Assessment of the Three-Column Format SERVQUAL: An Experimental Approach

Over the past decade, SERVQUAL has emerged as perhaps the most emerged as perhaps the most popular standardized questionnaire to measure service quality. Despite numerous criticisms, popular standardized questionnaire to measure service quality. This study investigates the usefulness of the three-column format SERVQUAL proSERVQUAL continues to appeal to both academics and practitioners. Originally developed and tested in the consumer posed by Parasuraman, Zeithaml and Berry, (Parasuraman, A., Zeithaml, V. A., Berry, L. L.: Reassessment of expectations as a comparison standard retail environment, more recent replications have extended in measuring service quality: Implications for further research, Journal of to the industrial market (Westbrook and Peterson, 1998; Marketing 58 (January 1994): 111–124. Using a six-group experimental Young and Varble, 1997), the public sector (Orwig, Pearson, design, group 1 respondents are administered the revised SERVQUAL and Cochran, 1997), higher education (Anderson, 1995a), instrument containing minimum expectations, desired expectations, and information systems (Pitt, Watson, and Kavan, 1995), real performance items. The remaining five groups are given different combinaestate (Nelson and Nelson, 1995), hospitals (Anderson, tions of the one and/or two column questionnaire. The findings indicate 1995b; Youssef, Nel, and Bovaird, 1995) and the legal profesthat the perception battery is the salient component, raising new concerns sion (Witt and Stewart, 1996). Moreover, the number of SERVregarding the usefulness of the revised expectations scale in service quality QUAL applications outside the United States continues to measurement. Management implications and future research issues are increase (see for example, Lam and Woo, 1997; Johns and discussed. J BUSN RES 2000. 49.57–65.  2000 Elsevier Science Inc. Tyas, 1997). All rights reserved Notwithstanding the considerable research that has been conducted using SERVQUAL, there are still a number of unresolved issues associated with service quality measurement. These include the need for further research on expectations, Definitions of service quality revolve around the idea as well as elaboration of gap issues and the underlying factor that it is the result of the comparison that customers structure (Carman, 1990; Babakus and Boller, 1992; Bouman make between their expectations about a service and and van der Wiele, 1992; Cronin and Taylor, 1992, 1994; their perception of the way the service has been performed Teas, 1993; Dabholkar, Thorpe, and Rentz, 1996; Gagliano (Lehtinen and Lehtinen, 1982; Lewis and Booms, 1983; Grönand Hathcote, 1994). In response to these criticisms, Parasuraroos, 1984; Parasuraman, Zeithaml, and Berry, 1985, 1988). man, Zeithaml, and Berry (1994b) developed and investigated In developing their GAPS model, the latter authors employ three alternative SERVQUAL formats. From their empirical generally accepted psychometric procedures (cf., Churchill, research, the authors conclude that “the three-column format 1979) that have resulted in the operationalization of service questionnaire seems most useful” as it can be used for diagnosquality by the SERVQUAL instrument. SERVQUAL has tic purposes and offers the possibility of using the perception items separately for those interested in maximizing predictive power. Among other improvements, the three-column format Address correspondence to A. Caruana, Department of Marketing, University incorporates the recent reconceptualization of the expectations of Malta, Msida MSD06, Malta. The research for this paper was conducted when Albert Caruana was a side of the GAPS model into desired and minimum expectavisiting research fellow with the School of Marketing at Curtin Business School in Perth, Western Australia. tions and the use of nine-point scales. Adjustments to the