Conceptual Issues in the Structural Analysis of Consumption Emotion, Satisfaction, and Quality: Evidence in a Service Setting

Richard L. Oliver, Vanderbilt University
ABSTRACT - Recent research linking consumer satisfaction and affect has raised issues which require elaboration before models incorporating consumption affect will become useful. Among these are the dimensional structure of affect and the role these dimensions play alongside consumption cognition (e.g., expectancy disconfirmation, quality). Previous affect work using the Izard (1977) typology has been limited due to the heavy emphasis on negative affect, as opposed to positive affect and arousal. In this paper, the contributions of positive and negative affect are examined with reference to affect frameworks based on dimensions of pleasantness and arousal. In a related topic, the hypothesized duality of consumption affect and cognition, with the specific inclusion of quality judgments, will be discussed. Drawing on Oliver (1989), a structural representation of parallel affect and cognitive mechanisms affecting satisfaction is proposed. Both the dimensional and structural suggestions are tested with data from an adolescent health-care setting. Results show that the parallel representation is supported.
[ to cite ]:
Richard L. Oliver (1994) ,"Conceptual Issues in the Structural Analysis of Consumption Emotion, Satisfaction, and Quality: Evidence in a Service Setting", in NA - Advances in Consumer Research Volume 21, eds. Chris T. Allen and Deborah Roedder John, Provo, UT : Association for Consumer Research, Pages: 16-22.

Advances in Consumer Research Volume 21, 1994      Pages 16-22


Richard L. Oliver, Vanderbilt University


Recent research linking consumer satisfaction and affect has raised issues which require elaboration before models incorporating consumption affect will become useful. Among these are the dimensional structure of affect and the role these dimensions play alongside consumption cognition (e.g., expectancy disconfirmation, quality). Previous affect work using the Izard (1977) typology has been limited due to the heavy emphasis on negative affect, as opposed to positive affect and arousal. In this paper, the contributions of positive and negative affect are examined with reference to affect frameworks based on dimensions of pleasantness and arousal. In a related topic, the hypothesized duality of consumption affect and cognition, with the specific inclusion of quality judgments, will be discussed. Drawing on Oliver (1989), a structural representation of parallel affect and cognitive mechanisms affecting satisfaction is proposed. Both the dimensional and structural suggestions are tested with data from an adolescent health-care setting. Results show that the parallel representation is supported.


Affect as an essential component of purchase and usage is beginning to attract the attention of consumer researchers, particularly within the context of service satisfaction and postpurchase response. This is so because affect is now known to be a more primitive motivational response in living organisms and because it may interact with cognition as both an antecedent and consequent (Lewis, Sullivan, and Michalson 1984). Thus, the role of affect is intertwined with other cognitive postpurchase responses either because the affect state generated in consumption "recruits" certain cognitions (Chattopadhyay and Alba 1988) or because it is a natural outcome of cognitive processes such as attribution (e.g., Weiner, Russell, and Lerman 1979).

Recently, researchers have begun to entertain affective influences in consumer response. Beginning with work in advertising (e.g., Holbrook and Batra 1987), and continuing in the domain of products (e.g., Westbrook 1987), some strides have been made in understanding the role of affect in consumption. Westbrook and Oliver (1991), for example, find that affect co-exists with cognition (e.g., expectancy disconfirmation) in the formation of satisfaction judgments and that both make independent contributions to satisfaction. Other work (Mano and Oliver 1993; Oliver 1992, 1993a; Oliver and Westbrook 1993) elaborates on this phenomenon and illustrates the attribute-specific and cognitive evaluation bases for affect-augmentation (Oliver 1993a) in consumer response. This paper reviews the work to date in the area of product and service postpurchase judgment, identifies specific issues in the work so far, and provides a new test of affect in the context of service satisfaction.

In doing so, the role, if any, of consumption affect in quality judgments will be explored. The fields of quality and satisfaction are merging (Oliver 1993b), although causality issues remain to be resolved. Given the newly discovered importance of affect in satisfaction, a reasonable question is the extent to which quality judgments are similarly affected.

In the following discussion, the dependent variable under study is limited to that of satisfaction. Justification for this is the extreme importance of satisfaction as an antecedent to repurchase or repatronage intentions, as a motivator for word-of-mouth, and as a diagnostic measure for managerial action. After describing efforts to conceptualize affect in consumption generally, a test of a conceptual model is performed in the service sector. As noted, service quality and satisfaction are now generating a great deal of interest in the field of services. The present study attempts to add to that literature.

