yuksel (2001) the expectancy disconfirmation paradigm

http://jht.sagepub.com

Journal of Hospitality & Tourism Research

DOI: 10.1177/109634800102500201 2001; 25; 107 Journal of Hospitality & Tourism Research

Atila Yüksel and Fisun Yüksel The Expectancy-Disconfirmation Paradigm: A Critique

http://jht.sagepub.com/cgi/content/abstract/25/2/107 The online version of this article can be found at:

Published by:

http://www.sagepublications.com

On behalf of:

International Council on Hotel, Restaurant, and Institutional Education

can be found at:Journal of Hospitality & Tourism Research Additional services and information for

http://jht.sagepub.com/cgi/alerts Email Alerts:

http://jht.sagepub.com/subscriptions Subscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://jht.sagepub.com/cgi/content/refs/25/2/107 Citations

at Universiteit Maastricht on June 7, 2010 http://jht.sagepub.comDownloaded from

http://www.chrie.org

http://jht.sagepub.com/cgi/alerts

http://jht.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav

http://jht.sagepub.com/cgi/content/refs/25/2/107


JOURNAL OF HOSPITALITY & TOURISM RESEARCHYüksel, Yüksel / EXPECTANCY-DISCONFIRMATION

THE EXPECTANCY-DISCONFIRMATIONPARADIGM: A CRITIQUE

Atila YükselAdnan Menderes University

Fisun YükselSheffield Hallam University

Accurate measurement of customer satisfaction is a prerequisite for developing effectivemanagement strategies. Only with reliable customer feedback, gathered through an ade-quate and appropriate assessment framework, can managers be in possession of facts thatwill allow them to implement satisfaction improvement programs. The Expectancy-Disconfirmation Paradigm (EDP) has become the dominant framework employed in theassessment of customer satisfaction with hospitality and tourism services. However, de-spite its dominance, there remain a number of unresolved issues concerning this model.Given the lack of attention to its limitations, this article reviews a number of conceptualand operational issues relating to the EDP and questions its reliability in assessing cus-tomer satisfaction with tourism and hospitality services. Suggestions for future researchare also provided.

KEYWORDS: expectation; performance; disconfirmation; comparative standards;validity.

There have been various frameworks developed to explain customer satisfac-tion, including the Expectancy-Disconfirmation Theory, the Equity Theory, theAttribution Theory, the Value-Percept Theory, the Dissonance Theory, the Con-trast Theory, the Comparison Level Theory, the Importance-Performance Theory,and the Evaluative Congruity Theory (Yi, 1990). In general, these theories sug-gest that consumer satisfaction is a relative concept, which is always judged inrelation to a standard. Some of these theories posit that consumers judge satisfac-tion in relation to values and desires (the Value-Percept Theory), whereas otherssuggest that the standard used is the predictive expectations (the Expectancy-Disconfirmation Paradigm [EDP]), or the experience-based norms (the Compari-son Level Theory). Some theorize that satisfaction results from the comparisonbetween consumer inputs and outputs (the Equity Theory), whereas others sug-gest that satisfaction is the result of the discrepancy between expectations and per-ceived performance; the latter led to the conceptualization of the EDP (Oliver,1980), which has become the most widely applied method of consumer satisfac-tion and dissatisfaction assessment (Oh & Parks, 1997; Weber, 1997).

Journal of Hospitality & Tourism Research, Vol. 25, No. 2, May 2001, 107-131© 2001 International Council on Hotel, Restaurant and Institutional Education

107



The conceptualization of tourist satisfaction as a comparison process betweentourists’ expectations and tourists’ experiences has been employed by severaltourism and hospitality investigators (Dorfman, 1979; Hughes, 1991; Pizam &Milman, 1993; Reisinger & Waryszak, 1996; Tribe & Snaith, 1998; Weber, 1997).However, given the attention, surprisingly limited research has been carried out toexamine whether the EDP possesses any theoretical and/or methodological limi-tations, and whether it is possible to apply the model in every situation. This arti-cle, therefore, sets out to discuss the validity of the EDP in assessing customer sat-isfaction within hospitality and tourism services. The first part of the articlepresents the main assumptions of the EDP, followed by a discussion of its concep-tual and operational limitations. Suggestions for managers and future research arethen provided.

THE EDP

The EDP implies that consumers purchase goods and services with prepur-chase expectations about anticipated performance. The expectation level thenbecomes a standard against which the product is judged. That is, once the productor service has been used, outcomes are compared against expectations. If the out-come matches the expectation, confirmation occurs. Disconfirmation occurswhere there is a difference between expectations and outcomes. A customer iseither satisfied or dissatisfied as a result of a positive or negative differencebetween expectations and perceptions. Thus, when service performance is betterthan initially expected, there is a positive disconfirmation between expectationsand performance that results in satisfaction, and when service performance is asexpected, there is a confirmation between expectations and perceptions thatresults in satisfaction. In contrast, when service performance is not as good asexpected, there is a negative disconfirmation between expectations and percep-tions, which causes dissatisfaction.

The notion of discrepancy theory may be traced back to Howard and Sheth’s(1969) definition of satisfaction, which states that it is a function of the degree ofcongruency between aspirations and perceived reality of experiences. There are,basically, two methods of investigating confirmation/disconfirmation of expecta-tions. The first is the inferred approach (or the subtractive approach), and the sec-ond is the direct approach (or the subjective approach)(Meyer & Westerbarkey,1996; Prakash & Loundsbury, 1992). The inferred approach involves the compu-tation of the discrepancy between expectations and evaluations of performance.This requires researchers to draw separate information relating to customer ser-vice expectations and perceived performance. These scores are then subtracted toform the third variable, the confirmation-disconfirmation, or difference, score.The direct approach requires the use of summary judgment scales to measure con-firmation/disconfirmation, such as “better than expected to worse than expected.”The calculation of the difference scores by the researcher is avoided, because therespondents can be asked directly the extent to which the service experienceexceeded, met, or fell short of expectations. As an alternative approach, subjectivedisconfirmation represents a distinct psychological construct encompassing a

108 JOURNAL OF HOSPITALITY & TOURISM RESEARCH



subjective evaluation of the difference between product performance and thecomparison standard (Churchill & Surprenant, 1982; Oliver, 1980). That is, sub-jective disconfirmation encompasses a set of psychological processes that maymediate perceived product performance discrepancies.

Both the inferred and the direct methods of EDP have been used by hospitalityand tourism researchers in various studies that assess international travelers’ sat-isfaction levels as well as in studies investigating customer satisfaction with hotelservices (e.g., Barsky, 1992; Barsky & Labagh, 1992; Cho, 1998; Danaher &Haddrell, 1996; Pizam & Milman, 1993; Reisinger & Waryszak, 1996; Tribe &Snaith, 1998; Weber, 1997; Whipple & Thach, 1988). The majority of hospitalityand tourism satisfaction studies have assumed that the EDP is a valid and reliableframework that can be confidently used to determine customer satisfaction withhospitality and tourism services. These studies, however, do not seem to havescrutinized the extent to which the EDP measures what it intends to measure,although verification of its validity in determining customer satisfaction is essen-tial. Some of the conceptual and operational issues relating to the EDP are dis-played in Table 1.

THE EDP—LIMITATIONS

Despite its dominance, the validity and reliability of the EDP in assessing cus-tomer satisfaction with hospitality and tourism services, which are experiential innature, may be disputable for a number of reasons. First, the use of expectationsmight be less meaningful for experiential services than for tangible consumergoods that are easy to evaluate prior to purchase (Hill, 1985). The EDP predictsthat customers will be satisfied (dissatisfied) when their initial expectations aremet (unmet); however, this may not necessarily be the case in every consumptionsituation. Depending on the situation, some customers may be satisfied with theservice experience even when the performance falls short of their predictiveexpectations but above the minimum tolerable level.

