An empirical examination of a model of perceived service quality and satisfaction

An empirical examination of a model of perceived service quality and satisfaction

An Empirical Examination of a Model of Perceived Service Quality and Satisfaction RICHARD A. SPRENC Michigan State University ROBERT D. Butler Unive...

1020KB Sizes 22 Downloads 81 Views

An Empirical Examination of a Model of Perceived Service Quality and Satisfaction

RICHARD A. SPRENC Michigan State University

ROBERT D. Butler University

MACKOY

Perceived service quality and satisfaction have generally been conceptualized but there isn’t a good understanding tle research has empirically

of their relationship.

examined the distinction.

to be distinct constructs,

While rhe two conswuc[s are very similar, lit-

This study discusses the conceprual arguments

for the distinction, and examines the empirical distinction by testing a recently proposed model of service quality and satisfaction.

Results indicate that the hvo constructs are, in the present case, distinct.

and there is some support for the model, with several modifications.

In recent years there has been a great deal of interest in the conceptualization and measurement of consumer satisfaction and perceived service quality, by both managers and academic researchers. Consumer satisfaction and service quality have each been the subject of extensive, but separate, research, although many studies of consumer satisfaction have been conducted in service settings (Cadotte, Woodruff and Jenkins, 1987; Fomell, 1992; Oliver, 1980; Oliver and DeSarbo, 1988; Oliver and Swan, 1989; Swan, 1988; Swan and Trawick, 1980). There seems to be a great deal of similarity between these two concepts, yet researchers are usually careful to state that these are different constructs (for exceptions see Dabholkar, 1993; Oliva, Oliver, and MacMillan, 1992; Spreng and Singh, 1993). For example, Cronin and Taylor (1992, p. 56) state that “This distinction is important to managers and researchers alike because service providers need to know whether their objective should be to have consumers who are ‘satisfied’ with their performance or to deliver the maximum level of ‘perceived service quality.“’ Most researchers in the services domain have maintained that these two constructs are distinct (Bitner, 1990; Boulding, Kalra, Staelin and Zeithaml, 1993; Carman, 1990; Parasuraman, Zeithaml and Berry, 1988; Taylor and

Richard A. Spreng, Michigan State University, Broad Graduate School of Management. Department of Marketing and Logistics, East Lansing, MI 48824. Robert D. Mackay, Butler University, Department of Marketing, Indianapolis, IN 46208. Journal of Retailing, Volume 72(2), pp. 201-214, ISSN: 0022-4359 Copyright 0 1996 by New York University. All rights of reproduction

201

in any form reserved.

202

Journal

of Retailing Vol. 72, No. 2 1996

Baker, 1994), yet there have been repeated calls for research investigating the relationship between the two constructs (Anderson and Fomell, 1994; Rust and Oliver, 1994). Greater understanding of the relationship between perceived service quality and satisfaction is needed. If they are distinct constructs as has generally been claimed, then we need to understand how they are different. If they are not distinct, then we don’t have to waste time on surveys asking for both or confuse managers by telling them they have to be concerned with both. Further, researchers in the service quality area continue to state that satisfaction is the result of a comparison with predictive expectations (Bolton and Drew, 1991; Bitner, 1990; Parasuraman et al., 1988, 1994; Oliver, 1993; Rust and Oliver, 1994; Zeithaml, Berry and Parasuraman, 1993), despite many criticisms of the disconfirmation of expectations model (Barbeau, 1985; Cadotte et al., 1987; LaTour and Peat, 1979; Spreng and Olshavsky, 1993; Westbrook and Reilly, 1983). If satisfaction is not simply a result of meeting expectations, then managers may not be focusing on the correct things. Closely related to this are the recommendations to manage expectations. Those who advocate this usually mean that managers should attempt to lower expectations so that it will be possible to provide service that is “better than expected” which will then produce higher satisfaction (Davidow and Uttal, 1989; Peters, 1987). If this is not a complete picture of how expectations influence satisfaction, then it may be that managers are being urged to do the wrong things. This paper will extend this research by empirically testing a new model of service quality and satisfaction. Specifically, we will first discuss one of the conceptual distinctions made in past research between service quality and satisfaction. We then test a model by Oliver (1993) that proposes to integrate satisfaction and perceived service quality.