Affect in Product Satisfaction Contexts

Westbrook (1987) is credited with introducing affect to the existing stream of research on the cognitive basis of satisfaction (Oliver 1980). Using Izard's (1977) emotional typology, he was able to show that two affect constructs, consisting of summated positive and negative affect, correlated in the predicted directions with overall satisfaction over two product categories (automobiles and cable TVCa service). Moreover, these relations held up to the introduction of cognition (i.e., expectancy disconfirmation) in regressions explaining satisfaction.

Later, Westbrook and Oliver (1991) performed a dimensional analysis, also using the Izard (1977) scheme, and found that three affective dimensions could be justified as underlying explanations for the satisfaction judgment. Respectively, these were hostility (a constellation of negative affect), pleasant surprise (positive affect and surprise), and interest. Of note is the fact that they were also able to identify emotional segments of consumers such that each segment was characterized with distinctively different emotional profiles. In order of their positive affectivity toward the tested product (autos), the groups could be described as pleasantly surprised, happy/content, unemotional, unpleasantly surprised, and angry/upset.

Later, Oliver (1993a), again using Izard's (1977) framework, provided evidence that the negative affect dimension could be viewed as having three subdimensions based on the attributional agency. Respectively, these consisted of the externally attributed affects of anger, disgust, and contempt, the internally attributed affects of guilt and shame, and the situation-specific affects of fear and sadness. The two positive affects in the Izard scheme, interest and joy were found to be separately processed for a product used in Oliver's study (autos), but not for the service context (course instruction). Oliver was able to show that the positive and negative affect constellations partially mediated attribute experience. That is, attribute experiences appeared to be the underlying causes for the affects reported in consumption.

Mano and Oliver (1993) replicated this finding in the context of the nature of product evaluation. This study differed on a number of dimensions from those described previously. First, the authors used affect items from Mano's (1991) work, based on the PANAS scale of Watson, Clark, and Tellegen (1988), which describes affect on the two dimensions of positive-negative affect and arousal. This two-dimensional space is sectioned into eight octants (see Mano and Oliver 1993) which are represented by two or three indicators. Second, the product/service context was unrestricted as subjects were allowed to select purchases (including services) which satisfied either a high or low involvement situation, as determined by survey instructions. Third, the selected product was evaluated not on the basis of attributes but on whether utilitarian (i.e., functional) or hedonic evaluation was most pertinent, based on the work of Batra and Ahtola (1990). Results showed that positive and negative affect and arousal were key dimensions underlying affect, as theory predicted. When input into a causal modeling framework, arousal was found to be a function of higher hedonic evaluation and lower utilitarian evaluation and positive and negative affect were found to be a function of arousal. Positive affect was also a function of hedonic evaluation. Finally, satisfaction was a function of positive and negative affect in the respective directions, and utilitarian evaluation. Once again, satisfaction is shown to be a function of (aroused) affect.

Specific attribute influences as influencing satisfaction through affect are only recently coming under investigation. Early exploratory attempts have been made by Oliver (1992) and Oliver and Westbrook (1993). Oliver used an MDS approach to position overall satisfaction among attribute satisfactions. Results showed satisfaction to be centrally located among two dimensions of attribute space which the author referred to as static and dynamic. Oliver and Westbrook posed specific predictions about which attribute judgments would predict specific positive and negative affects. Their results were encouraging but not consistently significant. Of note is the fact that clusters of consumers were found roughly in accord with those found in Westbrook and Oliver (1991). Replicated clusters of delighted (pleasantly surprised), contented, unemotional, and angry consumers were found as before. However, two new cluster descriptions, described as tentative and guilty/ashamed, were discovered, thus adding further emotional content to the satisfaction response.

Taken together, these investigations point to a consistent pattern of affective response in the satisfaction judgment. Apparently, attribute experience, in addition to having a direct effect on satisfaction, also drives the affective response the consumer has toward the product. Moreover, affective response appears to be well-specified by positive and negative affect and arousal which combine into states which describe more complex affects such as delight. This research stream, however, is tentative and would benefit from further corroboration and extension. More specifically, the affect octant approach of Mano (1991) and Watson et al. (1988) requires further testing in consumer (service) contexts if its validity characteristics are to be known. Additionally, the specific attribute constellation commonly referred to as quality has not been framed within the emotion space. In fact, little is known about the psychological interpretation of quality. The next section speaks to that issue.

