The Effect of Motivation to Process on Consumers' Satisfaction Reactions

Barry J. Babin, The University of Southern Mississippi
Mitch Griffin, Bradley University
Laurie Babin, The University of Southern Mississippi
ABSTRACT - This paper investigates potential moderating effects of the relationship between important explanatory variables and consumer satisfaction. Recent developments in assimilation-contrast theory suggest that involvement may change the satisfaction judgment process. Specifically, as involvement increases so should the likelihood of contrast, resulting in relatively more extreme satisfaction scores. Consistent with this reasoning, experimental results reported here support the moderating ability of involvement. Conversely, the potential priming effect of mood on reference standards is evidenced only under relatively low involvement.
[ to cite ]:
Barry J. Babin, Mitch Griffin, and Laurie Babin (1994) ,"The Effect of Motivation to Process on Consumers' Satisfaction Reactions", in NA - Advances in Consumer Research Volume 21, eds. Chris T. Allen and Deborah Roedder John, Provo, UT : Association for Consumer Research, Pages: 406-411.

Advances in Consumer Research Volume 21, 1994      Pages 406-411

THE EFFECT OF MOTIVATION TO PROCESS ON CONSUMERS' SATISFACTION REACTIONS

Barry J. Babin, The University of Southern Mississippi

Mitch Griffin, Bradley University

Laurie Babin, The University of Southern Mississippi

ABSTRACT -

This paper investigates potential moderating effects of the relationship between important explanatory variables and consumer satisfaction. Recent developments in assimilation-contrast theory suggest that involvement may change the satisfaction judgment process. Specifically, as involvement increases so should the likelihood of contrast, resulting in relatively more extreme satisfaction scores. Consistent with this reasoning, experimental results reported here support the moderating ability of involvement. Conversely, the potential priming effect of mood on reference standards is evidenced only under relatively low involvement.

INTRODUCTION

The disconfirmation paradigm provides the most popular theoretical basis for studying consumers' postconsumption reactions (Woodruff, Cadotte, and Jenkins 1983; Woodruff et al. 1991). Clearly, consumer disconfirmation is well established as an important causal agent of consumer satisfaction (Oliver 1980; Swan and Trawick 1981; Churchill and Surprenant 1982). Other variables, such as perceived performance, a priori expectations, and prior attitude, have also been considered as predictors of satisfaction with varying degrees of success (e.g., Westbrook 1980; Churchill and Surprenant 1982; Tse and Wilton 1988). The process with which these variables affect satisfaction is largely consistent with the disconfirmation paradigm's theoretical rootsCassimilation-contrast theory (LaTour and Peat 1979). For example, a consumer's subjective evaluation of performance requires a comparative process using some internalized standard as a basis for assimilation or contrast.

Despite the attention the satisfaction construct has received (see Perkins 1991), the number of variables that have been found to relate to consumer satisfaction are few when compared to other key consumer behavior variables (e.g., brand choice). In light of this, some authors have argued for a need to focus more on the invariance of the relationship between antecedent and consequence variables (Westbrook 1987; Westbrook and Oliver 1991). These studies espouse a more molecular position by examining changes in the evaluative process across consumption contexts (Oliver and DeSarbo 1988).

This paper's purpose is to report results of an investigation of the impact of consumers' processing motivations on satisfaction judgments. More specifically, this study examines experimentally the effect of differences in cognitive effort, corresponding to differences in motivational levels, on the satisfaction judgment process. The study's results add to our understanding of consumer satisfaction and provide evidence consistent with previous reasoning that questions the assumption of a linear, monotonic relationship between disconfirmation and satisfaction (Woodruff et al. 1983).

CONCEPTUAL BACKGROUND

Assimilation-Contrast Theory

Attempts at explaining postconsumption reactions using assimilation-contrast logic predate its maturation into the "disconfirmation model" (Anderson and Hair 1972). Using assimilation-contrast rationale, outcomes of consumption experiences are compared to internalized standards (e.g., expectations, performance norms, etc.) that serve as anchors for subsequent evaluations (LaTour and Peat 1979). Two results are possible.

