A Contingency Framework For Predicting Causality Between Customer Satisfaction and Service Quality

ABSTRACT - There is extensive literature on customer satisfaction and service quality, but little investigation of the causal sequence between these two constructs. Past models of this sequence are based on unique conceptualizations of the constructs with little agreement on the issue. A contingency framework is presented wherein different causal sequences between customer satisfaction and service quality occur under different service situations. By examining the determinants of causality between satisfaction and quality, the framework enhances our understanding of the customer evaluation process. Furthermore, the different contingency-related sequences discussed in the framework are predicted to have varying effects on future customer behavior, thus being of critical interest to practitioners. An agenda for future research on causality between satisfaction and quality and its impact on customer behavior is outlined. Strategic implications for service design and employee training are suggested.

Citation:

Pratibha A. Dabholkar (1995) ,"A Contingency Framework For Predicting Causality Between Customer Satisfaction and Service Quality", in NA - Advances in Consumer Research Volume 22, eds. Frank R. Kardes and Mita Sujan, Provo, UT : Association for Consumer Research, Pages: 101-108.

Advances in Consumer Research Volume 22, 1995      Pages 101-108

A CONTINGENCY FRAMEWORK FOR PREDICTING CAUSALITY BETWEEN CUSTOMER SATISFACTION AND SERVICE QUALITY

Pratibha A. Dabholkar, University of Tennessee

ABSTRACT -

There is extensive literature on customer satisfaction and service quality, but little investigation of the causal sequence between these two constructs. Past models of this sequence are based on unique conceptualizations of the constructs with little agreement on the issue. A contingency framework is presented wherein different causal sequences between customer satisfaction and service quality occur under different service situations. By examining the determinants of causality between satisfaction and quality, the framework enhances our understanding of the customer evaluation process. Furthermore, the different contingency-related sequences discussed in the framework are predicted to have varying effects on future customer behavior, thus being of critical interest to practitioners. An agenda for future research on causality between satisfaction and quality and its impact on customer behavior is outlined. Strategic implications for service design and employee training are suggested.

INTRODUCTION

Businesses today are well aware that they must satisfy customers and offer quality services in order to be competitively viable. For several years now service firms have been measuring customer satisfaction and service quality to gauge how well they are meeting customer needs. It is interesting that these measures of customer satisfaction and service quality are often used interchangeably (Cooper, Cooper, and Duhan 1989) indicating that practitioners may not see much difference between the two constructs. Consequently, there has been little discussion or research on the causal link between the constructs. Recently, however, researchers have suggested that the direction of the relationship between customer satisfaction and service quality be explored in order to increase our understanding of service evaluation processes used by customers (Bolton and Drew 1991; Cronin and Taylor 1992).

Some researchers (e.g., Iacobucci, Grayson, and Ostrum 1994) suggest that causality between customer satisfaction and service quality may be relevant only for an understanding of the consumer's evaluation but not necessarily for managerial considerations. It is true that practitioners are mainly concerned about customer satisfaction and service quality as predictors of customer behavior (Reichheld 1993) and not about the constructs themselves. This is certainly understandable; it is customer behavior, such as repeat purchase and word-of-mouth, that directly affects the viability and profitability of the firm. However, if the direction of the causal link between customer satisfaction and service quality were to lead to different customer behavior outcomes, this issue would have considerable managerial relevance in addition to enhancing our understanding of the process of customer evaluation.

The purpose of this paper is to propose a contingency framework where the causal link between customer satisfaction and service quality varies for each contingency and where it impacts customer behavior in different ways based on the causality. Such a framework would allow us to gain a deeper understanding of how customers evaluate services and to predict the type of evaluation processes that would lead to repeat purchases and customer loyalty.