PERCEIVED SERVICE QUALITY

AND SATISFACTION

Perceived service quality is defined by Parasuraman et al. (1988) as “a global judgment, or attitude, relating to the superiority of the service,” and many researchers in the service quality literature concur with this definition (Boulding et al., 1993; Bolton and Drew, 1991; Cronin and Taylor, 1992). The definition of consumer satisfaction is not so clear (see Yi, 1990 for a review). While there is not a clear consensus regarding the definition of satisfaction, most definitions would involve “an evaluative, affective, or emotional response” (Oliver, 1989, p., 1). Both satisfaction and service quality literatures have emphasized the idea that consumers make a comparison between the performance of the product or service and some standard. The service quality literature has maintained that the distinction between perceived service quality and satisfaction is that they use different standards of comparison (Bitner, 1990; Parasuraman et al., 1988; Zeithaml et al., 1993). These authors have argued that the standard of comparison in forming satisfaction is predictive expectations, or what the consumer believes wiZEhappen, while perceived service quality is the result of a comparison of performance and what the consumer feels a firm should provide. In a conceptual model that attempts to integrate service quality and satisfaction, Oliver (1993) similarly argues that while the antecedent of quality perceptions is the disconfirmation of ideals, the antecedents

An Empirical Examination of a Model of Perceived Service Quality and Satisfaction

203

of satisfaction are disconfirmation of predictive expectations (regarding quality and nonquality dimensions), and perceived quality. Oliver suggests that the constructs are distinct, in part, because they use different standards. That is, he claims that one can be satisfied with low quality if the performance is better than one’s prediction of the performance. However, Oliver (1993, p. 81) states that “... verification of the use of ideal expectations for quality and predictive expectations for satisfaction is needed.” Research in consumer satisfaction has extensively examined the issue of the standard against which performance is compared in producing satisfaction. While the construct of predictive expectations has dominated, a number of other standards have been suggested and tested, such as experience-based norms, ideals, values, desires, equity, and others (for reviews see Spreng and Dixon, 1992 and Woodruff, Clemons, Schumann, Gardial and Burns, 1991).

Oliver’s

Satisfaction/Service

Quaiity

Model

Oliver (1993) has proposed a model that is intended to integrate the satisfaction and the service quality literatures (see Figure 1 for a modified version of this model). He proposes that while service quality is formed by a ~ompa~son between ideals and perceptions of performance regarding quality dimensions, satisfaction is a function of the disconfkmation of

Disconfirmation

Figure 7.

Modified

Satisfaction-Service

Quality

Model

204

Journal of Retailing Vol. 72, No. 2 1996

predictive expectations regarding both quality dimensions and non-quality dimensions. Further, perceived service quality is proposed to be an antecedent to satisfaction. There are several limitations to this model. First, Oliver’s model suggests that satisfaction is not related to disconfirmation of ideals, except through service quality perceptions. However, there is a fair amount of evidence in the satisfaction literature that ideals or desires are an important antecedent to satisfaction (Barbeau, 1985; Cadotte et al., 1987; Spreng and Olshavsky, 1993; Swan and Trawick, 1980; Westbrook and Reilly, 1983). Second, Oliver’s model specifies that expectations do not influence perceptions of performance, despite an extensive literature showing this effect (for a review, see Yi, 1990, pp. 83-84). In order to investigate these issues, a study was undertaken that would be able to unambiguously assess several key aspects of the perceived service quality-satisfaction relationship. The purposes of this study are to assess the distinction between perceived service quality and satisfaction, and examine the impact of different standards of comparison. Desires were used to operationalize the standard of comparison that Oliver portrays in his model as “Ideals.” The service quality literature has used the term “desires” in describing their standard (Parasuraman et al., 1988), and this standard has had some support in the satisfaction literature (Barbeau, 1985; Spreng and Olshavsky, 1993).