Second, in line with the conventional EDP, many of these studies have usedpredictive expectations as the comparative standard. However, there is inade-quate research evidence on whether consumers use only predictive expectationsin their postpurchase product evaluations, whether they use other standard(s),which they bring into the consumption experience (e.g., minimum tolerable level,desires, ideals), or other standards that may emerge after the purchase (e.g., whatothers have received). Satisfaction processes may differ across products/services(Churchill & Surprenant, 1982), and the standard used may depend on situation(e.g., low-high involvement). The use of different comparative standard ques-tions may yield different levels with which the performance is compared, andmay produce different results in terms of customer satisfaction (Woodruff,Cleanop, Schumn, Godial, & Burns, 1991). If the relationship betweendisconfirmation and affective outcomes (e.g., satisfaction) or behavioral out-comes (e.g., return intentions) differs, depending on the comparative standardused, commitment to the conventional EDP might be wrong. For instance, meet-ing or exceeding predictive expectations may result in indifference (Swan &

Yüksel, Yüksel / EXPECTANCY-DISCONFIRMATION 109



Trawick, 1980), whereas the resultant satisfaction and future behavioral intentionlevels may be different, and possibly higher, when consumer values, desires, orideals are met.

Another limitation relates to the fact that the EDP cannot accommodate thedynamic nature of expectations. Consumers’ initial expectations of a product orservice might be substantially different from their expectations if measured after aservice experience that involves several encounters, as in the case of many hospi-tality and tourism services. If consumers are using these retrospective expecta-tions in their postpurchase evaluations, then the reliability of suggesting theoccurrence of a positive or negative confirmation/disconfirmation of initiallymeasured expectations is disputable. Another limitation is that the model isunable to accommodate the potential effect of customer perceptions of perfor-mance of alternative product(s) on evaluation judgments of the focal product/ser-vice. The EDP does not seem to provide comparative information on performanceof competitors, which is necessary for managers in planning competitive actions.


Table 1Conceptual and Operational Issues

Relating to the Expectancy-Disconfirmation Paradigm

ConceptualPrepurchase Without expectations, disconfirmation cannot occur. How

expectations realistic would it be to expect customers to have firmexpectations of all attributes prior to purchase in everyconsumption situation?

Meaning of Would an expectation question signify the same meaning toexpectations everyone?

Single or multiple Does customer satisfaction come from disconfirmation ofcomparison expectations alone?

Logical inconsistency Would meeting low expectations generate satisfaction as themodel predicts?

Why do customers report overall satisfaction when their ratingsindicate service performance falling short of their initialexpectations?

Disconfirmation Would the disconfirmation process operate in every consumptionprocess situation?

OperationalTiming of the Should it be measured before or after the service experience?

expectationmeasurement

“I have high If scores on expectations are consistently and constantly high,expectation” norm then it may never be possible to exceed them.

Possibility of Would meeting a high expectation with a high performance andmisleading meeting a low expectation with a low performance signify equalconclusions satisfaction in each case?

Dual administration Answering the same set of questions twice might bore theand possibility of respondents.response-tendency-bias



Moreover, the assessment of customer satisfaction as a difference score betweencustomer expectations and perceived performance, a common methodology usedby researchers, may hold administrative difficulties, which may consequentlyhave an impact on the reliability of the collected data. These conceptual and oper-ational issues are now reviewed in detail in the following section.

PREPURCHASE EXPECTATIONS

One of the problems related to the EDP is the suggested sequence of the model,which presupposes that everyone has precise expectations prior to the serviceexperience. It is obvious that without these prior expectations, confirmation/disconfirmation of expectations cannot occur (Halstead, Hartman, & Schmidt,1994). However, the logic of the EDP, which states that everyone has firm expec-tations of all attributes prior to service experiences, might be less meaningful insituations where customers do not know what to expect until they experience theservice. Unlike tangible goods, with which search attributes are dominant, tour-ism and hospitality services are experiential in nature and contain high percent-ages of experience and credence properties (Reisinger & Waryszak, 1996).Search properties refer to those attributes that a consumer evaluates before engag-ing in the service. These properties are primarily tangibles, which are physicalrepresentations of the service (e.g., facilities, equipment, and appearance of per-sonnel). Experience properties are those attributes that can be only assessed afterpurchase or during consumption such as taste, value, and purchase satisfaction(Zeithaml, 1981). Credence properties are the attributes that the consumer findsimpossible to evaluate even after purchase and consumption (e.g., the backstagehygiene conditions). In general, those services that are based heavily on experi-ence and credence properties, such as hospitality and tourism services, may bedifficult to predict and evaluate (Hill, 1985). Moreover, the variability in the ser-vice level that is provided from encounter to encounter in hospitality and tourismservices may create uncertainty, which may inhibit the formation of preciseprepurchase expectations (Jayanti & Jackson, 1991). Thus, the assumption thatthe formation of firm and realistic attribute-specific expectations occurs prior toevery purchase in the hospitality and tourism context may be incorrect.

Customers with little or no brand experience of products and services consti-tute a special case in the EDP, as it is not clear how the EDP may be applied to theevaluation of services for which the consumer has little information or experienceto generate a meaningful expectation (Halstead et al., 1994; McGill & Iacobucci,1992). Customer expectations of completely unfamiliar experiences (e.g., firsttime travel to Eastern Europe) are almost meaningless (Halstead et al., 1994).“Though one might assume that expectations based on travel to other parts ofEurope would be an appropriate proximate, this too, may have little relevance tothe actual experience” (Halstead et al., 1994, p. 116). Lack of any kind of previousexperience with the service, or not knowing what to expect as a result of theabsence of prepurchase information, may result in tentative and uncertain expec-tations (Crompton & Love, 1995; Mazursky, 1989; McGill & Iacobucci, 1992).




In these situations, regarding expectations as firm criteria against which makeevaluative judgments is likely to be fallacious (Crompton & Love, 1995).

Learning from previous service experiences may result in more accurate andstable expectations (Day, 1977). Experienced customers may, therefore, makebetter choices when repurchasing, they may have more realistic expectations, andthey may be more satisfied with their choices (Fisk & Coney, 1982; Halstead et al.,1994; Westbrook & Newman, 1987). On the other hand, inexperienced customersmay rely on external sources of information (Halstead et al., 1994) such as theorganization’s promotional material and word-of-mouth communication to shapetheir expectations, leading to expectations that are weaker, less complete, less sta-ble, and more superficial (Halstead et al., 1994; Mazursky, 1989; McGill &Iacobucci, 1992). Thus, measuring expectations may not be valid in situationswhere consumers do not have well-formed expectations prior to service experi-ence (Carman, 1990). In such situations, as Carman (1990) notes, expectationsmay be assumed to be zero, and expectation measures do not need to be obtainedevery time the perception measures are obtained.

Another problem with the EDP is that postpurchase evaluations may not bebased on initial expectations. For instance, McGill and Iacobucci (1992, p. 571)reported that

in contrast to what might have been expected from the literature on thedisconfirmation paradigm, that comparison of subjects’ listing of features thataffected their level of satisfaction in the post-experience questionnaire were notentirely consistent with the listing of factors that they expected to affect theirlevel of satisfaction in the pre-experience questionnaire.

Similarly, Whipple and Thach (1988, p. 16) stated that expectations may beimportant indicators of choice preference and “there is evidence that pre-purchasechoice criteria and post-purchase choice criteria are not the same.” If differentevaluative criteria are used before and after a service experience then “the initialexpectation framework is disregarded and is of little value for measuringsatisfaction.”

INTERPRETATION OF EXPECTATION QUESTION

Another problem with the EDP relates to the meaning of the expectation ques-tion to the respondent (i.e., whether the expectation question signifies the samemeaning to everyone).

Expectation represents a baseline against which performance is compared, andit may vary from a minimum, tolerable level of performance and estimates ofanticipated performance to some concept of ideal or perfect service (Ennew,Reed, & Binks, 1993). Given the confusion about the precise meaning of expecta-tion, the use of this concept as a means to conceptualize comparison standards hasbeen criticized by a number of researchers (Woodruff et al., 1991). The expecta-tion component of both service quality and satisfaction investigations might haveserious discriminant validity shortcomings, which causes the performance-




minus-expectation measurement framework to be a potentially misleading indi-cator of customer perceptions of services. For instance, findings reported in Teas’s(1993) study clearly suggest that not all respondents interpret the question ofexpectation in the same way and there may be a considerable degree of confusionamong respondents concerning the actual question being asked. Teas identifiedthat some responses suggested the expectation questions involved an importancemeasure, whereas other respondents used the scale to predict the performancethey would expect (i.e., forecasted performance). A few respondents interpretedthe question in terms of the ideal point concept (i.e., the optimal performance,what performance can be), and minimum tolerable concept (i.e., what perfor-mance must be).