Assimilation occurs if a consumption experience is perceived as consistent with internalized standards. The effect of this process on satisfaction ranges from negligible to modest as long as any discrepancy between a standard and an outcome is small. Thus, a "zone of indifference" results where a consumption outcome is perceived as equivalent to the norm, and the satisfaction response function is relatively nonresponsive to disconfirmation or performance perceptions (Woodruff et al. 1983; Woodruff et al. 1991; Zeithaml, Berry, and Parasuraman 1993). These zones of indifference are consistent with assimilation-contrast based findings in the pricing literature showing zones where consumer preference is nonresponsive to changes in price discounts and promotions (Kalwani and Yim 1992).

Alternatively, contrast results if consumer perceptions of the outcome differ notably from internalized standards. The result is substantially higher or lower satisfaction levels (depending on the direction of contrast) than would be observed under assimilation and a more responsive satisfaction function. Under contrast, changes in perceived performance relative to expectations (disconfirmation) will produce changes in satisfaction that are not realized under assimilation.

The Effect of Greater Effort

Due to the pervasiveness of assimilation-contrast explanations of social judgments, a significant amount of research has investigated contextual factors that enhance the likelihood of either reaction (Manis, Nelson, and Shedler 1988; Wedell, Parducci, and Geiselman 1987). One line of research suggests that enhanced cognitive effort, such as that associated with increased involvement or motivation, makes assimilation less likely. Martin (1986) reports a series of studies showing an increased likelihood of contrast when a subject's motivation to form an impression is disrupted. The result is explained in terms of a "set/reset" model (Martin 1986; Martin, Seta, and Crelia 1990). Nondisrupted subjects are better able to "partial out" a convenient standard of judgment and "reset" their frame of reference to provide their "genuine" reaction to a target.

An important implication of the set/reset model is that contrast is more cognitively complex than is assimilation. When one is unable or unwilling to expend a great deal of cognitive effort, he/she is more likely to use the less effortful alternative (assimilation) (Gilbert, Pelham, and Krull 1988). As illustration, one experiment required subjects to form an impression of an ambiguously described person (Martin et al. 1990, experiment 1). Subjects were primed with either a positive or negative concept and either distracted from completing the impression task or allowed to finish it with no distractions. Results showed that subjects in the distraction condition were more likely to rate their impression of the person as positive (negative) following a positive (negative) prime than those not distracted. That is, the supplied reference (prime) had a greater impact when subjects' cognitive effort was attenuated. A second experiment manipulating willingness to expend cognitive effort in place of ability duplicated these results (Martin et al. 1990, experiment 2). Specifically, subjects were more likely to assimilate (give an impression consistent with a prime) when motivation to form the impression was low. Paralleling the earlier result, reliance on the prime was more likely under low involvement conditions despite an individual's characteristics. Thus, a target's actual characteristics appear more predictive of impressions when the ability and/or willingness of subjects to expend cognitive effort is comparatively high. For a more detailed discussion of assimilation and contrast mechanisms compare Herr (1989) and Martin et al. (1990).

Implications for the Satisfaction Judgment

We propose that factors that influence the assimilation-contrast process will also influence satisfaction judgments. This is consistent with conceptual evidence suggesting that the size of "zones of indifference" in satisfaction response functions, where assimilation is a likely result, may vary situationally (Zeithaml et al. 1993). One might expect that under conditions of high (low) involvement the zone of indifference would be relatively small (large) given the associated increase (decrease) in willingness to process information (Petty, Cacioppo, and Schumann 1983).

Thus, the chances of contrast are enhanced as involvement increases and reported satisfaction is expected to be more responsive to differences in outcomes. This rationale is congruous with previous research investigating generalized negativity which found consumer appeal for automobiles was affected by discrepancies from expectations only under conditions of high involvement (Oliver 1976).

P1: Consumer satisfaction is more (less) responsive to disconfirmation under conditions of relatively high (low) involvement.

Priming effects may also influence the satisfaction judgment. The studies cited above (i.e., Martin 1986; Martin et al. 1990) show how simple affective primes can influence assimilation-contrast under certain conditions. These findings are also consistent with elaboration likelihood theory (Cacioppo and Petty 1989). It suggests that primes can influence salient information in a prime-congruent manner under conditions of low ability or unwillingness to process information. Essentially, such primes serve as a cognitive economizer. Thus even if the prime is irrelevant, it can affect the evaluative process.