BACKGROUND

There has been much recent debate regarding the overlap between the customer satisfaction and service quality constructs at the conceptual level and in terms of measurement (e.g., Bitner and Hubbert 1994; Dabholkar 1993; Iacobucci, Grayson, and Ostrom 1994; Parasuraman, Zeithaml, and Berry 1994; Spreng and Singh 1993). Researchers have proposed different conceptualizations of the two constructs and their relationship to each other. Most of these proposed relationships are dependent on the definitions of the constructs themselves. For example, some researchers (Bitner 1990; Oliver 1981; Parasuraman, Zeithaml, and Berry 1988) visualize customer satisfaction as experiential or occurring at the transactional level and service quality as an attitude at the global level; thus, they propose a causal link from customer satisfaction to service quality. Other researchers (Bolton and Drew 1992; Drew and Bolton 1991) point out that customers can and do evaluate service quality at the transactional level and customer satisfaction may be quite meaningful at the global level; therefore, they propose a causal link from service quality to customer satisfaction. Still others (Anderson and Sullivan 1993; Oliver 1993) view service quality as one of the antecedents of customer satisfaction and propose a link from service quality to customer satisfaction.

At first glance it may appear that the prediction of causality between customer satisfaction and service quality is merely academic and based superficially on the conceptualization of satisfaction and quality. However, most of this literature may be quite consistent if it can be drawn into a broad contingency framework. It is proposed that different conceptualizations of customer satisfaction and service quality are related to different contingencies. An understanding of these contingencies should allow us to predict the causal link between customer satisfaction and service quality across situations and to study the influence of this directionality on customer behavior.

DEVELOPING A CONTINGENCY FRAMEWORK

Evaluations of a service may occur prior to experiencing the service (based on word-of-mouth and/or promotion), during and immediately after a service experience, or after several experiences with a given service. In general, if a customer has not experienced the service, s/he could express an evaluation of service quality (based entirely on word-of-mouth and/or promotion) but satisfaction would be impossible to evaluate without experience (see Oliver 1993). Thus, causality between the two constructs is irrelevant prior to experiencing the service. Similarly, after experiencing several instances of the service delivery, the customer would integrate his/her evaluations of the service across all these instances and form an overall evaluation of service quality which would be very similar to his/her overall satisfaction with the service and both would evolve simultaneously over time. Thus, causality between the two constructs at the global level is also not relevant. When it comes to the actual service encounter, however, evaluations of quality and satisfaction may occur in different ways for different situations and different people. Here, causality could be quite meaningful especially if it impacts future customer behavior in different ways. Thus, the contingency framework is developed with a focus on specific service encounters.

There appears to be agreement in the literature that the evaluation of service quality is mainly cognitive (Bitner 1990; Parasuraman, Zeithaml, and Berry 1988), while the evaluation of customer satisfaction is a combination of cognitive and affective elements (Mano and Oliver 1993; Woodruff, Cadotte, and Jenkins 1983; Yi 1990). Some researchers suggest that customer satisfaction is in fact mainly affective (Oliver 1989; Westbrook 1987), which would be particularly true for service experiences. This difference between the two constructs can be used to predict different causal links for different service experiences. It is proposed that for service experiences where cognitions are formed first, the causal link will be from service quality to customer satisfaction, provided that some type of affect follows the initial cognitions. However, if there is little or no affect following the cognitions in relation to the experience, there will be an overlap between the two constructs. Finally, if there is a strong emotional reaction to the service, this will influence later cognitions about the service, and the causal link will be from customer satisfaction to service quality. Even if initial cognitions are formed in assessing discrepancies in the service that lead to emotional reactions, these cognitions will take second place to the strong emotional reaction that would influence later cognitions about the service.

TABLE 1

CONTINGENCY FRAMEWORK FOR PREDICTING CAUSALITY BETWEEN CUSTOMER SATISFACTION (CS) AND SERVICE QUALITY (SQ)