If there is a difference between customers’ interpretations of the expectationquestion, then the scores obtained from the performance-minus-expectation pro-cess can be misleading. Assume that two respondents rated different scores on theexpectation question concerning a visually appealing restaurant (1 and 7, respec-tively), and rated the same score of 7 on perceived performance. As a result of thelow score on expectation, the calculation suggests a positive gap (+6) in the firstcase, whereas the latter’s ratings suggest a gap score of 0, which consequentlyimplies that the satisfaction is higher in the first case. However, what if therespondent rated low on the expectation question because she wishes to savemoney and desires a visually unappealing restaurant, or the visual appealing issueis unimportant to the respondent? The last probability in particular represents apotential measurement validity problem. Although the first respondent rated 1 onthe expectation scale because the visually appealing issue is an unimportant fac-tor, the resultant performance-minus-expectation (P–E) score (+6) suggests ahigher level of quality/satisfaction in that situation than the quality/satisfactionlevel suggested in the second respondent’s case, in which the performance is highon an important attribute. Thus, it is illogical to assume that “scores with high per-formance on attributes of low importance items should reflect a higher servicequality [satisfaction] than equally strong performance on attributes of highimportance” (Teas, 1994, p. 134).

COMPARATIVE STANDARD ISSUE

The majority of past studies have assumed that consumers use only the predic-tive expectations in their postpurchase evaluation process as the comparisonbaseline. The validity of this contention is, however, questionable. Consumersmight be more likely to use predictive expectations based on external communi-cations (e.g., advertisement) before the purchase (in their choice decision mak-ing), whereas different standards may be used after the purchase (Woodruff et al.,1991) (e.g., what others have received, price paid vs. service delivered, perceivedperformance of alternative products). Furthermore, different customers mightuse different comparative standards, and depending on the situation, the form ofthe comparative standard may change. Apart from expectations, customers mayuse other standards to arrive at satisfaction judgments. For example, consumershigh in experience with the product category might use one of the experience-




based norms, such as average performance, favorite, or last used as a standard,whereas those who have little experience may use what others have received orwhat is promised (Spreng & Dixon, 1992). “Difficult-to-please” (or overlydemanding) type customers may use ideal expectations in their assessment,whereas “tolerant” customers may use minimum acceptable as a standard.

The comparative standard issue is extremely important because different typesof standards may yield different levels against which perceived experience iscompared (Woodruff et al., 1991). There is a possibility that respondents may besensitive to the type of the standard used in a customer satisfaction/dissatisfactionresearch (Olander, 1979; Woodruff et al., 1991), which in turn might have animpact on the resultant level of customer satisfaction identified by the researcher.If so, this gives rise to an interesting question. Would different satisfaction resultsbe obtained if the wording of a disconfirmation question were changed (i.e., com-parison of performance relative to a competitor instead of to the expected perfor-mance)? Would meeting or exceeding predictive expectations created by adver-tisements produce more or less satisfaction and return intention levels than whenthe customer finds the performance delivered by the organization was better thanthe best experience she had? If satisfaction ratings are contingent on which thestandard being used, “then historical commitment to expectancy disconfirmationby the academics and practitioners may be detrimental to advancing knowledgecritical to understanding customer satisfaction/dissatisfaction” (Woodruff et al.,1991, p. 103).

SINGLE OR MULTIPLE COMPARISON

Equally important, the majority of past hospitality and tourism satisfactionstudies using the EDP assume that consumers engage in a comparison processthat involves a single comparative standard. There is, however, emerging litera-ture suggesting that no single model or unique comparison process may fullyexplain consumer satisfaction/dissatisfaction judgments (Boulding, Karla,Staelin, & Zeithaml, 1993; Erevelles & Leavitt, 1992; LaTour & Peat, 1979;Sirgy, 1984; Tse & Wilton, 1988).

Several theories, including the Comparison Level Theory, the Equity Theory,and the Evaluative Congruity Theory, suggest that there are several basic determi-nants of product comparison. According to the Comparison Level Theory, thesemight include (a) consumers’ prior experiences with similar products, (b) situa-tionally produced expectations (those created through advertising and promo-tional efforts), and (c) the experience of other consumers who serve as referentpoints (LaTour & Peat, 1979). Similarly, the Equity theory (Adams, 1963)assumes that satisfaction exists when consumers perceive their output/input ratioas being fair. Whether a person is treated equitably (thus satisfied) may depend onvarious factors including the price paid, the benefits received, the time and effortexpended during the transaction, previous transactions, and the outcomes of allparties sharing the same experience (Woodruff, Ernest, & Jenkins, 1983). Simi-larly, Sirgy’s (1984) Evaluative Congruity model suggests that consumers mightuse multiple standards to arrive at satisfaction judgments. However, Sirgy argued




that (in)congruities may take place between different perceptual and evoked ref-erent states. Such (in)congruities may take place between (a) new product perfor-mance after usage and expected product performance before use, (b) new productperformance after use and old product performance before use, (c) expected prod-uct performance after purchase and ideal product performance before purchase,and (d) expected product performance after purchase and deserved product per-formance after use. Sirgy found that each of these congruities either separately orin combination significantly influenced customer satisfaction.

These theories suggest the occurrence of not only a single but also a multiplecomparison process, including complex interactions, which may take place eithersequentially or simultaneously (Oliver & DeSarbo, 1988), and that consumer sat-isfaction may not be the result of disconfirmation of predictive expectationsalone. In this sense, as some researchers suggest (Erevelles & Leavitt, 1992), abetter model of consumer satisfaction/dissatisfaction should take into account theoccurrence of multiple processes and multiple standards of comparison.

LOGICAL INCONSISTENCY

An additional problem related to the EDP is its main presumption. The currentlogic of the EDP predicts customers will evaluate a service favorably as long astheir expectations are met or exceeded (Iacobucci, Grayson, & Ostrom, 1994).However, this may not be the case every time. In situations where consumers areforced to buy an inferior, less desirable brand because their preferred brand is notavailable, then consumers may not necessarily experience disconfirmation of apreexperience comparison standard (LaTour & Peat, 1979). “If a less desirablebrand was indeed as undesirable as the customer had expected it to be, the con-sumer would experience no disconfirmation, and yet could be quite dissatisfied”(Iacobucci et al., 1994, p. 16). In addition, users of new brands who experienceunfavorable disconfirmation of a high preexperience standard that was generatedthrough advertising may still be satisfied with the brand if it has more of thedesired attributes than competing brands (LaTour & Peat, 1979).

The key role played by expectations in determining the level of satisfaction isquestionable. Consumers may show satisfaction or dissatisfaction for aspectswhere expectations never existed (McGill & Iacobucci, 1992; Yi, 1990). In herresearch on tourist satisfaction, Hughes (1991, p. 168) reported that “surprisingly,even though experiences did not fulfill expectations, a considerable number oftourists were relatively satisfied.” Similarly, Pearce (1991) maintained that tour-ists may be satisfied even though their experiences did not fulfill their expecta-tions. In a study of service quality perceptions of clinic customers, Smith (1995)reported a similar finding that respondents described themselves as extremelypleased with the clinic even where an aggregate performance-minus-expectationscore was negative (P < E). Similarly, Yüksel and Rimmington (1998) found thatcustomers might be reasonably satisfied even if the service performance does nottotally meet their initial expectations. These findings cast doubts on the logicalconsistency of the expectancy-confirmation/disconfirmation model, as it predictsthe customer to be dissatisfied when initial expectations are not met.




One possible explanation for this could be that some consumers may use mini-mum acceptable as a comparative standard in certain situations, and the perfor-mance above the minimum acceptable but below the predicted expectations maynot necessarily create dissatisfaction. The latitude-of-acceptance (Anderson,1973) or zone-of-indifference (Woodruff et al., 1983) concept might explain whythose customers whose expectations are unmet report satisfaction. This conceptsuggests that purchasers are willing to accept a range of performance around apoint estimate as long as the range could be reasonably expected (Oliver, 1997). Ifcustomer tolerance of some deviation from expectations exists, a level of serviceless than the expected does not generate dissatisfaction (Saleh & Ryan, 1991).Perceived performance within the zone of indifference probably does not causemuch attention to be directed toward the evaluation process. In contrast, perceivedperformance outside the zone of indifference is unusual and attention getting,leading to an emotional response (Woodruff et al., 1983). An alternative explana-tion could be that customers might engage in a trade-off process, wherein astrength of an attribute may compensate for the weakness(es) of another attributeand may lead to overall satisfaction (Lewis, Chambers, & Chacko, 1995).