METHOD

Setting and Subjects

The study examined student assessment of undergraduate advising, and subjects received extra credit in an undergraduate marketing class for participating. A total of 273 usable responses were obtained. The students in the study were all Juniors (36%) or Seniors (64%), and thus had at least one previous year of personal experience with the undergraduate advising office. Most students (95%) had had at least one previous appointment in the last year. While this is clearly a service setting, some may argue that there is not an exchange taking place. This is incorrect, since students view the services they receive as something they pay for through their tuition fees. This would be similar to a situation in which someone sees a physician, but the actual payment for the service comes from money the patient has paid to an insurance company. Further, there is ad~tional sacrifice by the students in terms of time and effort in seeing an advisor. Finally, if the level of service is poor (e.g., students get wrong or incomplete advice) the service failure can have extremely adverse consequences to the student (e.g., the student may not graduate on time, may waste money taking the wrong classes, etc.). Thus, we believe that the elements necessary for developing real reactions to a service are present in this situation. Subjects filled out a pre-experience questionnaire immediateIy before an ap~in~ent with an advisor. To prevent advisors from knowing that they were being evaluated, these questionnaires were filled out in offices that were in another part of the building from the

An Empirical Examination of a Model of Perceived Service Quality and Satisfaction

205

advising offices. After completing the pre-experience questionnaire, subjects went to their appointment, and then immediately returned to fill out the post-experience questionnaire. Our procedure has some advantages over many past studies of perceived service quality. First, we have measures of pre-experience constructs that were taken before the experience. Second, the subjects had a real, naturally occurring service experience, with perceptions collected both before and after the service encounter.

Measures

Desires, predictive expectations, perceived performance, desires congruency and expectations disconfirmation were each measured for ten attributes. These attributes were developed from reviewing past research in the academic advising literature, discussions with the director of the undergraduate advising office, and two focus groups with undergraduate students. Previous administration of the questionnaire permitted the measures to be refined, and this resulted in a set of 10 attributes of advising (convenience in making an appointment, friendliness of the staff, advisor listened to my questions, the advisor provided accurate information, the knowledge of the advisor, the advice was consistent, advisor helped in long range planning, the advisor helped in choosing the right courses for career, advisor was interested in personal life, and the offices were professional). The pre-experience questionnaire asked about students’ desires and expectations regarding the 10 attributes. The desires questions asked for the level of service the student wants to receive, while the expectations questions asked about the level of service the student believed they would actually receive, each on seven-point Likert scales (see Appendix). After their advising appointment, subjects filled out the post-experience questionnaire. Perceived performance was measured by asking for the student’s perception of their advising experience, and used seven-point Likert scales for the 10 attributes. For all three of these measures (Desires, Expectations, and Perceive Performance) the average over the ten scales was the measure used in the analysis described below. There has been a great deal of criticism of gap scores as measures of disconfirmation constructs (Teas, 1993), although not all researchers agree that difference scores are problematic, and in fact can provide more diagnostic information (Parasuraman et al., 1994). In order to measure desires congruency and expectations disconfirmation in a way that avoids some of the problems with past measures, they were both measured in a new way. To measure desires congruency for each attribute the subjects were asked for their subjective assessment of the “...difference between what you desired and what you received” on seven-point scales anchored, 1 = “Exactly as I desired,” and 7 = “Extremely different than I desired,” with “Somewhat different than I desired’ as the midpoint. This asked the subject to assess how close the product was to what was desired. Following this scale for each attribute was an eleven-point scale that asked “How good or bad was this difference?’ with -5 labeled as “Very bad,” +5 labeled “Very good,” and 0 labeled as “Neither good nor bad.” Desires congruency was operationalized by multiplying these two scales for each attribute. The ten products were then averaged to get an “Average Desires Congruency” score that was used in the analysis. Thus, these measures represent a belief regarding the