For example, a customer goes to the store where the service is fairly good (or somewhat substandard) and no unforeseen circumstances arise. There is little chance in this situation for strong emotion. The customer may say to him/herself "the service wasn't bad (or wasn't too good)" and then begin to feel good (bad) about it. This would mean that the customer evaluated the service quality at the cognitive level and then experienced satisfaction at the affective level based on the cognitive evaluation of service quality. Or, the customer may not feel good (or bad) about the service and may merely note at the cognitive level that the service was okay (or not). If service quality and customer satisfaction were to be measured in such situations, there would be no discriminant validity between the two constructs and no causal sequence. Now, if the customer goes to a service provider where something happens to delight (or disgust) him/her, this customer will experience strong emotion which would form a significant part of his/her satisfaction with the service experience. For the customer who is delighted, even if the rest of the service was not quite up to standard, the delightful aspect and the associated emotion will color cognitions that are subsequently formed about the service. In the same way, if everything had been proceeding fairly well, but something happens to upset the customer, s/he will experience strong emotion (perhaps anger) and this will influence later cognitions about the service. Various situations where the two possible causal sequences are likely to occur are presented in Table 1.

Zone of Indifference

If the customer's evaluation of the service is within an acceptable range called the zone of indifference (Erevelles and Leavitt 1992), the customer is unlikely to experience strong emotion and the causal link will be from service quality to customer satisfaction. If on the other hand, the customer's evaluation of the service falls outside the zone of indifference, emotional reactions (delight or anger) are more likely (Bloemer and Poiesz 1989) and the causal link will be from customer satisfaction to service quality.

Early definitions of customer satisfaction have included an element of surprise (Oliver 1981). Customers are likely to be surprised if the service experience falls outside the zone of indifference. As Spreng and Droge (1994) point out, surprise, itself, although often included in lists of emotional reactions (e.g., Izard 1977), is not an emotion given that it can be both positively and negatively valenced. It is a cognition that acts as an emotion intensifier (Ortony, Clore, and Collins 1988). Thus, it is surprise that gives rise to strong emotion when the evaluation of the service is outside the zone of indifference and creates a feeling of strong satisfaction or dissatisfaction with the service. If the customer is not surprised, this indicates that the experience was within the range acceptable to the customer, who should therefore be able to rationally and cognitively evaluate the quality of the service.

Essential Aspects versus Service Enhancers

All services have certain aspects that are essential to the service and others that are not essential but enhance the service experience (e.g., satisfiers versus dissatisfiers, Cadotte and Turgeon 1988; core versus peripheral (factors), Iacobucci et al. 1994). If the essential factors are present, they are often not even noticed. Customers expect to get medical attention at the doctor's, to reach the intended destination when taking a flight, and to have a place to park at any given service site. Such outcomes are not noticed because they are expected as a matter of course and when they are present, customers are likely to evaluate the service at a cognitive level. If, however, these factors are absent, i.e., the customer ends up not seeing the doctor, the flight is rerouted to another airport, or there is absolutely no parking available, the customer is likely to be upset. These experiences are outside the zone of indifference and will be noticed. They are likely to cause strong emotional reactions which will influence customers' subsequent evaluations of the service.

On the other hand, when service enhancers are absent, they may not be noticed, but if they are present, they will be. If the waiting time at the doctor's office is typical or the customers eventually find what they are looking for in the store, they may not be too concerned that the service could have been better. They did receive adequate service based on their expectations and they will therefore evaluate the experience cognitively. However, if the customer is called in to the doctor's examining room a minute after arrival and examined almost immediately or the store personnel go out of their way to help and find what the customer wants right away, the customer will feel very good about such experiences and will base subsequent cognitions on these feelings. Even if something else (minor) goes wrong during the service delivery, the good feelings will tend to override the evaluation of such discrepancies.

Type of Service

Some service situations themselves may be inherently more emotional than others, such as going to the hospital emergency room. In this case, the emotional evaluation of the service is likely to influence cognitions, and the directionality will be from customer satisfaction to service quality. Other services inherently will have low emotional possibilities such as shopping for groceries and customers will tend to evaluate such services cognitively. It may not be possible to neatly classify all service encounters into this dichotomy and hence, type of service should be viewed more as a continuum. Getting a haircut does not compare with the emergency room for emotional possibilities, but a bad haircut may be quite emotional for the customer and may influence subsequent evaluations of the service provider.

