1997 Kettinger Abstract
In this research note, we join the debate between Van Dyke, Kappelman, and Prybutok and Pitt, Watson, and Kavan pertaining to the conceptual and empirical relevance of SERVQUAL as a measure of IS service quality. Adopting arguments from marketing, Van Dyke et al. (1997) question the SERVQUAL gap measurement approach, the interpretation and operationalization of the SERVQUAL expectation construct, and the reliability and validity of SERVQUAL dimensionality. In a response to those arguments, Pitt et al. (1997) defend their previous work (1995) in a point-by-point counterargument that suggests that the marginal empirical benefit of a perceptual-based (SERVPERF) service quality measure does not justify the loss of managerial diagnostic capabilities found in a gap measure. While siding with many of the positions taken by Pitt et al. (1997), we attempt to add value to the debate by presenting discrepancies we have with the two other research teams and by suggesting alternative approaches to resolve, or at least alleviate, problems associated with SERVQUAL. We believe that the theoretical superiority of an alternative IS service quality measure should be backed by empirical evidence in the IS context, hence answering some of the criticism by Van Dyke et al. and offering a construct valid version of the IS-adapted SERVQUAL. From a pragmatic viewpoint, we believe that the justification of using SERVQUAL's gap measure should be driven by more effective ways to utilize expectations in IS service management. To this end, we introduce the newer Parasuraman et al. (1994b) measures, the concept of a "zone of tolerance" for expectation management and an illustration of its practical use in an IS setting. Overall, we attempt to set the direction of where we think this debate should lead the IS field, namely, toward practical and timely IS service quality measures.