Quality and Affect

A number of current works propose a linkage between quality and satisfaction (Bitner 1990; Bolton and Drew 1991; Cronin and Taylor 1992; Swartz and Brown 1989; Zeithaml, Berry, and Parasuraman 1993). None, however, explicitly considers affect. In fact, in Swartz and Brown and Zeithaml et al., the "quality" gap between expectations and performance is viewed as surrogate satisfaction, which some have likened to affect or emotion (e.g., Hunt 1977).

In Bitner (1990) and Bolton and Drew (1991), satisfaction is viewed as a function of expectancy disconfirmation. Service quality, in contrast, is positioned as subsequent to satisfaction under the implicit assumption that quality judgments result from satisfying or dissatisfying service encounters. Cronin and Taylor (1992) indirectly tested this assumption across eight service companies. Of note is the finding that, of the two reciprocal paths (satisfaction ¦ quality and quality ¦ satisfaction), only the quality ¦ satisfaction path was significant.

No study, however, has investigated affect in the context of quality. Cadotte, Woodruff, and Jenkins (1987) tested a "feelings" satisfaction model where satisfaction was defined as a constellation of affective adjectives (e.g., warm/cold). Their model showed that service attributes (in a restaurant setting) impacted this feelings criterion only through the disconfirmation concept. Thus, it is not known how quality is affected, if at all, by affect.

Oliver (1993b) has suggested that quality is largely a performance concept. More to the point, quality relies on assessments of "performance excellence criteria" (Zeithaml et al. 1993). These will tend to be exemplary levels of performance characteristics, those on which the service will truly excel. As such, quality judgments are posited to be a direct function of performance and may, likewise, influence satisfaction directly.

A summary model of the concepts discussed above is shown in the Figure.


The preceding discussion suggests the following hypotheses:

H1: Affect in service consumption can be described by a dimensionality consisting of positive affect, negative affect, and arousal.

H2: Positive and negative affect are a direct function of arousal and a positive/negative function of perfor- mance, respectively.

H3: Quality is a direct function of (high) service performance.

H4: Satisfaction is a function of affect, quality, disconfirmation, and performance.



Parents of adolescent patients who had convalesced from various ailments in a short-term hospital recovery center over a seven month period were surveyed as to their satisfaction with their child's stay. The count of individual patients in this time period numbered 377, of which 40 were used in pretesting the list of performance attributes (to be discussed). Parents of the remaining 337 patients were sent surveys, of which 65 usable responses were returned in a one month "response window" allowed for the studyCrepresenting a response rate of 19.3%. The average respondent was female, a mother of the patient, had some college, and was somewhat upscale with a family income between $30-50,000.


For the purpose of this study, measures of performance, affect, quality, disconfirmation, and satisfaction were constructed as follows:

Performance. A list of performance features was generated through two procedures. First, the laddering technique (Reynolds and Gutman 1988) was administered to 20 randomly selected parents from the original list of respondents. This approach resulted in a list of 24 key elements of service delivery. Second, the critical incidence technique (Bitner, Booms, and Tetreault 1990) was administered to a second set of 20 respondents. This resulted in service delivery outcomes which were fully contained in the previous list, thus corroborating the earlier results. In collaboration with the study sponsors, the final list was condensed to 19 delivery features for use in the final study. This outcome list was scored by respondents on five-point bipolar adjective scales (e.g., The nurses would be: rude ... friendly). A summated index was created by aggregating over all outcomes. Note that factor analysis was not used to reduce the dimensionality of the features. Two highly correlated attributes, which would load on a single dimension, could have singular, additive impacts on quality or satisfaction. This additivity would be obscured if dimensions were used (Oliver 1993a).

The affects. A set of affects completing the octal dimensional affect solution of Watson et al. (1988) was obtained from their short inventory which measures all eight segments of affect space. Lately, Larsen and Diener (1992) have refined this scale, making it more social and less clinical. The items used in this study were selected from those of Watson et al. and Larsen and Diener so as to provide two items from each octant to represent each of the eight affect categories. All were measured on frequency scales. Respondents were asked to describe "during your experience with the care unit, how frequently you felt" each of the emotions on 5-point scales ranging from "never" (1) to "always" (5).

Quality. Based on work by Garvin (1984), Steenkamp (1990), and Zeithaml (1988), a six-item quality scale was constructed using bipolar adjectives addressing excellence, superiority, value, quality, high standards, and best/worse.

Disconfirmation. Disconfirmation was measured on a disconfirmation scale reported in Oliver (1980). It contains three "better-worse than expected" items sampling benefits, problems, and overall performance.