Type of Customer

Some customers may be more cognitive than others and be more likely to evaluate experiences rationally at the cognitive level. Individuals with a high need for cognition (Cacioppo and Petty 1984) tend to look for information and to think about different aspects of their experiences. Such customers will evaluate service quality first and then decide whether they are satisfied. Other customers may be more emotional and are likely to evaluate experiences at the emotional or feeling level. Individuals with a high affect intensity (Larsen and Diener 1986) tend to magnify their emotions and feel experiences more deeply than others. Such customers will experience (dis)satisfaction with the service at an emotional level and then form a judgment about the service based on this emotion.

Customer's Mood

The customer's mood in the service situation is also likely to influence causality between customer satisfaction and service quality. Even mild moods can have a significant impact on behavior and evaluation (Gardner 1985). If the customer is in a very bad (or good) mood it is likely that s/he will evaluate the service experience emotionally and the causal sequence will be from customer satisfaction to service quality. If the customer is in a somewhat neutral mood (recognizing that there is no such thing as a completely neutral mood), then s/he is likely to evaluate the service at a cognitive level and the directionality will be from service quality to customer satisfaction.

INTERACTION OF CONTINGENCY SITUATIONS

The situations discussed above and presented in Table 1 are not necessarily independent. In fact, there is likely to be considerable interaction among these factors, and the causal sequences are most likely to occur if there are several factors present. For example, if an emotional customer in a bad mood has a service experience which is outside (and below) the zone of indifference, the directionality will be from customer (dis)satisfaction to service quality and the effect will be quite strong. It may also be noted that there are other factors that may influence the sequence of evaluation of the service that are not indicated in Table 1 because these may not have a clear main effect on the sequence. For example, the shopping situation, i.e., whether one is shopping alone or with friends/family is likely to affect whether one reacts emotionally or not, but different people may react differently in such situations. Thus, the influence of the shopping situation on the directionality between customer satisfaction may be manifest only through interaction with other factors.

Three possible interactions of the factors listed in Table 1 are presented in Figure 1. In practice, many more such interactions are possible, but these three examples are quite representative of the gamut of possibilities and each is just a little different to suggest the range of influences on the directionality between customer satisfaction and service quality.

As seen in Figure 1A, customers who rate low (or high) on both need for cognition and affect intensity will not favor a particular sequence for forming evaluations of the service in terms of service quality and customer satisfaction. The constructs are likely to overlap in these two cases. The individual who is low on both factors may simply form an overall sense of satisfaction and quality and thus the two constructs will overlap. The individual who is high on both factors may think in great detail about each aspect of the service, but may simultaneously experience emotions about each aspect. Thus, this individual too will form evaluations of satisfaction and quality simultaneously, but the evaluations will be richer in content. Now, customers who have a high need for cognition and low affect intensity are more than likely to evaluate experiences cognitively and then decide if they are satisfied with the service or not. Thus, their evaluations will show a causal sequence from service quality to customer satisfaction. On the other hand, customers with a low need for cognition and high affect intensity are more than likely to evaluate experiences on an emotional basis and then form cognitions about the service based on these emotions. Thus, they will go from customer satisfaction to service quality.

The interactions in Figure 1B are a little more complicated. Here, if the essential aspects of a service are present and service enhancers are absent, the presence or absence of these factors is unlikely to be noticed and thus an emotional reaction is unlikely. The customer will first form cognitions about the service and the directionality of evaluation will be from service quality to customer satisfaction. If the essential aspects of a service and service enhancers are both present, the service enhancers will be noticed and their presence will be appreciated and will make customers feel good about the service, and in turn, influence cognitions about the service. Thus, the causal sequence of evaluation will be from customer satisfaction to service quality.