A prime that may occur naturally in situations involving satisfaction judgments is a consumer's concurrent mood. While mood has been suggested as a nuisance factor sometimes related to consumer satisfaction (Peterson and Wilson 1992), it may be that consumer mood is serving as a prime for subsequent thoughts. Thus the relationship is something more than a nuisance. Peripheral cues like these would be most influential under low motivation and/or low ability to process issue-relevant information. Under these conditions, assimilation may lead to evaluations consistent with the prime (mood). Given sufficient motivation however, consumers are more likely to exert greater cognitive effort diminishing the effect of these primes.

P2: Mood is more (less) highly related to satisfaction under conditions of low (high) involvement.

RESEARCH METHODS

Study Description

An experiment manipulating product category involvement (2 levels) was conducted to test the propositions developed above. This manipulation was accomplished by assigning subjects randomly to one of two involvement conditions. Approximately half of all subjects were assigned to evaluate their last experience at a restaurant and half their last experience at a supermarket. Based on prior discussions with consumers representative of the subjects, the restaurant was assumed to be relatively more involving. This form of manipulation was considered desirable because both restaurants and supermarkets evoke product category involvement magnitudes quite representative of common, everyday consumption experiences. At the same time, it was expected that this manipulation would effectively result in significant differences in involvement.

Specific steps were taken to help insure a wide range of satisfaction scores and avoid ceiling effects commonly inhibiting consumer satisfaction measures (Peterson and Wilson 1992). To accomplish this, subjects were assigned a stimulus to rate based on a favorable or unfavorable image condition rather than allowing them to rate a stimulus of their choice. This approach was used only to avoid a severely skewed satisfaction distribution and not to test hypotheses regarding image. Thus, pretesting was conducted to identify restaurants and supermarkets that the subjects would be familiar with, but that would vary in image. Counterbalanced with the involvement manipulation, half of the subjects rated a relatively favorable stimulus and half a relatively unfavorable stimulus.

Subjects in the experiment were 87 undergraduate students residing in a small university town. Prior to beginning each session, subjects were given the opportunity to leave the classroom rather than participate (no subjects chose to leave). The survey booklet cover explained that the research was being conducted to rate consumer reactions to local retail establishments. Subjects first filled out a 4-item mood scale (e.g., "Currently, I am in a good mood" using a 5-point Likert Scale) (Peterson and Sauber 1983). Afterwards, index cards were passed out containing the name of either a favorable or unfavorable restaurant or supermarket. The card instructed subjects to raise their hand if they were either unfamiliar with the retailer on the card or had not been there in the past six months. This procedure resulted in reassignment of four subjects.

Subjects were then instructed to turn the page and respond to the remainder of the survey at their own pace. First, subjects were asked to state how long it had been since they last visited the assigned place and then to describe their last experience there in writing. Next, subjects recorded their reactions to the consumption experience. Disconfirmation was assessed using a 5-point Likert statement (e.g., "The restaurant turned out better than I expected.") (Oliver 1980). In addition, a graphical depiction of disconfirmation was obtained by having subjects indicate on a line how far above or below expectations the experience was. Distances were measured to the nearest tenth of an inch. Five items assessed consumer satisfaction. These items were patterned after previous efforts (Westbrook 1980; Oliver 1980; Holbrook and Batra 1987; Babin, Griffin, and Darden 1993) and contained items using a number of different scaling approaches (e.g., "I was satisfied with my experience at the restaurant," 100-point rating scales, etc.) to avoid common problems associated with measuring satisfaction. Finally, a manipulation check for product category involvement (e.g., "I think restaurant decisions are quite involving" on a 5-point Likert Scale) was administered, and subjects were given a chance to guess the purpose of the study. No subjects were considered either contingency or demand aware based on this analysis. Subjects were then debriefed and dismissed.

Measurement Results

The central dependent measure in this study is consumer satisfaction. Thus, a thorough analysis of the measurement quality of this variable was undertaken. The five items discussed above displayed a high degree of consistency as indicated by coefficient a (.91). In addition, a maximum likelihood factor analysis was conducted providing evidence of unidimensionality. The c2 goodness of fit statistic with 5 degrees of freedom is 3.21 (p > .668), suggesting a good fit to the data. The Goodness of Fit and Comparative Fit indices are .98 and .99, respectively, confirming this result (Bentler 1990). Based on these analyses, the five items were normalized to a constant metric (100 points) and averaged to provide a satisfaction score for each respondent.