Satisfaction. Satisfaction was measured using five items from Oliver's Likert-type satisfaction scale used in Westbrook and Oliver (1981), Oliver and Swan (1989), Oliver and Bearden (1985), and other studies. In addition, this scale was augmented with five items reflecting satisfaction with the child's experience, specially constructed for this study. Because both parent and child interacted with the service delivery, parents were also expected to reflect satisfying and dissatisfying experiences through the eyes of their child, a form of reflected appraisal.




Dimensional Structure of Affect

Factor analysis of the sixteen affects revealed four factors having eigenvalues greater than unity and explaining 61% of the item variance. This solution, however, contained one variable which did not load at the recommended 0.5 cutoff (annoyed) and one which loaded singularly on the fourth factor (enthused). Thus, the three factor solution, representing fourteen affects, was used to describe the affect structure emerging from this data set. See Table 1.

The results show that the two primary dimensions of positive and negative affect were supported as was the additional dimension of arousal. This corroborates earlier studies of affect in consumption (Westbrook 1987; Westbrook and Oliver 1991; Oliver 1993a), based on the Izard (1977) scheme, and also suggests that the role of arousal has been somewhat neglected in consumer research. Based on these findings, scales of positive and negative affect and arousal were constructed by adding the respective items loading on the factors.

Table 2 shows the descriptive statistics, correlation matrix, and alpha reliabilities of the affect scales and other study variables. This pattern of results is best discussed in terms of the model coefficients, tested here with two-stage least squares (TSLS) due to the recursive nature of the conceptual structure (see the Figure). Table 3 shows the TSLS results of the base model, a submodel of quality as a function of the affects, and a model entertaining a direct effect from performance to satisfaction.





These results show that, in accord with Mano and Oliver (1993), both positive and negative affect are a function of the arousal inherent in the service experience. However, the results differ in the present study in that arousal is negatively related to negative affect, suggesting that the lack of arousal is a displeasing state. In this context, subjects (i.e., parents) apparently become aroused in anticipation of pleasureCthe rehabilitation of their children. Additionally, similar to the results in Oliver (1993a), both positive and negative affect are a function of performance in the predicted direction. Apparently, service consumers do take their affect cues from high and low performance observations.

As expected, quality was also a function of performance ratings. In a second analysis (not hypothesized), the two affects were added to the quality equation. Neither was significant indicating that affect has little relation to the more cognitive quality judgment.

Finally, all predictors of satisfaction were significant in the hypothesized direction when performance was excluded from the regression. Note that the affects contributed to the variance explained despite the fact that two cognitive determinants were also in the equation (disconfirmation and quality). When performance, a third cognitive variable, was added to the satisfaction equation, however, multicollinearity effects were severe. Neither performance, quality, nor positive affect remained significant and satisfaction could be described as a function only of disconfirmation and negative affect.


The goal of this study was to provide further evidence of the role and structure of affect in service consumption and to corroborate the combined "two-appraisal" cognitive-affective model of Oliver (1989). Based on prior work, specific hypotheses were proposed and tested. The following discussion focuses, in turn, on the affect structure found in the present study versus that of prior work, the validity of the two-appraisal model in predicting satisfaction, and the relation between quality and satisfaction. Study limitations and future directions round out the discussion.

The Structure of Affect

Prior work (Westbrook 1987; Westbrook and Oliver 1991; Mano and Oliver 1993; Oliver 1993a) has suggested two or three primary affective dimensions underlying consumption. Westbrook suggested that these may be limited to positive and negative affect whereas Westbrook and Oliver found that arousal may coexist with positive affect and that interest (a low arousal form of affect) may be a third dimension. Later, Oliver found both positive and negative affect dimensions in one sample of respondents and the potential for a third, interest, dimension in a second. In all of these studies, the Izard (1977) DES was used which has no pure arousal elements. In contrast, Mano and Oliver and the present study used the octal representations of Watson et al. (1988). Mano and Oliver found a three-dimensional solution consisting of aroused positive affectivity, negative affectivity, and low arousal; a two dimensional solution showed only the aroused positive and negative factors. The present study supports the separate positive and negative dimensions and, additionally, suggests that arousal may be viewed as a separate third dimension.



Thus, a convergence of findings is beginning to appear suggesting that positive and negative affectivity are the primary dimensions underlying affect in consumption. Moreover, the role of arousal appears to be study-specific. In some samples, it tends to align with positive affect as if to suggest that positive affect is the result (or cause) of such arousal. In others, arousal appears separately, as it did here, or is associated with low arousal states of inactivity or interest. Thus, the role of arousal may be a more interesting phenomenon in consumption, as a research question, than is the valence of affect.