FIGURE 1

INTERACTION OF CONTINGENCY FACTORS FOR PREDICTING DIRECTIONALITY BETWEEN CUSTOMER SATISFACTION (CS) AND SERVICE QUALITY (SQ)

If the essential aspects of a service and service enhancers are both absent, the absence of the essential aspects will be noticed and this will be resented making customers feel bad about the service. This bad feeling will then influence cognitions about the service. Thus, the directionality of evaluation here will also be from customer satisfaction to service quality. Finally, if the essential aspects of a service are absent and service enhancers are present, both aspects will be noticed. While the customer may appreciate the presence of the service enhancers s/he will resent the absence of the essential factors. The reaction is therefore likely to be emotional and the causal sequence will be from customer satisfaction to service quality. Whether the overall experience is perceived as good or bad will depend on the service recovery process used to placate the customer about the missing essential aspects.

The last example of interactions (see Figure 1C) depicts three possible levels for each factor. A customer may be in a very good mood (e.g., elated), in a somewhat neutral mood, or in a very bad mood (e.g., upset). The service experience may be outside the zone of indifference (and below it) implying a really bad service experience, within the zone of indifference implying an acceptable service experience, or outside the zone of indifference (and above it) implying a really good service experience. The customer in the somewhat neutral mood who experiences a service within the zone of indifference is not likely to experience an emotional reaction. Instead the evaluation is likely to be cognitive and the directionality will be from service quality to customer satisfaction. If this customer's experience falls outside the zone of indifference, then given that expectations are not met (or are exceeded), it is possible that an emotional reaction will occur but this may be tempered by the customer's "normal" mood. Thus, we could expect a possible directionality from customer satisfaction to service quality or an overlap of the two constructs.

If the customer is in a very good or very bad mood, even if the experience is within the zone of indifference, the customer's mood is likely to temper the experience and evaluation will be mainly emotional. Thus, the causal sequence will be from customer satisfaction to service quality. It is possible, however, that the customer is able to prevent his/her mood from influencing the evaluation of the service recognizing that it was after all what s/he expected. In this case, there may be an overlap between service quality and customer satisfaction. If the customer in a very good (bad) mood experiences a service outside the zone of indifference and above (below) it, then the strong emotional reaction will be compounded by the customer's mood. The directionality should be from customer satisfaction to service quality and should strongly influence future behavior.

Finally, if the customer in a very good (bad) mood experiences a service outside the zone of indifference and below (above) it, there is a conflict between the initial feeling that the customer was experiencing and that caused by the service experience. In these situations, it is predicted that one (or more) of three things could happen. The customer in the very good mood will face the bad service philosophically and rather than get perturbed will evaluate it at a cognitive level, thus indicating a sequence from service quality to customer satisfaction. Similarly, the customer in the very bad mood will appreciate the excellent service, perhaps not enough to be jubilant, but still recognizing that it is more than s/he expected, and will evaluate it at a cognitive level, thus indicating a sequence from service quality to customer satisfaction. Or, in either situation, the customer may form both evaluations simultaneously resulting in an overlap of the two constructs. Alternatively, or in conjunction with either of these sequences of evaluation, the customer's mood may change from very good (bad) to somewhat neutral and perhaps even a little "down (up)" due to the service experience.

PREDICTING THE IMPACT ON CUSTOMER BEHAVIOR

As researchers we are interested in how the evaluation of service quality and customer satisfaction as well as the causal sequence of their formation influence subsequent customer intentions for repurchase. Under what conditions are repurchase intentions and loyalty strongest? When there is overlap between customer satisfaction and service quality in a given situation, it implies that the situation evoked little or no emotion (see Figure 2A). In this case, both evaluations are cognitive and while they could lead to future intentions, there is likely to be little commitment. When the evaluation of service quality takes place prior to the evaluation of customer satisfaction, and the customer then forms intentions for repurchase, the order of evaluation is cognitive, then affective, then conative, as predicted in traditional attitudinal models (see Figure 2B). Given that the evaluation is based on cognitions and then fortified by the affect that follows, the intentions should be well formed. Yet, the customer may be open to competitors' services if s/he has not yet tried them.