206

Journal of Retailing

Vol. 72, No. 2 1996

degree to which the attribute matched the subject’s desires and an evaluation of this belief. See Appendix for an example of these measures. Expectations congruency was operationalized in a similar way. Subjects were first asked to assess the difference between what they anticipated they would receive and what they actually received, and then to evaluate this difference. The two scores for each attribute were multiplied for each attribute. The ten products were then averaged to get an “Average Expectations Congruency” score that was used in the analysis. Overall satisfaction was measured by asking “Overall, how do you feel about the advising services you just received?’ Following this stem there were four seven-point scales anchored by “Very dissatisfied/very satisfied,” “Terrible/delighted,” “Very dissatisfied/not at all dissatisfied,” and “Not at all satisfied/very satisfied.” These are all widely used scales for measuring the level of satisfaction and dissatisfaction. Overall service quality was measured with respect to a long term, overall evaluation, and asked “Overall, what is the level of service quality you receive from advising services?’ Following this stem were three seven-point scales anchored by “Extremely poor/extremely and “Very low/very high.” Table 1 shows the correlations, stangood, ” “Awful/excellent,” dard deviations, and means of the variables.

RESULTS

Measurement

Results

A confirmatory factor analysis model of the constructs was estimated using LISRELS (Jbreskog and S&born, 1993). The fit of the model was good: chi-square = 45.00, 38 d.f. (p = .20), GFI = .97, AGFI = .95, CFI = 1 .OO(Table 2). All indicator factor loadings were significant, and the squared multiple correlations ranged from .7 1 to .93, indicating acceptable reliability. The average variance extracted (Fomell and Larcker, 198 1) for the satisfaction construct was .78 and for the overall service quality construct was .92. The average variance extracted (AVE) can also be used to evaluate discriminant validity (Fomell and Larcker, 1981). The AVE for each construct should be greater than the squared correlation between the construct and any other construct. Examining the correlations between the satisfaction and the perceived service quality constructs from Table 3 it is clear that this test holds, since the squared correlation between these two constructs is .53 (.732). Further evidence of the discriminant validity comes from an examination of the standardized residuals. There were no standardized residuals with an absolute value greater than 3.0. Only three out of 78 (4%) of the standardized residuals were over 2.0, and the largest positive residual was 2.99 and the largest negative residual was -2.82. Finally, note from the correlation matrix of the individual items in Table 1 that there are no measures of satisfaction that have a correlation with a measure of another construct that is higher than its correlation with the other satisfaction items. Likewise, none of the perceived service quality measures is more highly correlated with another measure than it is with the other perceived service quality measures. Therefore, measures of latent variables of Overall Satisfaction and Perceived Service Quality were used in structural equation model.

1

4.00 1.08

4.37 1.17

5.93 $89

1.oo .38 .I7 .27 .31 .30 .36 .51 .51 -51

l.ocl .55 .70 .61 .65 .64 .6l .67 .62 .62 .63

1 .oo .13 .23 -.I3 .oo -.ov -.06 -.ov -.05 -.03 -.04 -.06

SatisfactionZ:“Terrible/delighted.” Satisfaction3:“Very dissatisfied/notat all dissatisfied.” Satisfaction4:“Not at all satisfied/verysatisfied.” Perceived Service Quality 1: “Extremely poor/extremely good PerceivedService Quality 2: “Awful/excellent.” Perceived Service Quality 3: “Very low/very high.”

3.

2.

7.

-2.04 7.12

.73 .61 SV .59 .66 .58 58 .59

1.00

4.

1.52 6.13

1.oo 57 .!i6 51 .58 .41 .42 .43

5.

5.00 1.55

1 .OO .?8 .79 .83 .61 .59 .60

6.

4.70 1.19

1.00 .69 .80 .64 .64 .66

7.

5.30 1.51

1 .oo .77 .61 .59 .60

8.

Correlations, Means and Standard Deviations for Measures

Satisfartionl: ‘Thy dissatisfied/verysatisfied.”

Desires Perceived Performance Expectations Desires Congruency Expectations Congruency Satisfaction1 Satisfaction2 Satisfaction3 Satisfaction4 Perceived SQl Perceived SQ2 Perceived SQ3

Mean SD

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

TABLE

5.00 1.62

1.00 .65 .62 .64

9.

4.51 1.47

1 .oo .92 .93

10.