TABLE 1

EFFECT SIZE ESTIMATES

Three levels of disconfirmation of expectancies were created using the Likert item described above. Those respondents indicating disagreement (agreement) with the idea that the experience was better than expected were placed in negative (positive) disconfirmation group, while those giving a neutral response were placed in a third (confirmation) group. This classification was validated by a high correlation with the graphical disconfirmation item (r = .64).

The four item mood scale displayed high internal consistency (a = .87), replicating previous results (Peterson and Sauber 1983). Thus those items were summed to obtain a mood score for each individual. In addition, the multiple item measure of involvement displayed an acceptable coefficient a (.76) and was summed to form the appropriate measure.

Manipulation Check

An ANOVA model was conducted to assess the effectiveness of the involvement manipulation. Results show that the involvement manipulation successfully altered involvement scores (F = 6.25; p < .01). The means for each level confirm the direction of the manipulation (11.03 for restaurants and 10.00 for supermarkets with a scale range of 3 to 15). Additional analyses revealed no differences in the frequency of visits or familiarity of the restaurants and supermarkets used in the study.

RESULTS

Because mood is a continuous variable, ANCOVA was used to isolate the effects needed to test both research hypotheses. In the model, satisfaction (DV) is predicted by the two interactions of primary interest (involvement x disconfirmation and mood x involvement) as well as both main effects and the remaining interactions. The main effects and remaining interaction are not relevant to the research hypotheses but are included as control variables. The overall model predicts a significant proportion of the variance in subject satisfaction scores (F = 13.4; p < .0001).

Proposition 1

If the likelihood of contrast is greater under conditions of relatively high involvement, the effect would be captured by the interaction between subject disconfirmation and involvement. As can be seen in Table 1, this interaction significantly predicts subject satisfaction while controlling for all other variables (F = 3.27; p<.05). Further, the mean satisfaction scores by involvement and disconfirmation levels (Figure 1) show that this effect is in the hypothesized direction. Subjects in the high involvement condition displayed more extreme satisfaction scores. It can be inferred that high involvement subjects were less likely to assimilate a performance based on their more extreme reactions to outcomes both above and below expectations. To summarize, involvement appears to satisfy conditions for moderation of the relationship between disconfirmation and satisfaction (Baron and Kenny 1986).

Proposition 2

The second proposition suggests that mood can influence consumer satisfaction under conditions of relatively low involvement. This effect is supported by the significant mood by involvement interaction (F = 7.11; p < .05). Baron and Kenny (1986) recommend separate regression models to analyze the moderating role of a continuous variable. Thus, separate regressions were run for the low and high involvement conditions. Figure 2 depicts this effect graphically. While mood has little impact on satisfaction for subjects in the high involvement (restaurant) condition (b = -.08; B = -.67; p > .10), it significantly affects satisfaction scores for subjects in the low involvement (supermarket) condition (b = .38; B = 2.90; p < .05). A comparison of the raw regression coefficients for each model supports involvement as a moderator of the relationship between mood and satisfaction (F = 3.10; p < .05). Consistent with P2, subject satisfaction scores were unaffected by their reported mood in the high involvement condition, but satisfaction was positively associated with mood in the low involvement condition.

Although not central to tests of the propositions, neither involvement nor mood produced a significant main effect, while disconfirmation did, as supported by numerous previous efforts (Tse and Wilton 1988). The three-way interaction was also significant. Although space limits a detailed discussion, this result, as well as results of contrasts not reported here, suggest that the mood-involvement interaction has varying effects across disconfirmation levels.

DISCUSSION

These results demonstrate that the disconfirmation-satisfaction relationship is not invariant to changes in consumers' processing motivation. Specifically, it appears that as involvement becomes relatively high (low) and consumers are willing (unwilling) to exert correspondingly greater cognitive effort in deliberating consumption outcomes (Petty et al. 1983), contrast (assimilation) becomes more likely. This finding is consistent with intuitive reasoning that consumers are more likely to have extreme reactions to consumption outcomes they perceive as relatively important.