Lastly, when the evaluation of service quality takes place after the evaluation of customer satisfaction, this implies that there has been a strong emotional reaction to the service experience. It is commonly accepted that emotion is a much stronger predictor of behavior (Izard 1977; Westbrook 1987) especially when one is relatively new to the experience and still involved in the process of evaluation (Allen, Machleit, and Kleine 1992). Moreover, cognitions about service quality will be influenced by the emotion evoked during the situation and will be biased in one direction or another. The evaluation may not be as rational as in the earlier cases, but the impact on intentions should be much stronger including both direct and indirect effects (see Figure 2C). The models in Figure 2 extend Cohen and Areni's (1991) distinction between attitude as evaluative judgment and affect as a valenced feeling state by proposing three different possibilities in the sequence of evaluation formation and its impact on behavioral intentions.

From the provider's perspective, there are six possible outcomes for a service experience as presented in Figure 3. If there is an overlap between evaluations of satisfaction and quality and the customer receives a good (poor) service, this could be considered a good (poor) outcome. However, even in the case of the good outcome, there is much room for improvement because there is little commitment toward future behavior. If the causal sequence is from service quality to customer satisfaction and the customer has received good service, this should be considered a better outcome compared to the overlap situation (but not ideal). The customer is highly likely to come back, but s/he may try the competitors' services as well (Schlossberg 1993). The reason is that everything may have been done right, but the provider did not go the extra length to strike an emotional chord to win customer loyalty. For the same sequence, if the customer has received poor service, this should be considered a worse outcome compared to the overlap situation (but not terrible). The customer is highly likely not to come back, but then again, s/he may decide to give the firm a second chance given that this was a single occasion.

When the causal sequence for the evaluation is from customer satisfaction to service quality, and the customer has received good service, this is the best possible outcome for the firm. The customer will be delighted and is likely to be a customer for life (Schlossberg 1993). Having evaluated the experience on a highly emotional level, the customer will then translate this emotion to cognitions to become convinced to keep coming back. The worst outcome for the firm is if the customer receives poor service and evaluates the service emotionally. This customer is likely to be upset or disgusted, will form lasting cognitions based on this emotion, and will be lost forever to the firm. This customer is also likely to engage heavily in negative word-of-mouth.

FIGURE 2

IMPACT OF CAUSALITY ON INTENTIONS

FIGURE 3

POSSIBLE OUTCOMES FOR SERVICE EXPERIENCES

DISCUSSION

The literature on service quality and customer satisfaction does not appear to be in agreement regarding the causal order between these two constructs. Each researcher has his/her own conceptualization of these constructs and the directionality follows from this. At the same time, researchers realize that a causal order predicted entirely by suggesting that one construct is transactional and the other is global is somewhat arbitrary (e.g., Bitner and Hubbert 1994; Dabholkar 1993; Iacobucci, Grayson, and Ostrom 1994; Parasuraman, Zeithaml, and Berry 1994; Spreng and Singh 1993). Both service quality and customer satisfaction evaluations can occur at the transactional and the global level.

The contingency framework proposed in this paper attempts to show that different conceptualizations of customer satisfaction and service quality are indeed possible but under different contingencies. Researchers who predict that customer satisfaction leads to service quality are simply thinking about service situations where there is an element of surprise which enhances the emotion evoked by the encounter and colors subsequent cognitions about the experience (e.g., Bitner 1990; Oliver 1981; Parasuraman, Zeithaml, and Berry 1988). This view emphasizes the equivalence of satisfaction with experiential evaluation and of service quality with cognitions formed after the experience. On the other hand, those who predict that service quality leads to customer satisfaction are envisioning an attitudinal approach where cognitive evaluation precedes the formation of the associated affect. An example of this approach is the research on evaluations of telephone service (Bolton and Drew 1992; Drew and Bolton 1991) where cognitions appear to be formed first regarding the quality of the service and where the customer then decides whether s/he is happy with the service. In addition, the proposed framework outlines the various situations under which these different conceptualizations are valid.