4.46 1.36

1 .oo .91

I?.

4.46 1.45

1.00

12.

208

of

Journal

Retailing

Vol.

72,

No.

2 1996

TABLE2 Measurement

Model

Parameters

for Constructs

Completely

Average

Standardized Construct

Variance

Coefficient

Measure

t-value

Extracted

1 .oo

Desires’

Desires Congruency’

1 .oo

Expectations

1 .oo

-

SAT1

.91

-

SAT2

.86

20.88

SAT3

.84

19.99

SAT4

.92

24.28

PSQI

.97

-

Perceived

1 .oo

Performance’

1 .oo

Expectations Disconfirmation’

Satisfaction

Perceived

Notes:

.78

.92

Service Quality PSQ2

.95

37.5

PSQ3

.96

41.02

1. Construct used an average of 10 attributes as a single item mesure.

TABLE3 Factor Intercorrelations Factor lntercorrelations

1.

1. Desires

3.

4.

5.

6.

7.

1 .oo

2. Perceived

Performance

3. Expectations 4. Desires Congruency 5. Expectations

.13

1 .oo

.23

.55

1.00

.70

.38

1 .oo

.61

.17

.73

-.08

.73

.35

.70

.63

1 .oo

-.05

.65

.53

.61

.44

.73

-.13

Disconfirmation

6. Satisfaction 7. Perceived

2.

Service Quality

.oo

1.00 1 .oo

Table 3 shows the correlations between the constructs, and indicates that while the correlation between perceived service quality and satisfaction is high (.73), they do seem to be distinct. A model was estimated that constrained this parameter to be equal to 1.O, and the fit was significantly worse.

The Oliver

(1993)

Model

of Perceived

Service Quality

and Satisfaction

Figure 1 shows a modified version of the model proposed by Oliver (1993). Here, no “non-quality” dimensions are specified as determinants of satisfaction. Note that, in accordance with Oliver’s model, desires congruency is not specified to influence overall satis-

An Empirical Examination of

a Model of Perceived Service Quality and Satisfaction

209

faction, expectations disconfirmation is not specified to affect perceived service quality, and expectations are not specified to affect perceived performance. The fit of this model was only mediocre, with a chi-square of 276.66, 51 d.f. (pc.OOl), GFI = .86, AGFI = .79, and CFI = .93. When the relationship between expectations and perceived performance was allowed to be estimated, the fit improved greatly (chi-square = 177.13, 50 d.f., p<.OOl, GFI = .91, AGFI = .86, and CFI = .96). Thus, as we anticipated, expectations have a significant effect on perceptions of performance. Next, the path between desires congruency and overall satisfaction, and the path between expectations congruency and overall service quality were freed. The resulting decrease in chi-square was significant (change of 11.19, 2 d.f., p x.01). Desires congruency had a significant effect on satisfaction, while expectations disconfirmation did not affect overall service quality. Figure 2 shows the parameters of this model after the expectations disconfhmation-perceived service quality path was constrained to zero (chi-square = 164.94,49 d.f., GFI = .91, AGFI = .86, and CFl = .96). The amount of variation explained in overall satisfaction and in overall service quality is 67 percent and 37 percent respectively. While desires congruency and expectations disconfirmation have comparable direct effects on satisfaction, the total effects of desires congruency are stronger (standardized total effects of .52 (t = 7.87) than the total effects of expectations disconfirmation (standardized total effects of .26 (t = 3.94). This is due, of course, to the indirect effect of desires congruency on satisfaction through perceived service quality.

(10.96)

figure

Disconfwmation

2.