PERCEIVED PERFORMANCE AND DISCONFIRMATION

Another problem with the EDP is the assumption that the disconfirmationprocess will operate in every consumption situation. Depending on the productcategory and the nature of customers’ expectations, the customer assessment ofcertain services may not even rely on disconfirmation, but instead rely on per-formance evaluations only (Churchill & Surprenant, 1982; Halstead et al., 1994).For instance, when customer expectations have become passive (not actively pro-cessed), as in the case of continuously consumed services, or when there is a highfamiliarity with the service (repeat patronage), the confirmation/disconfirmationprocess may not operate unless performance is clearly outside the range ofexperience-based norms (Oliver, 1989).

Performance holds a preeminent role in the formation of customer satisfactionbecause it is the main feature of service that creates the consumption experience(Halstead et al., 1994). Performance may be the only tangible evidence on whichto base consumer service evaluation (Jayanti & Jackson, 1991), and consumers’judgment of the service organization’s performance alone can be an evaluation ofthe global quality judgment itself (Llosa, Chandon, & Orsingher, 1998). The merefact of asking a respondent to mark perceptions of the firms’ performance maylead a respondent to compare mentally perceptions against expectations (Llosaet al., 1998). In other words, estimates of perceptions might already include a per-ception-minus-expectation mental process. A similar suggestion is also found inGronroos’s (1993) work. He argued that

measuring expectations is not a sound way of proceeding anyway, because expe-riences are in fact perceptions of reality, and inherent in these perceptions are theprior expectations. Consequently if first, one way or the other, expectations are




measured and then experiences are measured, then expectations are measuredtwice. (p. 56)

OPERATIONAL DIFFICULTIES

In addition to the conceptual issues discussed above, there are also operationaldifficulties that make the assessment of customer satisfaction through the EDPquestionable.

TIMING OF THE EXPECTATION MEASUREMENT

There is a continuing debate on the timing of expectation measurement in cus-tomer satisfaction studies. Some researchers suggest that expectations should besolicited before the service experience (Carman, 1990), whereas others argue thatexpectations may be measured after the service experience (Parasuraman,Zeithaml & Berry, 1988).

Those supporting the measurement “before the experience” contend that “tobe of value expectations should be elicited prior to the service being provided,otherwise the risk is so great that expectations will be contaminated by percep-tions of the actual service provided” (Getty & Thomson, 1994, p. 8). This methodhas been employed by a number of researchers in tourism and hospitality litera-ture (e.g., Fick & Ritchie, 1991; Hughes, 1991; Johns & Tyas, 1996; Pizam &Milman, 1993; Tribe & Snaith, 1998; Weber, 1997; Whipple & Thach, 1988). Forinstance, in their investigation of satisfaction among first time visitors to Spain,Pizam and Milman (1993) solicited travelers’ expectations before they left andexamined their perception of 21 destination attributes after they returned fromtheir holiday. Weber (1997) adopted a similar approach in her research on satis-faction of the German travel market in Australia. Weber distributed question-naires to tourists on their arrival and asked them to complete the pretrip section onthe day of arrival and the posttrip section at the end of their holiday.

Although adopted by a number of researchers, measuring expectations prior toservice experience is problematic. It is reasonable to assume that in some situa-tions where the pre-post method is adopted, not all of the respondents answer tothe expectation part of the questionnaire at the required time (e.g., before the ser-vice experience). A number of respondents may wait to complete the expectationpart after the service experience, which may produce “hindsight bias” (Weber,1997; Yüksel & Rimmington, 1998). Prior expectations may be modified duringthe service encounter, and these modified expectations may be used in the com-parison process (Danaher & Mattsson, 1994; Gronroos, 1993; Iacobucci et al.,1994; McGill & Iacobucci, 1992). An observed effect (satisfaction) may be due toan event that takes place between the pretest and the posttest (Cook & Campbell,1979). According to Boulding et al. (1993, p. 9), “A person’s expectations justbefore a service contact can differ from the expectations held just after the servicecontact because of the information that enters the system between serviceencounters.” The unpredictability of tourism events lies at the heart of vacationexperience (Botterill, 1987), and events that are completely unanticipated prior to




a trip may become significant contributors to overall holiday satisfaction (e.g.,unanticipated sources or discoveries such as the weather, the travel companion,and unexpected adventures) (Pearce, 1980; Weber, 1997).

If learning takes place during the service encounter, and expectations are mod-ified as a consequence of this, then the use of initially measured expectations insatisfaction assessment is not logical. Satisfaction is most accurately measured atthe conclusion of the transaction, and thus the expectation referent relevant to sat-isfaction would be the one actually used by the consumer in satisfaction forma-tion, not necessarily the one measured before consumption (Oliver, 1997). Thisbecomes problematic if the consumer has updated (downgraded or elevated) hisor her expectations during consumption. Recent studies (e.g., Zwick, Pieters, &Baumgartner, 1995) suggest that updated expectations may be more influential insatisfaction judgments than preconsumption expectations. Given the evidenceabove, it seems reasonable to assume that a customer’s expectations prior to theservice experience may be different from those against which they compare theactual experience.

Given the complications that surround the measurement of expectations priorto the service experience, an alternative method of soliciting expectations is tomeasure them after the service experience or simultaneously with the serviceexperience. A number of researchers have used this method (Dorfman, 1979; Fick &Ritchie, 1991; Parasuraman et al., 1988). For example, Parasuraman et al. (1988)and Dorfman (1979) asked respondents to complete both expectations and per-ceptions questions at the same time. Based on their previous experience with theservice, respondents were asked what they expected and then were asked whatthey had experienced. However, this approach is also questionable, as expecta-tions might be over-/understated if the tourists have a very negative or positiveexperience (Yüksel & Rimmington, 1998). In addition, respondents’ capability tocorrectly remember prior expectations raises doubts about the validity of thesemeasures (Loundsbury & Hoopes, 1985). It is argued that if expectations are mea-sured after the service experience or simultaneously with the experience, it is notthe expectations that are being measured but something that has been biased bythe experience. For example, Halstead (1993) found that expectations that aremeasured after service experience were higher for dissatisfied customers than forsatisfied customers. This suggests that recalled expectations will be biased towardthe experienced performance (Oliver, 1997).

CONSISTENTLY HIGH EXPECTATION SCORES

From a practical perspective, Dorfman (1979) draws attention to the fact thatexpectations are generally rated very highly. That is, respondents may feel moti-vated to demonstrate an “I have high expectation” social norm, and the expectedlevel therefore may exceed the existing level for no other reason than this type ofresponse bias (Babakus & Boller, 1992). In the service quality area, for example,Babakus and Boller (1992, p. 257) pointed out that “in general when people areasked to indicate an expected level and an existing level of service they seldom




rate the expected level lower than the existing level.” If these scores are almostconstant, “then there is little point in including them in an instrument, since theywill not give responses significantly different from using the perception scoresalone” (Crompton & Love, 1995, p. 15).

The results of Fick and Ritchie’s (1991); Parasuraman, Zeithaml, and Berry’s(1988, 1991, 1994); Smith’s (1995); and Tribe and Snaith’s (1998) researchreveals that scores on expectations are indeed rated consistently higher than thescores on the performance component. For instance, Smith (1995) reports that themean score for the expectation scale in her study was 6.401 (standard deviation =0.347). In Parasuraman et al.’s (1991) study, the mean expectation score was6.22. Buttle (1995), in a comprehensive review of research studies on servicequality and satisfaction, reported that the average score of expectations was6.086. The results of Bojanic and Rosen’s (1995) study in a family restaurantenvironment demonstrated that consumers’ perceptions about the actual level ofservice provided fell significantly short of their expectations. Similarly, theresearch undertaken by Yüksel and Rimmington (1998) revealed that the meanexpectation scores for restaurant services was significantly higher than the meanperformance perception scores. These findings suggest that it would be difficultto satisfy tourists as expectations will never be met or exceeded.