In terms of behavioral outcomes of consumer satisfaction, this might translate into increased brand switching under high involvement at the same level of disconfirmation. Alternatively, consumers may be more likely to display brand loyalty as involvement decreases while holding disconfirmation constant. Other behavioral outcomes, such as word-of-mouth and complaining behavior, may also conform to this pattern. Thus, brands that perform superior to norms held by consumers may benefit from encouraging deliberation of the product outcome. Such encouragement may be facilitated by encouraging feedback on product performance using monetary or discount incentives. For example, a new product might be accompanied by a satisfaction survey that consumers could complete and return in exchange for a small monetary incentive. Brands that do not perform well compared to these reference points however, may benefit from satisfaction judgments more conducive to assimilation. This might be particularly important in product classes where the consumer can be expected to hold a favorable expectation (Woodruff et al. 1991).

FIGURE 1

THE EFFECT OF INVOLVEMENT AND DISCONFIRMATION ON SATISFACTION

FIGURE 2

THE EFFECT OF MOOD AND INVOLVEMENT ON SATISFACTION

The study also suggests that internal reference points besides expectations, even those that appear irrelevant to the satisfaction judgment, can influence consumer satisfaction. In this study, subject mood influenced reported satisfaction under low involvement only. It is also conceivable that beyond mood, any emotional traces evoked by recall of the stimulus concerned might influence the satisfaction judgment. If so, satisfaction of consumers under low involvement could be affected by emotions evoked during the communication, purchase, or usage situation so long as they are stored in conjunction with the stimulus.

Further research is needed on these issues. Based on the nature of relationships revealed in this study, it can be inferred that consumer zones of indifference or tolerance in consumption outcomes correspond to involvement levels. However, a direct test of this proposition awaits the development of measurement techniques capable of assessing the magnitude of these zones across consumers and consumption contexts. In addition, the involvement manipulation used here is assumed to have altered cognitive effort based on previous theory (Petty et al. 1983). However, alternative studies that control for product involvement, while manipulating cognitive effort directly, may dissect this effect even more precisely. The involvement manipulation here also could be criticized for encompassing a limited range. However, if a wider range were considered in future studies, the effects found here should be larger.

Additional research is needed to study mood and other factors potentially associated with satisfaction under low involvement. For example, does mood influence satisfaction judgments similarly under conditions of positive or negative affect? Another conceptual area that needs clarification is whether or not the results imply "true" shifts in consumer satisfaction, or does mood simply bias satisfaction measurement? The present experiment is unable to detect the true mechanism but leaves this interesting question for further study. Furthermore, other contextual factors may also prime this judgment. For instance, it is possible that the satisfaction outcome from consumption of one product might influence satisfaction with another consumed concurrently or in close proximity. For example, our satisfaction with one store in a mall may be influenced by our experience with a previously visited retailer.

In conclusion, this paper sought to investigate differences in the disconfirmation model of consumer satisfaction at different, yet comparable, levels of involvement. The results of the study suggest disconfirmation corresponds to satisfaction more strongly as motivation to process increases. This may help explain substantial differences in effect sizes for disconfirmation across studies of consumer satisfaction (cf., Swan and Trawick 1981; Oliver 1980; Tse and Wilton 1988). Further, disconfirmation models of consumer satisfaction may need to consider extraneous variables as potential moderating variables. This study showed how one previously thought nuisance variable, consumer mood, can influence consumer satisfaction at relatively low levels of involvement. More work is needed to reveal additional potential priming and framing effects.

REFERENCES

Anderson, Rolph E. and Joseph F. Hair, Jr. (1972), "Consumerism, Consumer Expectations, and perceived Product Performance," in Proceedings of the 3rd Annual Conference of the Association for Consumer Research, Provo, UT: Association for Consumer Research, 67-79.

Babin, Barry J., Mitch Griffin, and William R. Darden (1993), "An Empirical Comparison of Alterative Conceptualizations of Postconsumption Phenomena," Journal of Consumer Satisfaction, Dissatisfaction, and Complaining Behavior (in press).

Baron, Reuben M. and David A. Kenny (1986), "The Moderator-Mediator Variable Distinction in Social Psychological Research," Journal of Personality and Social Psychology, 51 (July), 1173-1182.

Bentler, P.M. (1990), "Comparative Fit Indices in Structural Models," Psychological Bulletin, 107 (March), 238-246.