Final Model, Completely

Standardized

Parameters

(t-values)

210

Journal of Retailing Vol. 72, No. 2 1996

DISCUSSION

The results indicate that service quality and satisfaction, as operationalized here, are distinct. A confirmatory factor analysis showed that the two constructs were, in this case, distinct, and structural equations modeling showed that they have different antecedents. Our results also provide some support for Oliver’s model of service quality and satisfaction. As he proposed, the discon~~ation of expectations does not have a significant effect on perceived service quality. Contrary to his model, desires cong~ency does influence satisfaction. However, this result supports past research in satisfaction showing that desires congruency is an important antecedent of satisfaction (Barbeau, 1985; Spreng and Olshavsky, 1993; Swan and Trawick, 1980). Thus, our results also address Anderson and Fornell’s (1994) call for continued research on the antecedents of satisfaction. While the discon~~ation of expectations did not have a direct effect on service quality, this does not mean that expectations are unimportant in the formation of perceived service quality. The effect of expectations is indirect, through perceived performance (indirect standardized effects of .24, t = 7.16). Thus, the indirect effect of expectations on perceived service quality is positive. Conversely, the indirect effects of desires is negative (st~d~diz~ effect of -.12, c = -4.92), as there is not a si~~c~t relationship between desires and perceived performance. Therefore, as found by Boulding et al. (1993), the two types of standards (i.e., desires and predictive expectations) have opposite effects. The above results have a number of implications for research and management. First, further evidence is presented that desires congruency is an important antecedent of consumer satisfaction. The service quality literature continues to state that satisfaction is formed by a comparison with predictive expectations (Bolton and Drew, 1991; Bitner, 1990; Parasuraman et al., 1988, 1994; Oliver, 1993; Rust and Oliver, 1994; Zeithaml et al., 1993), despite a great deal of contrary evidence (Barbeau, 198.5; Cadotte et al., 1987; LaTour and Peat, 1979; Spreng and Olshavsky, 1993; Westbrook and Reilly, 1983). Since this is one of the ways perceived service quality and satisfaction are suppose to differ, it is important to recognize that predictive expectations are not the only antecedent of satisfaction. Thus, managers should not believe that merely meeting (or exceeding) predictive expectations wil1 satisfy consumers. Second, the results show the importance of expectations in influencing perceptions of performance. However, note that expectations have a negative effect on satisfaction, through ~scon~~ation, but a positive effect on both satisfaction and perceived service quality, through perceived performance. Most consultants or authors who advocate “managing expectations” mean that firms should seek to lower expectations so that it will be possible to provide service that is “better than expected” which will then produce higher satisfaction (Davidow and Uttal, 1989; Peters, 1987). Thus, these people are focusing on the negative impact of expectations on satisfaction through discon~~ation {i.e., lower ex~tations causes higher positive discon~~ation, which causes higher satisfaction). However, the positive effect of expectations on satisfaction through perceived performance is being ignored. By lowering expectations a firm might also lower perceptions of perfor-

An Empirical Examination of a Model of Perceived Service Quality and Satisfaction

211

mance, leading to lower levels of satisfaction. Thus, managing expectations down can result in lower, not higher satisfaction. It appears that managers must balance the positive and negative effects of expectations. That is, in terms of satisfaction, it appears that raising expectations can increase satisfaction through perceived performance, but at the same time lower satisfaction through disconfirmation. This indicates that it may be problematic to “manage expectations” as some suggest. The prescription regarding desires is more straightforward and intuitive: a key determinant of both satisfaction and service quality is meeting customers’ desires.

APPENDIX The desires and the predictive expectations questions were asked in a two-column format (see Parasuraman et al., 1994). This was deemed to be better than separate pages for each set of measures because the two-column method makes it clearer to respondents the distinction between predictive expectations and desires. The questionnaire was as follows: DIRECTIONS: Below are a series of statements each of the statements: l

l

about the advising center and the services offered. For

on the first scale, indicate how much you agree/disagree that this is the amount of service you would want to receive. on the second scale, indicate how much you agree/disagree that this is the amount of service you believe you will actually receive.

1. It will be extremely

convenient

to make an advising appointment.

The level of service

The level of service I believe

I WANT TO RECEIVE Strongly Agree 1

I WILL ACTUALLY Strongly Disagree

2

3

4

The desires congruency follows:

5

6

7

and the expectations

RECEIVE

Strongly Agree 1 congruency

Strongly Disagree 2

3

4

5

6

7

pages were similar, and were as

DIRECTIONS: The following questions are related to the difference between what you desire (wanted receive) from advising and how well you think the advising center met your des&es. In comparison to what you desired, how big was the difference between what you wanted, and what the advising center actually provided? Since this difference can be either good or

to

212

Journal

of Retailing Vol. 72, No. 2 1996

bad, following each question is a scale that asks you to rate how good or bad this difference is. Put a circle around the number that represents your answer. 2. The degree to which it was extremely

convenient

to make an advising appointment.