In this sense, to ensure that expectations are exceeded, some researchers havesuggested that service providers should understate the destination or organiza-tion’s capability of delivering these experiences in promotional efforts. Forexample, Pizam and Milman (1993, p. 208) suggested that “it would be more ben-eficial to create modest and even below realistic expectations.” Although this is asensible and potentially effective suggestion in theory, it is questionable whetherit can, and should, be implemented practically (Weber, 1997). The problem is thattourists may not want to spend time and money in a destination in the first place ifpromotional efforts convey the possibility of the destination being unable todeliver adequate services. Moreover, establishing a threshold in which expecta-tions are raised high enough to attract customers, but low enough to allow forexpectations to be exceeded, is difficult (Weber, 1997). The present logic of theEDP encourages managers to lower expectations for a given service, and thenhave the users discover a superior outcome than expected, thus leading to greatersatisfaction. Obviously, low expectations would affect the motivation and wouldtherefore reduce purchase and consumption (Williams, 1998).

In addition, expectations are argued to have a direct positive effect on satisfac-tion because “without observing the performance, expectations may have alreadypredisposed the consumer to respond to the product in a certain way (the higherthe expectations, the higher the satisfaction or vice versa)” (Oliver, 1997, p. 89). Ifhigh expectations lead to more favorable ratings, then one may suggest that com-panies should strive to raise expectations substantially above the product perfor-mance to obtain a higher product evaluation (Yi, 1990). However, the validity ofthis suggestion is also questionable. Raising expectations substantially above theproduct performance and failing to meet these expectations may backfire as smalldiscrepancies may be largely discounted, whereas large discrepancies may result




in a very negative evaluation. Given these complications, it appears that increas-ing service performance may be the safest strategy.

RELIABILITY OF DIFFERENCE SCORES

The majority of consumer satisfaction studies adopted the use of the differencescore, which results from the subtraction of respondents’ expectation scores fromtheir scores based on performance. The resultant scores, however, may result inmisleading conclusions (Teas, 1993). For example, there are six ways of produc-ing a score of –1 (e.g., P = 1 – E = 2; P = 6 – E = 7) and seven ways of producing ascore of 0 (e.g., P = 1 – E = 1; P = 7 – E = 7). The point in question is, would thelevel of resulting satisfaction be the same in cases where low expectations are metby low performance and high expectations are met by high performance? Thelogic of EDP predicts that satisfaction in each case would be equal; however, dif-ferent combinations of expectations and performance (high performance/highexpectations; low performance/low expectations) would result in different satis-faction states (see Chon, 1992; Chon, Christianson, & Chin-Lin, 1998; Chon &Olsen, 1991).

Another problem related to the EDP is the possibility of a ceiling or a flooreffect (Oh & Parks, 1997; Yi, 1990), which might occur when an individual gavethe highest score on the expectation scale prior to consumption. When a productperforms well above these initial expectations, the individual can only give thesame score (i.e., the highest score on the scale) for perceived performance. In thiscase, the inferred disconfirmation would be zero although a positive discon-firmation occurs in the consumer’s mind (Yi, 1990). Moreover, the diagnosticability of the direct disconfirmation approach (better/worse than expected) maybe questionable because this scale cannot indicate whether the expectations beingconfirmed or dis/confirmed were high or low. Intuitively speaking, the case ofmeeting or exceeding low customer expectations is qualitatively different frommeeting or exceeding high expectations. Therefore, the direct disconfirmationscale seems to be of little use for diagnostic analysis (Yüksel & Rimmington,1998).

SUGGESTIONS AND CONCLUSION

The EDP has become the dominant framework employed in the assessment ofcustomer satisfaction with hospitality and tourism services. However, despite itsdominance, the reviewed literature demonstrates that there are a number of unre-solved issues concerning the validity and reliability of the model (see Table 1),and this raises doubts about whether formulating management and marketingstrategies based on results derived from the EDP is appropriate.

Empirical research is necessary to further analyze the concept of customer sat-isfaction in the tourism and hospitality context, and to test the reliability and valid-ity of the EDP in its measurement. There is a need to clarify whether the use ofprior expectations and disconfirmation measures are applicable in all situations,and whether the use of a direct measure of perceived performance is a better pre-




dictor of customer satisfaction in certain situations. There is also a need to under-stand the formation of expectations in the tourism and hospitality contexts; therole of expectations in customer choice, satisfaction, and behavioral intentions;and the relative ability of other comparison standards in determining customersatisfaction. From a management point of view, there is also a need to developstraightforward and informative frameworks that could provide guidelines onways in which elements in the service environment can be managed to increasecustomer satisfaction and repeat business. Several research suggestions are dis-cussed in the following paragraphs.

Prepurchase versus retrospective expectations. The main difficulty arisesfrom the model’s assumption that customers have precise expectations of all ser-vice attributes prior to the purchase. In certain situations, however, the validity ofthis assumption is questionable. When a service or a product is new, and meaning-ful (or precise) expectations have not been formulated, the use of expectation-based measurement in the assessment of consumer satisfaction seems to behighly controversial. The use of prior expectations as the standard of comparisonmay also be inappropriate when consumption involves multiple encounters tak-ing place over an extended period of time. In multiple-encounter service experi-ences, such as vacations or hotel-stay experiences, a shift in the postpurchasestandard of comparison is likely to occur as a result of change in needs, as well asin the importance of various product attributes. It is, therefore, reasonable toargue that standard(s) at the time of evaluation rather than prior expectations maybe used in the judgment process.

The use of prior expectations in the assessment of customer satisfaction mayalso be redundant when there is a high familiarity with the product or service, asthe level of expectations is likely to be very similar to the actual performance.Thus, in such situations a more direct measure of perceived performance may be amore useful and straightforward method to study customer satisfaction than morecomplex composite measures of expectation disconfirmation (Yüksel &Rimmington, 1998). To establish which framework is a better predictor of cus-tomer satisfaction and a more straightforward way of assessment in the hospital-ity and tourism contexts, further empirical research is needed. There is also a needto understand whether the expectation-disconfirmation process operates in everytourism and hospitality service consumption situation, and if it does, what are itscausal influences.

Ability of expectations in predicting satisfaction. The review of literature dem-onstrates that predictive expectation-based disconfirmation measures yield atbest only modest correlation with satisfaction measurements (Cadotte, Woodruff, &Jenkins, 1987). This gives rise to the question of whether predictive expectationstandard best predicts customer satisfaction. There are other forms of comparisonstandards, such as minimum tolerable, desired, deserved, ideal, and experience-based norms (Miller, 1977; Sirgy, 1984; Spreng & Olshavsky, 1993; Tse & Wil-ton, 1988; Woodruff et al., 1983) that have not been fully researched in hospitalityand tourism satisfaction studies. Understanding which standard (or combinationof standards) may be most appropriate to study consumer satisfaction in different




situations is imperative (i.e., would other forms of standards predict satisfactionbetter; do consumers use different comparative standards in different situations;do satisfaction processes differ across hospitality and tourism services). Under-taking a comparative research is warranted to test the relative ability of differentcomparison standards, to predict customer satisfaction within different hospital-ity and tourism settings, and to ascertain whether the use of different standardsyields different customer satisfaction results. For instance, it would be interestingto investigate whether comparing predictive expectations, perceived desires, orexperience-based norms to postpurchase performance will yield the same resultsin terms of customer satisfaction.

A comparative analysis in this nature requires that respondents complete atleast two sets of comparative standard questions along with another set necessaryto measure the service performance. However, such a survey may suffer fromlength and respondent fatigue (e.g., if there were 15 important service attributes,then 45 separate questions would be necessary). One way of dealing with thisproblem could be to design separate instruments including a single comparativestandard questions set and a performance set and use a multiple sample of respon-dents (Oliver, 1997; Yüksel & Rimmington, 1998). However, researchers need toestablish that differences found in the analysis could reasonably be attributed toreasons other than the sample differences.

Logical inconsistency. One of the main difficulties with the EDP is its assump-tion that customers will evaluate services favorably as long as their expectationsare met or exceeded. However, when consumer behavior is constrained to pur-chasing a product that the consumer expects to perform poorly, and it does, theEDP wrongly assumes that the consumer will be satisfied. This logical inconsis-tency needs to be addressed.

