Note A NOTE ON THE MEAsuREMENT OF SERVICE QUALITY-A (FRmNnLV) CRITIQUE OF STANK Kelly C. Strong Steven A. Taylor Illinois State University Normal, Dlinois Introduction Theodore Stank's (1993) recent article in the Journal of Business Strategies offers a timely discussion of the importance of quality as a strategic tool within service indus- tries. We emphatically agree with his argument that improving service quality is an effective way to gain competitive advantage through differentiation. We also applaud his suggestion that quality must be consistently and effectively measured from the customer's perspective. Stank should also be commended for suggesting that service firms should consider service quality feedback prior to making resource allocation decisions within the context of strategic planning. In short, we are delighted to see that the spe- cial needs of service firms are receiving attention within the strategy literature and that quality of service is being discussed in strategic terms instead of being narrowly treated as a marketing construct. Because we believe there to be a chronic shortage of illuminating discussions of strat- egies for service firms, (including quality assessment), we are reluctant to criticize one of the few service quality/service strategy articles published in a strategy journal. There- fore, we wish to emphasize the adjective "friendly" in the title to this note. We do not wish to detract from the importance of service quality in developing strategies for ser- vice firms. However, we feel management educators, strategy researchers, and strategic decision makers should be aware of the rapidly evolving literature and research on ser- vice quality assessment which calls into question the foundations of the SERVQUAL (parasuraman, Zeitharnl, and Berry 1985; 1988; 1991) scale which forms the basis of Stank's recommended method. The SERVQUAL scale (parasuraman, Zeitharnl, and Berry 1985; 1988; 1991; hereaf- ter referred to as PZB) was devised and promoted as a generalizable instrument for measuring service quality across a variety of service industries. The logistics service quality measurement tool proposed by Stank (1993) is based largely on the SERVQUAL scale, with minor adaptations tailored to the nuances of the logistics setting. Thus, it would appear that many of the criticlsms being leveled at the SERVQUAL scale in the emerging literature apply equally to the logistics service quality instrument offered by Stank. In the following section, we will attempt to present a summary of those criti- cisms, and will then offer some suggestions for future directions in service quality as- sessment. Spring 1994 Strong & Taylor: Measurement ofService Summary of Findings Regarding the SERVQUAL Scale 77 The theory upon which the SERVQUAL scale is based was developed through ex- ploratory focus--group research conducted by PZB (1985, 1988). From these focus--group interviews with consumers and managers, PZB concluded that customers bring some prior expectation of firm performance to a service encounter, and then compare their percep- tions of the firm's actual performance with this preconceived expectation. This compari- son of expected performance to perceived performance, termed expectancy disconfumation, forms the basis of service quality within the SERVQUAL methodology advocated by PZB. Expectancy disconfirmation and its relationship to service quality are depicted in Equa- tion 1. Service Quality = f(perceptionsPerformance - ExpectationsPerformance) [1] The crux of equation [1] is that service quality is a function of the gap between percep- tions of performance and expectations of performance. When perceptions of performance meet or exceed expectations, positive service quality is perceived. When perceptions of performance are less than or equal to expectations, poor quality service is perceived. Therefore, in order to measure service quality using equation [1], both perceptions of performance and expectations of performance must be captured. Some strategy scholars (e.g. Quinn 1992; Stank 1993) have recently recommended the SERVQUAL scale (c.f., PZB 1988) as a survey instrument that purports to serve this purpose with broad generalizability and reliability across alternative service settings. In other words, consis- tent with PZB (1988), these strategists would argue that the SERVQUAL scale can and should be generalizable to alternative service settings such as logistics service quality. However, a number of emerging studies are questioning the use of the SERVQUAL scale as a basis for measuring service quality. Criticisms of the Conceptual Arguments Underlying the SERVQUAL Model The first major criticism of the SERVQUAL scale was reported by Cronin and Taylor (1992). Cronin and Taylor present an empirical analysis demonstrating that the PerceptionsPerfOlIllllllCC measures alone in