Difference between what I Desired and what I received Exactly as I desired

1

Somewhat different than I desired

2

3

4

How good or bad is this difference?

Extremely different than I desired

5

6

7

Neither bad nor good

Very bad

-5

-4

-3

-2

-1

0

Very good

1

2

3

4

s

REFERENCES Anderson, Eugene W. and Claes Fomell. (1994). “A Customer Satisfaction Research Prospectus.” Pp. 241-268 in Roland T. Rust and Richard L. Oliver (eds.), Service Quality: New Directions in Theory and Practice. Thousand Oaks, CA: Sage. Barbeau, 3. Bradley. (1985). “Predictive and Normative Expectations in Consumer Satisfaction: A Utilization of Adaptation and Comp~son Levels in a Unified Framework.” 4.27-32 in H. Keith Hunt and Ralph L. Day (eds.), Conceptual and Empirical Contributions to Consumer Satisfaction and Complaining Behavior. Bloomington, IN: Indiana University School of Business. Bitner, Mary Jo. (1990). “Evaluating Service Encounters: The Effects of Physical Surroundings and Employee Responses,” Journal of Marketing, 54(April): 69-82. Bolton, Ruth N. and James H. Drew. (1991). ‘A Longitudinal Analysis of the Impact of Service Changes on customer Attitudes,” Jounla~ of marketing, 55(Janu~}: 1-9. Boulding, William, Ajay Kaha, Richard Staelin and Valarie A. Zeitbaml. f1993). “A Dynamic Process Model of Service Quality: From Expectations to Behavioral Intentions,” Journal of Marketing Research, 30(February): 7-27. 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): 305-3 14. Carman, James M. (1990). “Consumer Perceptions of Service Quality: An Assessment of the SERVQUAL Dimensions,” Journal of Retailing, 66(Spring): 33-55. Cronin, J. Joseph, Jr. and Steven R. Taylor. (1992). “Measuring Service Quality: A Reexamination and Extension,” Journal ofMarketing, 56(July): 55-68. Davidow, William H. and Bro Uttal. (1989). Total Customer Service: The Udrimate Weapon. New York: Harper and Row. Dabholkar, Pratibha A. (1993). “Customer Satisfaction and Service Quality: Two Constructs or One?” Pp. lo-18 in David W. Cravens and Peter R. Dickson (eds.), Enhancing Knowledge Development in Marketing, Vol. 4. Chicago, IL: American Marketing Association. Fomell, Claes. (1992). “A National Customer Satisfaction Barometer: The Swedish Experience,” Journal of Marketing, 56(January): 6-21. Fomell, Claes and David F. Larcker. (1981). “Evaluating Structural Equation Models with Unobservable Variables and Measurement Error,” Journal of marketing Research, lS(Febmary): 39-50. Joreskog, Karl G. and Dag S&born. (1993). LZSREL 8: A Guide to the Program and Applications. Homewood, IL: Scientific Software International, Inc.

An Empirical Examination of a Model of Perceived Service Quality and Satisfaction