A Two-Appraisal Satisfaction Model

Three studies now exist to provide partial support for a "two-appraisal" representation of the determinants of satisfaction. The two types of appraisal referred to under this interpretation are cognitive judgment, represented here by quality and disconfirmation, and the affect experienced from consumption, represented here by positive and negative affect. Prior to this study, Oliver (1993a) showed satisfaction to be a function of the affects, attribute satisfaction/dissatisfaction (as a proxy for performance), and disconfirmation, and Mano and Oliver (1993) showed satisfaction to be predicted by the affects and utilitarian appraisal (but not hedonic appraisalCwhich was closely related to affect). Now it is shown that satisfaction is a function of the affects and separate judgments of quality and disconfirmation. In all cases, the affective and cognitive variable sets made significant and independent contributions to satisfaction.

Taken together, the three studies suggest that Oliver's (1989) two-appraisal representation may be a reasonable approach to satisfaction formation. Apparently two mechanisms operate in tandem in consumers' minds, one involving the assessment of functional or comparative outcomes (what the product/service gives me) and one relating to how the product/service influences affect (how the product makes me feel). Oliver did allow for the possibility that cognition influences affect and, in fact, the Oliver (1993a) and Mano and Oliver (1993) findings display such effects. Unfortunately, the quality variable investigated here did not appear to be related to either positive or negative affect.

Quality, Satisfaction, and Affect

Introduction of a direct measure of quality into the satisfaction framework is a recent phenomenon. Previously, Bolton and Drew (1991) provided an indirect (performance-based) quality indicator within a satisfaction framework and showed influences of this quality surrogate on satisfaction. As far as this author is aware, the present study is the first to provide a scaled quality measure that was not performance-attribute based. This measure, which yielded a high alpha reliability of 0.90, was strongly related to satisfaction, although not as strongly as disconfirmation. Moreover, it was more strongly related to performance than any other study variable, indicating that it is a performance-related concept and derives from performance observations as suggested in Oliver (1993b). The data clearly show that it is not affect-based, thus distinguishing quality from satisfaction, which is affect-based according to the two-appraisal model. This finding adds support to the quality-influences-satisfaction camp (e.g., Cronin and Taylor 1992) which advocates satisfaction as a superordinate concept to quality.

Future Directions

A number of new research directions are suggested by these results. First, alternative affect schemes beyond those of Izard (1977) and Watson and his colleagues need to be tested to determine if the two and three dimensional interpretations of affect space in service consumption can withstand replication with other theories. Second, much more work on the independent contributions and interplay of affect and cognition in the satisfaction judgment needs to be performed. It appears that the separate contributions of affect and cognition are established; what is missing is evidence on their interaction, causal relationships, and joint dependence on causal agents in satisfaction formation. Thirdly, quality as integral to consumption experience and satisfaction requires greater conceptual and methodological effort if this critical variable is to be properly understood in context. Current writings, particularly in the popular press, assume that quality and satisfaction are isomorphic. It is this author's conclusion that quality is more cognitive and performance-oriented, while satisfaction is a complex summary consumption judgment. More will come of this controversy.


Perhaps the greatest limitation of the study beyond the usual caveats regarding static data, small sample restrictions, and measurement error is the possibility that the model as illustrated in the Figure and tested here is mis-specified. It is well-known that alternative model structures will give acceptable fits from the same data set. Of concern to the present study is the temporal ordering of the variables. For example, arousal may be a consequent of affect and not a determinant as suggested in Mano and Oliver (1993), the theoretical source used to frame the present model. Additionally, the observed multicollinearity between performance, quality, and positive affect which attenuated the regression coefficients when all three were resident in the same analysis, suggests that a submodel of effects is operating between the cognitive performance and quality variables and positive affect. Others are advised to thresh out alternative, reasonable interpretations of model structure for testing. This will require samples larger than that used here, especially if multiple indicators are to be employed. Nonetheless, the study findings presented here are offered as early evidence for emerging conceptualizations of affect, quality, and satisfaction in the service area.


Appreciation is expressed to Bob Neydon, Chris Opipari, Mark Stevens, and Terra Travis for their efforts in the design, data collection, and execution of the study. Richard Anton of the Vanderbilt University Medical Center provided valuable assistance in facilitating the sponsorship and conduct of the research. The author also thanks the Center for Services Marketing and the Dean's Fund for Faculty Research of the Owen Graduate School of Management of Vanderbilt University for providing partial support for this project.