equation [1] explain more of the variance in overall service quality perceptions than does the entire expectancy disconfirmation equation. Since a single variable from the SERVQUAL formula explains more variance in service qual- ity perception than does the derivative of that formula, one can only conclude that the theoretical foundations of the SERVQUAL scale appear flawed. (At the risk of oversim- plifying the argument, if one assumes that k is a function of i+j, but i explains more of k's variance than does i+j, then one must assume that i+j has no theoretical based rela- tionship to k). Specifically, Cronin and Taylor (1992) argue that the expectancy disconfirmation model was inappropriately generalized by PZB from the consumer satisfaction literature (c.f. Oliver 1980) to explain service quality perceptions and suggest the simplified relation- ship described in equation 2 as a more appropriate model of service quality. 78 Journal ofBusiness Strategies Service Quality = f(perceptionperformance) Vol. 11, No.1 [2] Subsequent research appears to support this argument, although readers should be aware that the debate in the literature is ongoing. Oliver (1993a), who developed the original expectancy disconfirmation model in 1980, recently reviewed the service quality and consumer satisfaction literature and concluded that service quality perceptions are sepa· rate from consumer satisfaction judgments along at least three dimensions (also see Patterson and Johnson (1993) for a similar discussion reaching similar conclusions, 1) the domains of service quality and consumer satisfaction are different (i.e., they are conceptually different things). Service quality is best considered a long-term attitude reflecting service superiority or excellence while consumer satisfaction is a transitory judgment reflecting fulfillment. 2) service quality perceptions can develop in the absence of personal experience whereas consumer satisfaction requires personal experience. 3) service quality perceptions appear to be antecedents to consumer satisfaction judg- ments (i.e., quality - satisfaction). Taylor (1993) extends these arguments by developing a model of consumer decision making that reconciles the emerging evidence concerning the conceptual domains of the relevant constructs. Taylor suggests that prior to personally experiencing a service pro- vider, initial service quality perceptions can be based on communication influences and vicarious experiences through friends, relatives, and coworkers. However, once a person has a personal experience, service quality becomes an aggregate representation of the summed influences of their previous service quality attitudes and their most recent satis- faction judgment from a service encounter (see equation 3). Service Qualityl+l = f (Service Qualityt_l' Satisfaction) [3] These models are at odds with the most recent SERVQUAL conceptualization. That is, the original authors of the SERVQUAL scale have maintained their support for the expectancy disconfumation model of service quality in spite of the growing arguments against the model. Zeithaml, Berry, and Parasuraman (1993) recently present a response to the growing criticisms of the conceptual underpinnings of the SERVQUAL scale. Here the authors modify their original position by suggesting that both consumer satis· faction judgments and service quality can form from the expectancy disconfumation process. Importantly, there is no empirical evidence to date supporting this attenuated position. However, there is empirical evidence to support other models that do not appear rec- oncilable with the gap model. For example, Teas (1993) recently assesses the SERVQUAL scale and concludes from a comprehensive empirical analysis that the gap formulation recommended by PZB appears to exhibit questionable validity. Indeed, in a recent article on service quality by one of the original authors of the SERVQUAL scale report results that they themselves suggest appear inconsistent with the gap model (Boulding, Kalta, Staelin and Zeitharnl 1993). Spring 1994 Strong & Taylor: Measurement ofService 79 The satisfaction literature similarly appears to support the argument that expectancy disconfinnation should not be used in service quality measurement. Oliver (1993b) pre- sents a compelling empirical analysis of consumer satisfaction which demonstrates that consumer satisfaction judgments result from the influences of both expectancy disconfinnation and (positive and negative) affect. We suggest that the weight of the evidence supports the position that expectancy disconfinnation belongs in models of con- sumer satisfaction and not service quality. Equation [4] presents the model of consumer satisfaction empirically validated by Oliver (1993b). Consumer Satisfactiont+l = f(Affect., Disconfirmationt> QUalityt_l) [4] Thus, the emerging literature and empirical