Having presented a contingency framework to show that the causal link between customer satisfaction and service quality varies according to the service situation and that the direction of the causal link influences future customer behavior, an agenda is suggested for future research. The contingencies presented in Table 1 (as well as other relevant contingencies) could be used to set up experiments where the effect of these factors or their interactions on the link between customer satisfaction and service quality (see Figure 1) can be investigated. Furthermore, the influence of this directionality under various contingencies on future customer behavior can also be investigated. However, such studies must deal with the issue of truly measuring emotion rather than measuring cognitions about emotion (see Allen, Machleit, and Kleine 1992; Oliver 1989). The use of critical incidents (see Bitner 1990; Bitner and Hubbert 1994) is another useful technique to determine the sequence between customer satisfaction and service quality. Yet another technique would be to monitor actual service delivery and/or the recovery process to determine emotional content and causal sequence and to decide which sequence leads to the strongest intentions regarding future behavior.

The framework presented in this paper also has several implications for practitioners. First, service firms should understand the three models presented for service evaluation (Figure 2), the six possible outcomes of service from the provider's perspective (Figure 3), and all the different contingencies (Table 1, Figure 1) where these evaluations and outcomes can occur. The firms should then train employees to provide the basic minimum service. For example, they can ensure that essential aspects of the service are present and that the service provided is within the zone of indifference for most customers. Firms should then motivate employees to strive for the best possible service delivery. For example, employees should try to provide service enhancers and strive to provide service above the zone of indifference. However, providing service at this level involves extra effort and cost, and is not always possible. Yet, if the customer is delighted, s/he is likely to become a customer for life. At the other extreme, employees must be trained to avoid the worst situation. If the customer is upset, s/he will be lost to the firm and may take other customers along through negative word-of-mouth. To avoid such situations, employees must take care of little problems and prevent a bad situation from getting worse. They must learn to judge situations especially for services with high emotional possibilities. Also, some customers are more emotional than others and employees must learn how to handle them. For services with little emotional possibility, the best approach is to offer efficient, effective services that will receive high cognitive evaluation.

REFERENCES

Allen, Chris T., Karen A. Machleit, and Susan Schultz Kline (1992), "A Comparison of Attitudes and Emotions as Predictors of Behavior at Diverse Levels of Behavioral Experience," Journal of Consumer Research, 18 (March), 493-504.

Anderson, Eugene W. and Mary W. Sullivan (1993), "The Antecedents and Consequences of Customer satisfaction for Firms," Marketing Science, 12 (Spring), 125-143.

Bitner, Mary Jo (1990), "Evaluating Service Encounters: The Effects of Physical Surroundings and Employee Responses," Journal of Marketing, 54 (April), 69-82.

Bitner, Mary Jo and Amy R. Hubbert (1994), "Encounter Satisfaction versus Overall Satisfaction versus Quality: The Customer's Voice," in Service Quality: New Directions in Theory and Practice, Roland T. Rust and Richard L. Oliver, eds., SAGE Publications, 72-94.

Bloemer, Jose M. M. and Theo B. C. Poiesz (1989), "The Illusion of Consumer Satisfaction," Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 2, 43-48.

Bolton, Ruth N. and James H. Drew (1991), "A Multistage Model of Customers' Assessments Service Quality and Value," Journal of Consumer Research, 17 (March), 375-384.

Cacioppo, John T. and Richard E. Petty (1984), "The Need for Cognition: Relationship to Attitudinal Processes," in Social Perception in Clinical and Counseling Psychology, Vol. 2, R. P. McGlynn et al., eds., Lubbock: Texas Tech University Press, 113-140.

Cadotte, Ernest R. and Normand Turgeon (1988), "Dissatisfiers and Satisfiers: Suggestions from Consumer Complaints and Compliments," Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 1, 74-79.

Cohen, Joel B. and Charles S. Areni (1991), "Affect and Consumer Behavior," in Handbook of Consumer Behavior, Thomas S. Robertson and Harold J. Kassarjian, eds., Englewood Cliffs, NJ: Prentice-Hall, 189-240.

Cooper, Ann R., Cooper, M. Bixby, and Dale F. Duhan (1989), "Measurement Instrument Development Using Two Competing Concepts of Customer Satisfaction," Journal of Consumer Satisfaction, Dissatisfaction Complaining Behavior, 2, 28-35.