Batra, Rajeev and Olli T. Ahtola (1990), "Measuring the Hedonic and Utilitarian Sources of Consumer Attitudes," Marketing Letters, 2 (April), 159-170.

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, Bernard H. Booms, and Mary Stanfield Tetreault (1990), "The Service Encounter: Diagnosing Favorable and Unfavorable Incidents," Journal of Marketing, 54 (January), 71-84.

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

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

Chattopadhyay, Amitava and Joseph W. Alba (1988), "The Situational Importance of Recall and Inference in Consumer Decision Making," Journal of Consumer Research, 15 (June), 1-12.

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

Garvin, David A. (1984), "What Does `Product Quality' Really Mean?," Sloan Management Review, 26 (Fall), 25-43.

Holbrook, Morris B. and Rajeev Batra (1987), "Assessing the Role of Emotions as Mediators of Consumer Responses to Advertising," Journal of Consumer Research, 14 (December), 404-421.

Hunt, H. Keith (1977), "CS/DCOverview and Future Research Directions," in Conceptualization and Measurement of Consumer Satisfaction and Dissatisfaction, H. Keith Hunt, ed., Cambridge, MA: Marketing Science Institute, 455-488.

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

Larsen, Randy J. and Edward Diener (1992), "Promises and Problems With the Circumplex Model of Emotion," in Review of Personality and Social Psychology, Vol. 13, Margaret S. Clark, ed. Newbury Park, CA: Sage, 25-59.

Lewis, Michael, Margaret Wolan Sullivan, and Linda Michalson (1984), "The Cognitive-Emotional Fugue," in Emotions, Cognition, and Behavior, Carroll E. Izard, Jerome Kagan, and Robert B. Zajonc, eds. Cambridge: Cambridge University Press, 264-288.

Mano, Haim (1991), "The Structure and Intensity of Emotional Experiences: Method and Context Convergence," Multivariate Behavioral Research, 26 (No. 3), 389-411.

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

Oliver, Richard L. (1980), "A Cognitive Model of the Antecedents and Consequences of Satisfaction Decisions," Journal of Marketing Research, 17 (November), 460-469.

Oliver, Richard L. (1992), "An Investigation of the Attribute Basis of Emotion and Related Affects in Consumption: Suggestions for a Stage-Specific Satisfaction Framework," in Advances in Consumer Research, Vol. 19, eds., John F. Sherry, Jr. and Brian Sternthal, Provo, UT: Association for Consumer Research, 237-244.

Oliver, Richard L. (1993a), "Cognitive, Affective, and Attribute Bases of the Satisfaction Response," Journal of Consumer Research, 20 (December), forthcoming.

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

Oliver, Richard L. and William O. Bearden (1985), "Disconfirmation Processes and Consumer Evaluations in Product Usage," Journal of Business Research, 13 (June), 235-246.

Oliver, Richard L. and John E. Swan (1989), "Equity and Disconfirmation Perceptions as Influences on Merchant and Product Satisfaction," Journal of Consumer Research, 16 (December), 372-383.

Oliver, Richard L. and Robert A. Westbrook (1993), "Profiles of Consumer Emotions and Satisfaction in Ownership and Usage," Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 6, forthcoming.

Reynolds, Thomas J. and Jonathan Gutman (1988), "Laddering Theory, Method, Analysis, and Interpretation," Journal of Advertising Research, 28 (February/March), 11-31.

Steenkamp, Jan-Benedict E. M. (1990), "Conceptual Model of the Quality Perception Process," Journal of Business Research, 21 (December), 309-333.

Swartz, Teresa A. and Stephen W. Brown (1989), "Consumer and Provider Expectations and Experiences in Evaluating Professional Service Quality," Journal of the Academy of Marketing Science, 17 (Spring), 189-195.

Watson, David, Lee Anna Clark, and Auke Tellegen (1988), "Development and Validation of Brief Measures of Positive and Negative Affect: The PANAS Scales," Journal of Personality and Social Psychology, 54 (June), 1063-1070.

Weiner, Bernard, Dan Russell, and David Lerman (1979), "The Cognition-Emotion Process in Achievement-Related Contexts," Journal of Personality and Social Psychology, 37 (July), 1211-1220.

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

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

Zeithaml, Valarie A. (1988), "Consumer Perceptions of Price, Quality, and Value: A Means-End Model and Synthesis of Evidence," Journal of Marketing, 52 (July), 2-22.

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.