Cacioppo, John T. and Richard E. Petty (1989), "The Elaboration Likelihood Model: The Role of Affect and Affect-Laden Information Processing in Persuasion," in Cognitive and Affective Responses to Advertising, Patricia Cafferata and Alice M. Tybout eds., Lexington, Mass.: Lexington Books.

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

Gilbert, D. T., B. W. Pelham, and D. S. Krull (1988), "On Cognitive Components of the Social Inference Process: When Person Perceivers Meet Persons Perceived," Journal of Personality and Social Psychology, 54 (October), 733-740.

Herr, Paul M. (1989), "Priming Price: Prior Knowledge and Context Effects," Journal of Consumer Research, 16 (June), 67-75.

Holbrook, Morris and Rajeev Batra (1987), "Assessing the Role of Emotions as Mediators of Consumer Responses to Advertising," Journal of Consumer Research, 14 (December), 404-20.

Kalwani, Manohar U. and Chi Kin Yim (1992), "Consumer Price and Promotion Expectations: An Experimental Study," Journal of Marketing Research, 29 (February), 90-100.

LaTour, Stephen A. and Nancy C. Peat (1979), "Conceptual and Methodological Issues in Consumer Satisfaction Research," in Advances in Consumer Research, Vol. 6, ed. William Wilkie, Ann Arbor, MI: Association for Consumer Research, 431-437.

Manis, M., T. E. Nelson, and J. Shedler (1988), "Stereotypes and Social Judgment: Extremity, Assimilation, and Contrast," Journal of Personality and Social Psychology, 55 (January), 28-36.

Martin, Leonard L. (1986), Categorization and Differentiation: A Set, Re-set, Comparison Analysis of the Effects of Context on Person Perception. New York: Springer-Verlag.

Martin, Leonard L., John J. Seta, and Rick A. Crelia (1990), "Assimilation and Contrast as a Function of People's Willingness and Ability to Expend Effort in Forming an Impression," Journal of Personality and Social Psychology, 59 (January), 27-37.

Oliver, Richard L. (1976), "Hedonic Reactions to the Disconfirmation of Product Performance Expectations: Some Moderating Conditions," Journal of Applied Psychology, 61, 246-250.

Oliver, Richard L. (1980), "A Cognitive Model of the Antecedents and Consequences of Satisfaction Decisions," Journal of Consumer Research, 27 (November), 460-468.

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

Perkins, Debra S. (1991), "A Consumer Satisfaction, Dissatisfaction and Complaining Behavior Bibliography: 1982-1990," Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, Vol. 4, 194-228.

Peterson, Robert A. and Matthew Sauber (1983), "A Mood Scale for Survey Research," in Proceedings of the American Marketing Association's Summer Educators Conference, Chicago: American Marketing Association.

Peterson, Robert A. and William R. Wilson (1992), "Measuring Customer Satisfaction: Fact and Artifact," Journal of the Academy of Marketing Science, 20 (Winter), 61-71.

Petty, Richard E., John T. Cacioppo, and David Schumann (1983), "Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement," Journal of Consumer Research, 10 (September), 135-146.

Swan, John E. and Frederick I. 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 Formation: An Extension," Journal of Marketing Research, 25 (May), 204-212.

Wedell, D. H., A. Parducci, and R. E. Geiselman (1987), "A Formal Analysis of Ratings of Physical Attractiveness: Successive Contrast and Simultaneous Assimilation," Journal of Experimental Social Psychology, 23, 230-249.

Westbrook, Robert A. (1980), "Intrapersonal Affective Influences on Consumer Satisfaction with Products," Journal of Consumer Research, 7 (June), 49-54.

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

Westbrook, Robert A. and Richard L. Oliver (1991), "The Dimensionality of Consumption Emotion Patterns and Consumer Satisfaction," Journal of Consumer Research, 18 (June), 84-91.

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.

Woodruff, Robert B., D. Scott Clemons, David W. Schumann, Sarah F. Gardial, and Mary Jane Burns (1991), "The Standards Issue in CS/D Research: A Historical Perspective," Journal of Consumer Satisfaction, Dissatisfaction and Complaint Behavior, 4, 103-109.

Zeithaml, Valarie A., Leonard L. Berry, and A. Parasuraman (1993), "The Nature and Determinants of Customer Expectations of Service," Journal of the Academy of Marketing Science, 21 (Winter), 1-12.

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