As Spreng, Mackenzie, and Olshavsky (1996) observe, the use of desires as thestandard of comparison may eliminate this logical inconsistency because itchanges the prediction of customer satisfaction level (if the less desirable productperformed as undesirable as the consumer had expected, the consumer could bequite dissatisfied). The use of desires as the standard of comparison may also bemore appropriate when a service experience or a product is new and a consumerhas no expectations about it. In particular, as desires or needs (e.g., desire for vari-ety) drive the consumption, then the postpurchase comparison process mayinvolve evaluating the extent to which the product or service fulfills customerdesires (Cote, Foxman, & Cutler, 1989) rather than expectations created by adver-tisement. Further research is necessary to test the ability of the desire-congruency,as well as the need fulfillment, to explain customer satisfaction in the tourism andhospitality contexts.

Ability of predictive expectations in determining behavioral intentions. Theconventional EDP suggests that if the product performance is consistent with con-sumer expectations, then customers will be satisfied and they will be more likelyto repurchase. However, as common experience suggests such relationshipsbetween satisfaction and repurchase, expectations may not always hold (Gupta &




Stewart, 1996). Consumers may switch to competitive brands even if their predic-tive expectations are met. The point in question is would fulfillment of brandexpectations, created by advertisement, play the most important role in the deter-mination of postpurchase behavior such as repurchase, switching, and recom-mendations.

There might be a difference between the nature of standards and the extent oftheir influence on the determination of affective (e.g., satisfaction) and behavioraloutcomes (e.g., repurchase). Satisfaction may be determined by evaluating aproduct’s ability to fulfill one’s current desires, wants, needs, or expectations.The brand expectations, however, might be less influential in the formation ofbehavioral intentions (Gupta & Stewart, 1996). In addition to product availabilityand accessibility, the behavioral intentions may simply be determined by theextent to which the brand performs better or worse in comparison with other alter-native products. If this is the case, then developing management strategies forrepeat business based on the results of expectation-disconfirmation measuresonly is inadvisable. Further research is necessary to determine the ability ofexpectations in predicting behavioral intentions, and understand whether con-sumers use different comparison standards when arriving at their satisfaction andbehavioral intention judgments.

Satisfaction—end result of a single or a multiple evaluation process? Anothermain difficulty with the EDP is the fact that it views expectations as the primarydeterminant of customer dis/satisfaction, however the cause of dis/satisfactionmay not be the confirmation/disconfirmation of expectations alone. For example,a cost conscious customer may feel highly dissatisfied with the service when shefinds out that the cost of the same service to another customer was less, eventhough the customer’s scores on predictive expectations and perceived perfor-mance may suggest satisfaction. Similarly, restaurant customers may developdissatisfaction if their perception of service transaction fairness (e.g., price paidvs. level of service delivered; service at the next table was more prompt) is beyondthe acceptable range. Alternatively, a hotel customer may be satisfied eventhough the performance delivered lagged behind what she had predicted, but atleast was above the minimum acceptable level. Thus, to increase the model’s abil-ity in accounting for customer satisfaction and provide a much richer picture of it,the effects of equity perceptions and customer tolerance levels on the develop-ment of customer satisfaction judgments need to be recognized.

The use of an equity construct may provide a richer picture of satisfaction thatmay not be available using the traditional EDP (Erevelles & Leavitt, 1992). Theequity construct might be useful in capturing satisfaction in situations where sat-isfaction is evaluated relative to other parties in an exchange, and the outcomes ofall parties sharing the same experience are considered in the evaluation (Oliver &DeSarbo, 1988). The determination of consumer tolerance levels (or the mini-mum acceptable service levels), on the other hand, may guide managers not to puttoo much time and effort on areas where high performance would not be noticedand rewarded by customers. Maintaining the service performance at adequatelevels in these areas might be a cost-effective strategy. It is, however, important to




note that different consumers may operate at different tolerance levels, and thatthe tolerance level might change depending on the importance of the occasion, themood of the customer, and so on. Identification of the tolerance level may, there-fore, be potentially difficult (see Parasuraman et al., 1994). Nevertheless, theinclusion of overall measures to assess customer opinions about the fairness of theservice experience in comparison with the time and money spent and with theiropinions about the adequacy level of services may lead to a greater diagnosticability.

AN ALTERNATIVE ASSESSMENT FRAMEWORK

As a result of the ever-intensifying global competitiveness, managers of tour-ism and hospitality establishments are increasingly in need of comparative infor-mation to identify the relative internal and external product strengths and weak-nesses, which is critical in formulating competitive actions against competitors.Competitive gaps may occur between an organization or destination’s service/product performance and another organization offering the same product or ser-vice. Organizations should strengthen their competitive edge by surpassing theservice performance of other organizations. It is highly likely that organizations,as well as destinations, that fail to pay attention to the performance delivered byrivals, do not know their relative shortcomings, and overlook the improvement ofcustomer experience and satisfaction stand to lose their market share (Yüksel,2000).

Comparison against other brands in the product class may be more relevant tosatisfaction and behavioral intention judgments than are ordinary predictions ofservice attribute performance (Cadotte et al., 1987; Oliver, 1997). The perceivedperformance of one product/service may affect evaluation judgments of another(Gardial, Woodruff, Burns, Schumann, & Clemons, 1993). Thus, designinginstruments that could capture customers’ perceptions of a given organization’sinternal and external performance is an essential means of pinpointing where anorganization stands on service performance relative to its competitors. It wouldalso enhance understanding of how competitive advantages can be gained.

Customer experiences with alternative products and/or services and their per-ceptions of what competition has to offer may affect evaluation judgments of thefocal product/service. Thus, a Relative Performance Assessment model (RPA) isproposed to assess customer perceptions of a given organization’s internal aswell as relative (external) performance. The suggested RPA model involves threemain stages: (a) identification of significant service attributes to customer satis-faction and repeat business (those attributes that are most valued by customers),(b) assessment of internal performance delivered on these key areas, and(c) assessment of relative performance (external).

As questionnaire-based surveys that analyze both internal and relative perfor-mance perceptions with a large number of service attributes questions may sufferfrom respondent fatigue, development of a shorter and straightforward survey




instrument is instrumental. Kreck (1998) has recently proposed a respondent-centered approach in assessing service quality. Kreck’s method includes two cat-egories of questions: (a) statement and ranking of the three most important ser-vice attributes to the respondent, and (b) assessment of performance delivered onthese attributes. A major strength of this approach is that it allows respondents tofreely state service attributes significant to them instead of being restricted toanswer to a preestablished set of attribute questions determined by researchers.

The diagnosticity of this instrument may be strengthened by the inclusion ofan additional set of questions dealing with the relative performance perceptions(see appendix). For instance, respondents can be instructed to rank the three mostimportant attributes that immediately come to their minds when staying at a hotelor dining at a restaurant. Respondents then can be asked to evaluate the perfor-mance of those attributes that they ranked on a numerical scale. Finally, respon-dents can be instructed to compare the performance of these attributes with theirlast used or best-brand experience. This consolidated method can enable manag-ers to acquire critical insights into the key service factors valued by the customers.It also helps assess the efficiency and effectiveness of the internal process andprocedures in meeting customers’ requirements (performance relative to custom-ers’ internal desires, values), and the competitive gaps that may exist betweentheir organizations and their competitors (performance relative to a competitor).Managers are suggested to incorporate the relative performance concept intotheir periodic customer satisfaction investigations because any information with-out relative performance is incomplete. The information obtained through theRPA would be useful in formulating competitive actions.

Finally, although everyone knows what satisfaction is, it clearly does notalways mean the same thing to everyone (Oliver, 1997). On one hand, it maymean minimum acceptability to some customers, whereas on the other hand itmay mean near perfection to others. Depending on the situation, consumer satis-faction may (or may not) result from a complex process involving simultaneousinteractions with more than one comparison standard, and it may be determinedby exogenous and interpersonal elements that are beyond management. As Oliver(1997) puts it, the work of the consumer’s mind is like a black box, an observercan see only what goes in and what comes out, not what occurs inside. This com-plex and variable nature of satisfaction suggests that development of a universalmeasurement framework that can fully capture the diverse elements of the con-sumer satisfaction in every consumption situation may not be possible. A com-bined approach, capitalizing on the potential strengths of other alternative satis-faction assessment frameworks (e.g., the Equity Theory, the Comparison LevelTheory, the Value-Percept Theory), however, may provide a richer picture of cus-tomer satisfaction in situations that may not be gained using the EDP alone.




APPENDIXSurvey Instrument

Q1.a Based on your experience as a user of hotels, what are the three most important ser-vice attributes that come to your mind immediately when using lodging facilities?