evidence do not support the gap model ad- vocated by Parasuraman, Zeithaml, and Berry for service quality measurement (i.e., SERVQUAL). Rather, the evidence supports the position that service quality should be measured by perceptions of service-firm performance only (see [2]), while satisfaction is unique from service quality and best modeled as equation [4]. This may lead one to wonder whether ExpectationsPerformance are at all useful from a managerial or research perspective. We suggest that they are, however, only as a unique construct. Boulding et al (1993) have recently presented evidence that consumers' ex- pectations of what service firms should and will provide exerts a positive influence on subsequent service quality judgments. This suggests that while the gap measures advo- cated by PZB, and by extension Stanks, may not be useful, expectations and perceptions of performance evaluations can contribute to a better understanding of consumer deci- sion-making processes when treated as unique constructs. Empirical Criticisms of the SERVQUAL Scale The psychometric properties of the SERVQUAL survey instrument have also recently been criticized by a number of authors. Cronin and Taylor (1992) were the first to point out that the generalizability of the SERVQUAL scale appears suspect. Specifically, Cronin and Taylor (1992) found that the factor structure underlying data collected using the SERVQUAL scale varies across service settings. This led Cronin and Taylor (1992) to suggest that the individual scales for expectations and performance perceptions be treated as unidimensional scales and summed-and-averaged in analysis. 1 Babalrus and Boller (1992) empirically assessed the psychometric properties of the SERVQUAL scale and report findings that support the conclusions of Cronin and Taylor (1992). Specifically, these authors also found that the dimensionality of service quality data captured using the SERVQUAL instrument appears to depend on the type of ser- vices under investigation. Babakus and Boller additionally found that the practices of using mixed-word item wording and the general use of gap measures should be approached with caution. Brown, Churchill, and Peter (1993) recently report an additional assessment of the psychometric properties of the SERVQUAL scale and also report serious problems from their analyses. First, higher reliabilities are found with the performance-based measures 80 Journal ofBusiness Strategies Vol. 11, No.1 of service quality than the gap-based measures advocated in SERVQUAL (c.f., Cronin and Taylor 1992). Second, the SERVQUAL items failed to achieve discriminant valid- ity from its component parts. Third, the SERVQUAL scale items tend to exhibit vari- ance restriction. Finally, the distribution of SERVQUAL scores are reported as non- normal in their study. Readers interested in further exploring these issues are directed to a recent debate in print concerning the relative merits of these models (c.f., Cronin and Taylor 1994; Parasuraman, Zeithaml, and Berry 1994; Teas 1994). A final note concerns the use of importance weights in models of service quality. Stank argues that the addition of importance weights is important for measuring logistics ser- vice quality. However, we ask that strategic decision makers take care to ensure that the data support this call in their own specific setting. Cronin and Taylor (1992) present evidence that not only do importance weights fail to contribute to the explained variance in many service settings, they can actually reduce one's ability to explain variations in service quality perceptions. Recommendations for Future Research A number of recommendations are apparent for consideration by strategists. First, researchers should identify as a high priority the identification of reliable and valid sets of scale items that can serve as measures of service quality and satisfaction in models of organizational strategy. A fundamental challenge in achieving this end will be achieving discriminant validity between the measures. Second, research investigating the anteced- ents and outcomes of service quality from a strategic perspective appears warranted. Third, ascertaining if there are perceived differences between the internal and external constitu- encies of the organization's service quality should be investigated. The impact of these differences, if any, can then be assessed relative to such important constructs as leader- ship, organizational culture, and firm performance. Summary The preceding discussion attempts to inform strategic managers and researchers of the rapidly evolving literature concerning the service quality and consumer satisfaction con- structs. We take the position that dialogue and interdisciplinary efforts enhance our un- derstanding of shared problems in the social sciences. We support Stank's (1993) call for using consumer