Cronin, Joseph J., Jr. and Steven A. Taylor (1992), "Measuring Service Quality: A Reexamination and Extension," Journal of Marketing, 56 (July), 55-68.

Dabholkar, Pratibha A. (1993), "Customer Satisfaction and Service Quality: Two Constructs or One?" in Enhancing Knowledge Development in Marketing, David W. Cravens and Peter Dickson, eds. Chicago: American Marketing Association, 4, 10-18.

Drew, James H. and Ruth N. Bolton (1991), "The Structure of Customer Satisfaction: Effects of Survey Measurement," Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 4, 21-31.

Erevelles, Sunil and Clark Leavitt (1992), "A Comparison of Current Models of Consumer Satisfaction/Dissatisfaction," Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 5, 104-114.

Gardner, Meryl P. (1985), "Mood States and Consumer Behavior: A Critical Review," Journal of Consumer Research, 12 (December), 281-300.

Iacobucci, Dawn, Kent Grayson, and Amy Ostrum (1994), "The Calculus of Service Quality and Customer Satisfaction: Theoretical and Empirical Differentiation and Integration," in Advances in Services Marketing and Management, Teresa A. Swartz, David A. Bowen, and Stephen W. Brown, eds., Volume 3, JAI Press.

Larsen, Randy J. and Ed Diener (1986), "Affect Intensity as an Individual Difference Characteristic: A Review," Journal of Research and Personality, 21, 1-39.

Izard, Carroll E. (1977), Human Emotions, New York: Plenum.

Mano, Haim and Richard L. Oliver (1993), "Assessing the Dimensionality and Structure of the Consumption Experience: Evaluation, Feeling, Satisfaction," Journal of Consumer Research, 20 (December), 451-466.

Oliver, Richard L. (1993), "A Conceptual Model of Service Quality and Service Satisfaction: Compatible Goals, Different Concepts," in Advances in Services Marketing and Management, Teresa A. Swartz, David A. Bowen, and Stephen W. Brown, eds., Volume 2, JAI Press, 65-85.

Oliver, Richard L. (1989), "Processing of the Satisfaction Response in Consumption: A Suggested Framework and Research Propositions," Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 2, 1-16.

Oliver, Richard L. (1981), "Measurement and Evaluation of Satisfaction Processes in Retail Settings," Journal of Retailing, 57 (Fall), 25-48.

Ortony, Andrew, Gerald L. Clore, and Allan Collins (1988), The Cognitive Structure of Emotions, New York: Cambridge University Press.

Parasuraman, A., Valarie A. Zeithaml, and Leonard L. Berry (1994),"Reassessment of Expectations as a Comparison Standard in Measuring Service Quality: Implications for Further Research," Journal of Marketing, 58 (January), 111-124.

Parasuraman, A., Valarie A. Zeithaml, and Leonard L. Berry (1988), "SERVQUAL: A Multi-Item Scale for Measuring Consumer Perceptions of Service Quality," Journal of Retailing, 64 (1), 12-40.

Reichheld, Frederick F. (1993), "Loyalty-Based Management," Harvard Business Review, 71 (2), 64-73.

Schlossberg, H. (1993), "Dawning of the Era of Emotion," Marketing News, 27 (4), 1.

Spreng, Richard A. and Cornelia Droge (1994), "A Comprehensive Model of the Effects of Expectations on Consumer Satisfaction, presented at the Society for Consumer Psychology Conference, St. Petersburg, FL, forthcoming.

Spreng, Richard A. and A. K. Singh (1993), "An Empirical Assessment of the SERVQUAL Scale and the Relationship between Service Quality and Satisfaction," in Enhancing Knowledge Development in Marketing, David W. Cravens and Peter Dickson, eds., Chicago: American Marketing Association. vol. 4, 1-6.

Westbrook, Robert A. (1987), "Product/Consumption-Based Affective Responses and Postpurchase Processes," Journal of Marketing Research, 24 (August), 258-270.

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

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

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