Most important attribute: _____________________________________Second important attribute: _____________________________________Third important attribute: _____________________________________

Q2.a Based on your current visit, how would you rate the hotel’s efforts in which youstayed on the three service attributes that are important to you?

Poor ExcellentFirst attribute: 1 2 3 4 5 6 7Second attribute: 1 2 3 4 5 6 7Third attribute: 1 2 3 4 5 6 7

Q3. Think about the best hotel stay you have experienced. Based on your best stay experi-ence, how would you rate this hotel in comparison to service characteristics that are impor-tant to you?

Better WorseFirst attribute: 1 2 3 4 5 6 7Second attribute: 1 2 3 4 5 6 7Third attribute: 1 2 3 4 5 6 7

Other comments:

Please state the name of the hotel where you have had your best stay experience:

[Demographic detail question]

[Behavioral intention questions]

a. Developed from Kreck (1998).

REFERENCES

Adams, S. J. (1963). Toward an understanding of inequity. Journal of Abnormal and SocialPsychology, 67, 422-436.

Anderson, E. R. (1973). Consumer dissatisfaction: The effect of disconfirmed expectancyon perceived product performance. Journal of Marketing Research, 10, 38-44.

Babakus, E., & Boller, W. G. (1992). An empirical assessment of the Servqual scale. Jour-nal of Business Research, 24, 253-268.




Barsky, J. D. (1992). Customer satisfaction in hotel industry: Meaning and measurement.Hospitality Research Journal, 16, 51-73.

Barsky, J. D., & Labagh, R. (1992, October). A strategy for customer satisfaction. The Cor-nell Hotel and Restaurant Administration Quarterly, 32-40.

Bojanic, D., & Rosen, D. L. (1995). Measuring service quality in restaurants: An applica-tion of the Servqual instrument. Hospitality Research Journal, 18(1), 3-14.

Botterill, T. D. (1987). Dissatisfaction with a construction of satisfaction. Annals of Tour-ism Research, 14, 139-141.

Boulding, W., Karla, A., Staelin, R., & Zeithaml, V. A. (1993, February). A dynamic pro-cess model of service quality, from expectations to behavioural intentions. Journal ofMarketing Research, 30, 7-27.

Buttle, A. F. (1995). What future for Servqual. Proceedings, 24th EMAC Conference,Essec, 211-231.

Cadotte, E. R., Woodruff, R. B., & Jenkins, R. J. (1987, August). Expectations and normsin models of consumer satisfaction. Journal of Marketing Research, 24, 305-314.

Carman, J. M. (1990, Spring). Consumer perceptions of service quality: An assessment ofthe Servqual dimensions. Journal of Retailing, 66, 35-55.

Cho, B. H. (1998). Assessing tourist satisfaction: An exploratory study of Korean youthtourists in Australia. Tourism Recreation Research, 23(1), 47-54.

Chon, K. (1992). The role of destination image in tourism, an extension. Tourist Review,4(7), 2-8.

Chon, K., Christianson, J. D., & Chin-Lin, L. (1998). Modelling tourist satisfaction: Japa-nese tourists’ evaluation of hotel stay experience in Taiwan. Australian Journal of Hos-pitality Management, 2(1), 1-6.

Chon, K., & Olsen, M. D. (1991). Functional and symbolic congruity approaches to con-sumer satisfaction/ dissatisfaction in consumerism. Journal of the International Acad-emy of Marketing Research, 1, 2-23.

Churchill, G. R., & Surprenant, C. (1982). An investigation into determinants of customersatisfaction. Journal of Marketing Research, 19, 491-504.

Cook, D. T., & Campbell, T. D. (1979). Quasi experimentation: Design, analysis issues forfield settings. Boston: Houghton Mifflin.

Cote, J. A., Foxman, E. R., & Cutler, B. D. (1989). Selecting an appropriate standard ofcomparison for post-purchase evaluations. Advances in Consumer Research, 16,502-506.

Crompton, L. J., & Love, L. L. (1995, Summer). The predictive validity of alternativeapproaches to evaluating quality of a festival. Journal of Travel Research, 34(1), 11-25.

Danaher, P. J., & Haddrell, V. (1996). A comparison of question scales used for measuringcustomer satisfaction. International Journal of Service Industry Management, 17(4),4-26.

Danaher, P. J., & Mattsson, J. (1994). Cumulative encounter satisfaction in the hotel con-ference process. International Journal of Service Industry Management, 5(4), 69-80.

Day, R. (1977, April). Consumer satisfaction, dissatisfaction and complaining behaviour.In R. Day (Ed.), Symposium proceedings. Bloomington: University of Indiana.

Dorfman, P. W. (1979). Measurement and meaning of recreation satisfaction: A case studyin camping. Environment and Behaviour, 11(4), 483-510.

Ennew, T. C., Reed, V. G., & Binks, R. M. (1993). Importance-performance analysis andthe measurement of service quality. European Journal of Marketing, 27(2), 59-70.




Erevelles, S., & Leavitt, C. (1992). A comparison of current models of consumer satisfaction/dissatisfaction. Journal of Consumer Satisfaction/Dissatisfaction and ComplainingBehaviour, 5, 104-114.

Fick, G. R., & Ritchie, B.J.R. (1991, Fall). Measuring service quality in the travel and tour-ism industry. Journal of Travel Research, 30, 2-9.

Fisk, R. P., & Coney, A. K. (1982). Post-choice evaluation: An equity theory analysis ofconsumer satisfaction and dissatisfaction with service choices. In H. K. Hunt & L. R.Day (Eds.), Conceptual and empirical contributions to consumer satisfaction and dis-satisfaction and complaining behaviour (pp. 9-16). Bloomington: Indiana UniversitySchool of Business.

Gardial, F. S., Woodruff, B. R., Burns, M. J., Schumann, W. D., & Clemons, S. (1993).Comparison standards: Exploring their variety and the circumstances surrounding theiruse. Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behaviour, 6,63-73.

Getty, M. J., & Thomson, N. K. (1994). The relationship between quality, satisfaction, andrecommending behaviour in lodging decisions. Journal of Hospitality and LeisureMarketing, 2(3), 3-22.

Gronroos, C. (1993). Toward a third phase in service quality research: Challenges anddirections. In T. A. Swart, D. E. Bowen, & S. W. Brown (Eds.), Advances in servicemarketing and management (2nd ed., pp. 49-64). Greenwich, CT: JAI.

Gupta, K., & Stewart, D. W. (1996). Customer satisfaction and customer behaviour, the dif-ferential role of brand and category expectations. Marketing Letters, 7(3), 249-263.

Halstead, D. (1993). Exploring the concept of retrieved expectations. Journal of ConsumerSatisfaction, Dissatisfaction and Complaining Behaviour, 6, 56-62.

Halstead, D., Hartman, D., & Schmidt, L. S. (1994). Multi source effects on the satisfactionformation process. Journal of the Academy of Marketing Science, 22(2), 114-129.

Hill, D. J. (1985). Satisfaction and consumer services. Advances in Consumer Satisfaction,13, 311-315.

Howard, A. J., & Sheth, N. J. (1969). The theory of buyer behaviour (p. 147). New York:Wiley.

Hughes, K. (1991). Tourist satisfaction: A guided tour in North Queensland. AustralianPsychologist, 26(3), 166-171.

Iacobucci, D., Grayson, A. K., & Ostrom, A. L. (1994). The calculus of service quality andcustomer satisfaction: Theoretical and empirical differentiation and integration. InT. A. Swart, D. E. Bowen, & S. W. Brown (Eds.), Advances in service marketing andmanagement (3rd ed., pp. 1-67). Greenwich, CT: JAI.

Jayanti, R., & Jackson, A. (1991). Service satisfaction: An exploratory investigation ofthree models. Advances in Consumer Research, 18, 603-610.

Johns, N., & Tyas, P. (1996). Use of service quality gap theory to differentiate betweenfoodservice outlets. The Service Industries Journal, 16(3), 321-346.

Kreck, A. L. (1998). Service quality: Who should determine it? Research and practice.Journal of International Hospitality, Leisure and Tourism Management, 1(4), 63-77.

LaTour, S. T., & Peat, N. C. (1979). Conceptual and methodological issues in consumersatisfaction research. Advances in Consumer Research, 6, 431-437.