213

LaTour, Steven A. and Nancy C. Peat. (1979). “Conceptual and Methodological Issues in Consumer Satisfaction Research.” Pp. 43 l-437 in William L. Wilkie (ed.), Advances in Consumer Research, Vol. 6. Ann Arbor, MI: Association for Consumer Research. Oliva, Terence A., Richard L. Oliver and Ian C. MacMillan. (1992). “A Catastrophe Model for Developing Service Satisfaction Strategies,” Journal OfMarketing, 56(July): 83-95. Oliver, Richard L. (1980). “A Cognitive Model of the Antecedents and Consequences of Satisfaction Decisions,” Journal ofMarketing Research, 17(November): 460-469. . (1989). “Processing of the Satisfaction Response in Consumption: A Suggested Framework and Research Propositions,” Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 2: 1-16. (1993). “A Conceptual Model of Service Quality and Service Satisfaction: Compatible _. Goals, Different Concepts.” Pp. 65-85 in Advancesin Services Marketing andManagement, Vol. 2. in Satisfaction Oliver, Richard L. and Wayne S. DeSarbo. (1988). “Response Determinants Judgments,” Journal of Consumer Research, lrl(March): 495507. 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. Parasuraman, A., Valarie A. Zeithaml and Leonard L. Berry. (1988). “SERVQUAL: A Multi-Item Scale for Measuring Consumer Perceptions of Service Quality,” Journal of Retailing, 64(Spring): 13-40. Parasuraman, A., Valarie A. Zeithaml and Leonard L. Berry. (1994). “Alternative Scales for Measuring Service Quality: A Comparative Assessment Based on Psychometric and Diagnostic Criteria,” Journal of Retailing, 70(Fall): 201-230. Peters, Tom. (1987). Thriving on Chaos. New York: Harper and Row. Rust, Roland T. and Richard L. Oliver. (1994). “Service Quality: Insights and Managerial Implication From the Frontier.” Pp. l- 19 in Roland T. Rust and Richard L. Oliver (eds.), Service Quality: New Directions in Theory and Practice. Thousand Oaks, CA: Sage. Spreng, Richard A. and Andrea L. Dixon. (1992). “Alternative Comparison Standards in the Formation of Consumer Satisfaction/Dissatisfaction.” Pp. 85-91 in Leone and Kumar (eds.), Enhancing Knowledge Development in Marketing, Vol. 3. Chicago, IL: American Marketing Association. Spreng, Richard A. and Richard W. Olshavsky. (1993). “A Desires Congruency Model of Consumer Satisfaction,” Journal of the Academy of Marketing Science, Zl(Summer): 169-177. Spreng, Richard A. and A.K. Singh. (1993). “An Empirical Assessment of the SERVQUAL Scale, and the Relationship Between Service Quality and Satisfaction.” Pp. l-6 in David W. Peter R. Cravens and Dickson (eds.), Enhancing Knowledge Development in Marketing, Vol. 4. Chicago, IL: American Marketing Association. Swan, John E. (1988). “Consumer Satisfaction Related to Disconfirmation of Expectations and Product Performance,” Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 1: 40-47. Swan, John E. and I. Frederick Trawick. (1980). “Satisfaction Related to Predictive vs Desired Expectations.” Pp. 7-12 in H. Keith Hunt and Ralph L. Day (eds.), Refining Concepts and Measures of Consumer Satisfaction and Complaining Behavior. Bloomington, IN: Indiana University Press. Taylor, Steven A. and Thomas L. Baker. (1994). “An Assessment of the Relationship Between Service Quality and Customer Satisfaction in the Formation of Consumers’ Purchase Intentions,” Journal of Retailing, 70(2): 163-178. Teas, R. Kenneth. (1993). “Expectations, Performance Evaluation, and Consumers’ Perceptions of Quality,” Journal of Marketing, 57(0ctober): 18-34.

214

Journal of Retailing

Vol. 72, No. 2 1996

Westbrook, Robert A. and Michael D. Reilly. (1983). “Value-Percept Disparity: An Alternative to the Disconfirmation of Expectations Theory of Consumer Satisfaction.” Pp. 256-261 in Richard P. Bagozzi and Alice M. Tybout (eds.), Advances in Consumer Research, Vol. 10. Ann Arbor, MI: Association for Consumer Research. Woodruff, Robert B., D. Scott Clemons, David W. Schumann, Sarah F. Gardial and Mary Jane Bums. (1991). “The Standards Issue in CS/D Research: Historical Perspective,” Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 4: 103-109. Yi, Youjae. (1990). “A Critical Review of Consumer Satisfaction.” Pp. 68-123 in Valarie A. Zeithaml (ed.), Review of Marketing. Chicago, IL: American Marketing Association. 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, Zl(Winter): 1-12.