significant (.67 with the t-value of 3.0). These results supported the hypothesis that expectation is likely to have direct effects on CS for high ambiguity products.

Figure 2 provides a summary of key results in the model. Only significant paths are included in this figure to gain a better understanding of the CS processes. We can note several findings that are common to low and high ambiguity products. First, expectation and performance had significant effects on disconfirmation, which in turn had significant effects on CS. That is, expectation and performance had indirect effects on CS through disconfirmation for both products. Second, perceived performance was influenced by expectation under low and high ambiguity.

We can also note some interesting differences between low and high ambiguity products. First, the direct effect of expectation was significant under high product ambiguity, but not under low product ambiguity. Second, the direct effect of performance was significant under low product ambiguity, but not under high ambiguity. Third, the impact of expectation on performance was greater under high ambiguity than under low ambiguity. Fourth, the model fit was better under low ambiguity than under high ambiguity.

DISCUSSION

The findings of the present study have several implications. First, it suggests that the nature of CS process varies across product categories. This result challenges the implicit assumption of the conventional model that CS formation processes are the same across product categories. Specifically, when products are ambiguous, the direct effect of expectation increases and the direct effect of performance is reduced. When products are unambiguous, the direct effect of expectation decreases and the direct effect of performance increases. Thus, a marketer needs to know how ambiguous the product is for a better understanding of the CS process.

This study has implications for advertising as well. If a marketer has discovered that he or she has an ambiguous product, then the marketer might wish to create high expectations. The marketer can make claims that can neither be proved nor disproved. Emotional, rather than rational, appeals could be used in the promotion campaign to build up consumer expectations. When products are ambiguous and the performance cannot be directly observed or tested, consumer expectations are likely to influence product evaluations and satisfaction judgments. On the other hand, when products are unambiguous, the role of expectations will be reduced.

The finding has different implications for topdog marketers (market leaders) and underdog marketers (new or existing companies with market share below topdogs). The marketers of topdog products would benefit by increasing subjective criteria for judging their products, because consumers' expectations would color product evaluations. Provided that consumer expectations about topdog products are high, their evaluations would be favorable. On the other hand, marketers of underdog products should try to educate consumers on how products can be objectively judged, especially if the underdog product is in fact comparable to or better than the topdog product. Having a comparable or superior product may not lead to higher CS, if the product is viewed ambiguous by consumers.

Some limitations of the present study should be noted. We have measured product ambiguity in a survey, rather than manipulating it in an experiment. One may need to manipulate ambiguity in a controlled experiment. This would ensure that one is in fact testing the effects of ambiguity and not some other confounding factors. Also, the current operationalization of ambiguity is rather limited, and one should find a better way to measure or determine the level of ambiguity associated with a product. The following are additional questions that may help one to determine the degree of ambiguity associated with a product.

Do you think that everyone would evaluate this product on the same attributes?

How many attributes do you think people use to evaluate this product? List as many as you can.

How easy or difficult is it for you to choose criteria with which to judge this product?

List characteristics which you feel make this product easy or difficult to evaluate.

Which of the following words (phrases) best describes the process of evaluating this product: easy-difficult, quick-slow, consistent-inconsistent, etc.?

This research is in part supported by the Sanford R. Robertson Assistant Professorship at the Michigan Business School. I wish to thank Kent Nassen for his assistance in this research project.

REFERENCES

Alloy, Lauren B. and Naomi Tabachnik (1984), "Assessment of Covariation by Humans and Animals: The Joint Influence of Prior Expectations and Current Situational Information," Psychological Review, 91 (January), 112-149.

Bearden, William O. and Jesse E. Teel (1983), "Selected Determinants of Consumer Satisfaction and Complaint Reports," Journal of Marketing Research, 20 (February), 21-28.

Bobrow, Daniel G. and Donald A. Norman (1975), "Some Principles of Memory Schemata," in Representation and Understanding: Studies in Cognitive Science, eds. D. G. Bobrow and A. Collins, New York: McGraw-Hill.

Cadotte, Ernest R., R. B. Woodruff, R. L. Jenkins (1987), "Expectations and Norms in Models of Consumer Satisfaction," Journal of Marketing Research, 24 (August), 305-314.

Churchill, Gilbert A., Jr. and C. Surprenant (1982), "An Investigation into the Determinants of Customer Satisfaction," Journal of Marketing Research, 19 (November), 491-504.

Darby, Michael R. and Edi Karni (1973), "Free Competition and the Optimal Amount of Fraud," Journal of Law and Economics, 16 (April), 67-88.

Helson, Harry (1964), Adaptation-level Theory, New York: Harper & Row.

Herr, Paul, Steven J. Sherman, and Russell H. Fazio (1982), "On the Consequences of Priming: Assimilation and Contrast Effects," Journal of Experimental Social Psychology, 19, 323-340.

Hoch, Stephen J. and John Deighton (1989), "Managing What Consumers Learn from Experience," Journal of Marketing, 53 (April), 1-20.

Hoch, Stephen J. and Y. Ha (1986), "Consumer Learning: Advertising and the Ambiguity of Product Experience," Journal of Consumer Research, 13 (September), 221-233.

Holbrook, Morris B. (1978), "Beyond Attribute Structure: Toward the Informational Determinants of Attitude," Journal of Marketing Research, 15 (November), 545-556.

Hunt, H. Keith and Ralph L. Day (1982), Conceptual and Empirical Contributions to Consumer Satisfaction and Complaining Behavior, Bloomington, IN: Indiana University.

Hunt, H. Keith and Ralph L. Day (1985), Consumer Satisfaction, Dissatisfaction and Complaining Behavior, Bloomington, IN: Indiana University.

Oliver, Richard L. (1980), "A Cognitive Model of the Antecedents and Consequences of Satisfaction Decisions," Journal of Marketing Research, 17 (September), 46-49.

Oliver, Richard L. and William O. Bearden (1983), "The Role of Involvement in Satisfaction Processes," in Advances in Consumer Research, Ann Arbor, MI: Association for Consumer Research.

Oliver, Richard L. and Wayne S. DeSarbo (1988), "Response Determinants in Satisfaction Judgments," Journal of Consumer Research, 14 (March), 495-507.

Oliver, Richard L. and G. Linda (1981), "Effects of Satisfaction and Its Antecedents on Consumer Preference and Intention," in Advances in Consumer Research, Kent B. Monroe, ed. Ann Arbor, MI: Association for Consumer Research, 88-93.

Swan, John. E. and I. Frederick Trawick (1981), "Disconfirmation of Expectations and Satisfaction with a Retail Service," Journal of Retailing, 57 (Fall), 49-67.

Tse, David K. and Peter C. Wilton (1988), "Models of Consumer Satisfaction: An Extension," Journal of Marketing Research, 25 (May), 204-12.

Westbrook, Robert and Michael D. Reilly (1983), "Value-Percept Disparity: An Alternative to the Disconfirmation of Expectations Theory of Consumer Satisfaction," in Advances in Consumer Research, Richard P. Bagozzi and Alice M. Tybout, eds. Ann Arbor, MI: Association for Consumer Research.

Yi, Youjae (1990), "A Critical Review of Consumer Satisfaction," in Review of Marketing 1990, ed. Valarie A. Zeithaml, Chicago; American Marketing Association, 68-123.

----------------------------------------