Lewis, R. C., Chambers, E. R., & Chacko, E. H. (1995). Marketing leadership in hospital-ity. New York: Van Nostrand Reinhold.




Llosa, S., Chandon, J., & Orsingher, C. (1998). An empirical study of Sevqual’s dimen-sionality. The Service Industries Journal, 18(2), 16-44.

Loundsbury, L. W., & Hoopes, L. L. (1985). An investigation of factors associated withvacation satisfaction. Journal of Leisure Research, 17, 1-13.

Mazursky, D. (1989). Past experience and future tourism decisions. Annals of TourismResearch, 16, 333-344.

McGill, L. A., & Iacobucci, D. (1992). The role of post-experience comparison standardsin the evaluation of unfamiliar services. Advances in Consumer Research, 19, 570-578.

Meyer, A., & Westerbarkey, P. (1996). Measuring and managing hotel guest satisfaction.In D. M. Olsen, R. Teare, & E. Gummesson (Eds.), Service quality in hospitalityorganisations (pp. 185-204). New York: Cassell.

Miller, J. A. (1977). Studying satisfaction, modifying models, eliciting expectations, pos-ing problems, and making meaningful measurements. In H. K. Hunt (Ed.), Concep-tualisation and measurement of consumer satisfaction and dissatisfaction (pp. 72-92).Cambridge, MA: Marketing Science Institute.

Oh, H., & Parks, C. S. (1997). Customer satisfaction and service quality: A critical reviewof the literature and research implications for the hospitality industry. HospitalityResearch Journal, 20(3), 36-64.

Olander, F. (1979). Consumer satisfaction: A sceptic’s view. Aarhus, Denmark.

Oliver, R. L. (1980). A cognitive model of the antecedents of satisfaction decisions. Jour-nal of Marketing Research, 17, 46-49.

Oliver, R. L. (1989). Processing of the satisfaction response in consumption: A suggestedframework and research propositions. Journal of Consumer Satisfaction, Dissatisfac-tion and Complaining Behaviour, 2, 1-16.

Oliver, R. L. (1997). Satisfaction: A behavioural perspective on the consumer. New York:McGraw-Hill.

Oliver, R. L., & DeSarbo, W. S. (1988). Response determinants in satisfaction judgement.Journal of Consumer Research, 14, 495-507.

Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1988, Spring). Servqual: A multiple itemscale for measuring consumer perceptions of service quality. Journal of Retailing, 64,12-40.

Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1991). Refinement and reassessment ofthe Servqual scale. Journal of Retailing, 67(4), 420-450.

Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1994). Reassessment of expectations as acomparison standard in measuring service quality: Implications for further research.Journal of Marketing, 58, 111-124.

Pearce, P. L. (1980, Summer). A favorability-satisfaction model of tourists’ evaluations.Journal of Travel Research, 19, 13-17.

Pearce, P. L. (1991). Introduction the tourism psychology. Australian Psychologist, 26(3),145-146.

Pizam, A., & Milman, A. (1993). Predicting satisfaction among first time visitors to a des-tination by using the Expectancy Disconfirmation theory. International Journal ofHospitality Management, 12, 197-209.

Prakash, V., & Loundsbury, W. J. (1992). A reliability problem in the measurement ofdisconfirmation of expectations. In P. R. Bagozzi & M. A. Tybout (Eds.), Advances inconsumer research (Vol. 10, pp. 244-249).




Reisinger, Y., & Waryszak, R. (1996). Catering to Japanese tourists: What service do theyexpect from food and drinking establishments in Australia? Journal of Restaurant andFoodservice Marketing, 1(3/4), 53-71.

Saleh, F., & Ryan, C. (1991). Analysing service quality in the hospitality industry using theServqual model. Service Industries Journal, 11(3), 324-345.

Sirgy, J. M. (1984). A social cognition model of CS/D: An experiment. Psychology andMarketing, 1, 27-44.

Smith, M. A. (1995). Measuring service quality: Is Servqual now redundant? Journal ofMarketing Management, 11, 257-276.

Spreng, R. A., & Dixon, A. L. (1992). Alternative comparison standards in the formation ofsatisfaction/dissatisfaction. In Leone & Kumar (Eds.), Enhancing knowledge develop-ment, marketing (pp. 85-91). Chicago: American Marketing Association.

Spreng, R. A., Mackenzie, S. B., & Olshavsky, R. W. (1996). A re-examination of the deter-minants of consumer satisfaction. Journal of Marketing, 60, 15-32.

Spreng, R. A., & Olshavsky, W. (1993). A desires congruency model of consumer satisfac-tion. Journal of the Academy of Marketing Science, 21, 169-177.

Swan, J., & Trawick, F. I. (1980). Satisfaction related to the predictive vs. desired expecta-tions. In H. K. Hunt & R. Day (Eds.), Refining concepts and measures of consumer sat-isfaction and complaining behaviour (pp. 7-12). Indiana University.

Teas, R. K. (1993). Expectations, performance evaluation, and consumers’ perception ofquality. Journal of Marketing, 57, 18-34.

Teas, R. K. (1994). Expectations as a comparison standard in measuring service quality:An assessment of a reassessment. Journal of Marketing, 58, 132-139.

Tribe, J., & Snaith, T. (1998). From Servqual to Holsat: Holiday satisfaction in Varadero,Cuba. Tourism Management, 19, 125-134.

Tse, D. K., & Wilton, C. P. (1988). Models of customer satisfaction formation: An exten-sion. Journal of Marketing Research, 25, 204-212.

Weber, K. (1997). Assessment of tourist satisfaction, using the Expectancy Discon-firmation theory, a study of German travel market in Australia. Pacific Tourism Review,1, 35-45.

Westbrook, R. A., & Newman, W. (1987, August). An analysis of shopper dissatisfactionfor major household appliances. Journal of Marketing Research, 15, 456-466.

Whipple, W. T., & Thach, V. S. (1988, Fall). Group tour management: Does good serviceproduce satisfied customers. Journal of Travel Research, 28, 16-21.

Williams, C. (1998). Is the Servqual model an appropriate management tool for measuringservice delivery quality in the UK leisure industry. Managing Leisure, 3, 98-110.

Woodruff, R. B., Cleanop, D. S., Schumn, D. W., Godial, S. F., & Burns, M. J. (1991). Thestandards issue in CS/D research: Historical perspective. Journal of Consumer Satis-faction/Dissatisfaction and Complaining Behaviour, 4, 103-109.

Woodruff, R. B., Ernest, R. C., & Jenkins, R. L. (1983, August). Modelling consumer satis-faction processes using experience-based norms. Journal of Marketing Research, 20,296-304.

Yi, Y. (1990). A critical review of consumer satisfaction. In V. A. Zeithaml (Ed.), Review ofmarketing (pp. 68-123). Chicago: American Marketing Association.

Yüksel, A. (2000, January). Determinants of tourist satisfaction: Findings from Turkey.Paper presented at the 5th Graduate Research Student Conference, University of Hous-ton, TX.




Yüksel, A., & Rimmington, M. (1998, December). Customer satisfaction measurement:Performance counts. The Cornell Hotel and Restaurant Administration Quarterly, 39,60-70.

Zeithaml, V. A. (1981). How consumer evaluation processes differ between goods and ser-vices. In J. H. Donnelly & W. R. George (Eds.), Advances in consumer research(pp. 186-190). Chicago: American Marketing Association.

Zwick, R., Pieters, R., & Baumgartner, H. (1995). On the practical significance of hind-sight bias, the case of Expectancy Disconfirmation Model of customer satisfaction.Organisational Behaviour and Human Decision Process, 64(1), 103-117.

Submitted May 20, 1999First Revision Submitted February 14, 2000Second Revision Submitted May 8, 2000Accepted June 1, 2000Refereed Anonymously

Atila Yüksel, Ph.D. (e-mail: [email protected]), is a lecturer at Adnan Menderes Uni-versity, Turkey (Turizm Isl. Otel. Y.O., Candan Tarhan Bulvari, No. 6, Kusadasi-Aydin,09400 Turkey) and Fisun Yüksel (e-mail: [email protected]) is a Ph.D. candidate atSheffield Hallam University.




yuksel (2001) the expectancy disconfirmation paradigm

Documents

equity theory

theattribution theory

dissonance theory

valuepercept theory

importanceperformance

consumer satisfaction

comparison level theory

hospitality services