service quality judgments in strategic decision-making processes and hope that this discussion enhances these actions. Clearly, strategic management must be approached from a quality enhancement frame- work (Quinn 1992; Drucker 1992) for all firms, including service organizations (Nayyar 1992). This suggests the need for quality assessment tools designed for use within the service setting. The SERVQUAL scale as recommended by Stank (1993) represents a noble start in the right direction, but strategists should be aware that the weight of the evidence suggests that it is not adequate in its current form and method. We caution both researchers interested in service quality and service fInn executives to be fully aware Spring 1994 Strong & Taylor: Measurement ofService 81 of what they are (and are not) getting when they use the SERVQUAL method of ser- vice quality assessment. Endnote 1 Cronin and Taylor (1992) in effect suggest using these two set of items as indices rather than as factor-based scales. The difference is that an index is an exact linear combination of observed items. The dimensionality of the items of an index used as an observed variable is not relevant. References Babakus, E. and G.W. Boller (1992). "An Empirical Assessment of the SERVQUAL Scale." Journal of Business Research 24: 253-268. Brown, T.J., GA Churchill, and J.P. Peter (1993). "Improving the Measurement of Service Qual- ity." Journal of Retailing 69 (1): 127-39. Boulding, William, Ajay Kalra, Richard Staelin, and Valeri~ Zeithaml (1993). "A Dynamic Pro-- cess Model of Service Quality: From Expectations to Behavioral Intentions." Journal of Market- ing Research XXX: 2-27. Carman, J.H. 1990. "Consumer Perceptions of Service Quality: An Assessment of the SERVQUAL Dimensions." Journal of Retailing 66 (1): 33-55. Cronin, J., and S.A Taylor (1992). "Measuring Service Quality: A Reexamination and Extension." Journal of Marketing 56 (July): 55--68. Cronin, J., and S.A Taylor (1994). "SERVPERF versus SERVQUAL: Reconciling Perfor- mance-Based and Perceptions-Minus-Expectations Measurement of Service Quality." Journal of Marketing 58 (1), 125-3 1. Drucker, P. (1992). Managing for the Future: The 19908 and Beyond. New York: Truman Talley Books/Dutton. Nayyar, P.R. (1992). "Performance Effects of Information Asymmetry and Economies of Scope in Diversified Service Firms." Academy of Management Journal 36 (1): 2B-57. Oliver, Richard L. (1980), "A Cognitive Model of the Antecedents and Consequences of Satisfac- tion Decisions." Journal of Marketing Research XVII (November 1980): 460-469. Oliver, Richard L. (1993a), "A Conceptual Model of Service Quality and Service Satisfaction: Compatible Goals, Different Concepts." In Advances in Services Marketing and Management 2: 65-85. 82 Journal ofBusiness Strategies Vol. 11, No.1 Oliver, Richard L. (1993b). "Cognitive, Affective, and Attribute Bases of the Satisfaction Re- sponse." Journal of Consumer Research 20 (December): 418-30. Parasuraman, A, V.A Zeithaml, and L.L. Berry (1985), "A Conceptual Model of Service Qual- ity and its Implications for Future Research." Journal of Marketing 49 (Fall): 41-50. Parasuraman, A, V.A Zeithaml, and L.L. Berry (1988), "SERVQUAL: A Multiple-Item Scale for Measuring Consumer Perceptions of Service Quality," Journal of Retailing 64 (1) 12- 40. Parasuraman, A, V A Zeithaml, and L.L. Berry (1991), "Refmement and Reassessment of the SERVQUAL Scale." Journal of Retailing 67 (4): 420-450. Parasuraman, A, VA Zeithaml, and L.L. Berry (1994). "Reassessment of Expectations as a Com- parison Standard in measuring Service Quality: Implications for Future Research," Journal of Marketing 58 (1), 111-24. Patterson, Paul G. and Lester W. Johnson (1993). "Disconfirmation of Expectations and the Gap Model of Service Quality: An Integrated Paradigm." Journal of Consumer Satisfaction, Dissatis- faction, and Complaining Behavior Volume 6, 90-9. Quinn, J.B. (1992), Intelligent Entemrise. New York: The Free Press. Stank, T. (l993), "Measuring Logistics Service Quality." Journal of Business Strategies 10 (Spring): 51-62. Taylor, Steven A (1993). "The Roles of Service Quality, Consumer Satisfaction, and Value in Quinn's (1992) Paradigm of Services." The Journal of Marketing Theory and Practice Volume 2, Number 1, (Fall 1993): 14-26. Teas, R. Kenneth (1993). "Expectations, Performance Evaluation, and Consumers' Perceptions of Quality." Journal of Marketing 57 (4): 18-34. Teas, R. Kenneth (1994). ''Expectations as a Comparison Standard in measuring Service Qual- ity: An Assessment of a Reassessment," Journal of Marketing 58 (1), 132-9. Zeitharnl, Valerie A, Leonard L. Berry, and A Parasuraman (1993). "The Nature and Detenni- nants of Customer Expectations of Service." Journal of the Academy of Marketing Science 21 (1): 1-12. A Note on the Measurement of Service Quality-A (Friendly) Critique of Stank