SERVQUAL and IS-SERVQUAL

---

Service quality is the most researched area of services marketing (Fisk et al., 1993). Parasuraman et al. (1985) suggest that service quality is an overall evaluation similar to attitude. They conducted a total of twelve focus group interviews with current or recent consumers of four difference services, including retail banking, credit card, securities brokerage, and product repair and maintenance, to conceptualize the construct of service quality. In developing the SERVQUAL instrument, Parasuraman et al. (1988) contend:

Perceived quality is the consumer's judgement about an entity's overall excellence or superiority (Zeithaml, 1987). It differs from objective quality (as defined by, for example, Garvin, 1983 and Hjorth-Anderson, 1984); it is a form of attitude, related but not equivalent to satisfaction, and results from a comparison of expectations with perceptions of performance.

Based on extensive focus group research (Parasuraman et al., 1985), the original 22-item SERVQUAL was developed (Parasuraman et al., 1988) for application in a broad spectrum of service sectors, including an intrafirm context (Zeithaml et al., 1990). Parasuraman et al. (1988) found, regardless of the type of service, customers used basically similar criteria in evaluating service quality and that these criteria span virtually all aspects of the service. Underlying the 22 items are five dimensions, including:

• Tangibles: Physical facilities, equipment, and appearance of personnel.

• Reliability: The ability to perform the promised service dependably and accurately.

• Responsiveness: The willingness to help customers and to provide prompt service.

• Assurance: The knowledge and courtesy of employees and their ability to convey trust and confidence.

• Empathy: Caring, individualized attention the service provider gives its customers.

Service quality for each dimension is captured by a difference (gap) score G (representing perceived quality for that item), where G = P - E and P and E are the average ratings of a dimension's corresponding perception and expectation statements, respectively.

Because service quality is a significant topic in marketing, the SERVQUAL (Parasuraman et al., 1988, 1991, 1993, 1994a, 1994b) has been subject to considerable debate regarding the use of difference scores (Peter et al., 1993; Cronin & Taylor, 1992), ambiguous definition of the expectations (Teas, 1993, 1994), applicability of SERVQUAL across industries and setting (Carman, 1990; Brown et al., 1993; Dabholkar et al., 1996), reliability of difference scores (Brown et al., 1993; Peter et al., 1993), predictive and convergent validity (cf., Babakus & Boller, 1992; Brown et al., 1993; Cronin & Taylor, 1992; Parasuraman et al., 1991, 1993, 1994b), and unstable dimensionality of the SERVQUAL instrument (e.g., Carman, 1990; Babakus & Boller, 1992).

Recognizing the need to more comprehensively measure information services quality, Kettinger & Lee (1994) adapted the SERVQUAL instrument to the IS context as an enhancement to the existing User Information Satisfaction (UIS) measure (Ives et al., 1983). Kettinger et al. (1995) then examined cross-culturally the strength of the IS-adapted SERVQUAL instrument using organizations in four different countries. Pitt et al. (1995) further extended the application of SERVQUAL in IS by placing service quality within the IS Success Model (Delone & McLean, 1992) and by independently testing SERVQUAL's reliability and validity in samples from three different organizations.

Van Dyke et al. (1997, 1999) have made an important contribution by bringing an ongoing debate from marketing to the IS community. They concentrated on the conceptual and empirical difficulties with the IS-adapted SERVQUAL (Kettinger & Lee, 1994; Pitt et al., 1995) and provided a sound summary of the marketing literature in this regard (see Table 2). Van Dyke et al. suggest the need for an alternative response format. One alternative is to use the perception-only method of scoring (SERVPERF). A second alternative, suggested by Carman (1990) and Babakus & Boller (1992), is to revise the wording of the SERVQUAL items into a format combining both expectations and perceptions into a single question.

Table 2: Problems Identified in the Literature

Issue	Problem	Literature
Use of gap scores	Poor choices as a measure of a complex psychological construct	Lord, 1958; Wall & Payne, 1973; Johns, 1981; Peter et al., 1993
Reliability	Cronbach's alpha overestimates the reliability of difference scores	Lord, 1958; Wall & Payne, 1973; John, 1981; Prakash & Lounsbury, 1983; Peter et al., 1993
Discriminant validity	Given the high correlation between the differences score and the perception score, it is difficult to demonstrate that the difference score is measuring some thing unique from the perceptions component.	Cronin & Taylor, 1992; Peter et al., 1993
Spurious correlations	Correlations between gap scores and other variables are artifacts of correlations with the components	Peter et al., 1993
Variance restrictions	"E" scores are consistently higher than "P" scores. This leads to a systematic variance restriction which is problematic for many types of statistical analyses.	Peter et al., 1993; Brown et al., 1993
Validity	The perception component of the perception-minus-expectation scores performs better as a predictor of perceived overall quality than the difference score itself.	Parasuraman et al., 1988; Cronin & Taylor, 1992, 1994; Babakus & Boller, 1992; Boulding et al., 1993; Brensinger & Lambert, 1990
Ambiguity of "expectation" construct	Multiple definitions of "expectations" result in a concept that is loosely defined and open to multiple interpretations. These various interpretations can result in serious measurement validity problems.	Tea, 1993, 1994
Unstable dimensionality	Atheoretical construction combined with the use of gap scores raise questions about the true factor structure of the service quality construct.	Babakus & Boller, 1992; Brensinger & Lambert, 1990; Finn & Lamb, 1991; Carman, 1990; Cronin & Taylor, 1992; Cronbach & Furby, 1970; Parasuraman et al., 1991
Source: Van Dyke et al. (1999)

Pitt et al. (1997) and Kettinger et al. (1997, 1999) subsequently responded to the research note in this issue by Van Dyke et al. (1997, 1999) concerning the use of SERVQUAL and are resistant to "throw the baby out with the bath water" until strong conceptual, empirical, and practical evidence pushes the IS-adapted service quality scale markedly away from the SERVQUAL. Specifically, Kettinger & Lee (1997) contend that "measures should not be discarded until such time as their underlying theory and practicality for IS have been conceptually and empirically discredited" (p. 236).

In sum, the SERVQUAL is widely used today as a diagnostic tool for uncovering areas of service quality strengths and shortfalls. In developing the SERVQUAL instrument, Parasuraman et al. (1998, 1991, 1994a) intended to derive a service quality measure that would transcend multiple measurement contexts. They suggest that, regardless of the types of service, customers use basically similar criteria in evaluating service quality and that these criteria span virtually all aspects of the service. They also assert that as core evaluation criteria, the SERVQUAL provides a basic "skeleton" underlying service quality that can undergo minor wording modifications. After examining several studies, Fisk et al. (1993) conclude that researchers generally agree that the SERVQUAL is a good predictor of overall service quality. Prior studies (Kettinger & Lee, 1994; Pitt et al., 1995; Watson et al., 1998) also indicate that the SERVQUAL is adequate for measuring IS service quality. Furthermore, research reports that practitioners find the SERVQUAL a useful tool for assessing service quality and determining actions for raising service quality (Pitt et al., 1995). Thus, it is justified to adopt the SERVQUAL as a base for the new EC-SERVQUAL instrument. Since the current SERVQUAL and IS-SERVQUAL instruments are targeted primarily towards either traditional retailing or information systems contexts, this study attempted to validate and refine an EC-adapted SERVQUAL instrument for assessing customer perceptions of web site service quality in digital and service contexts

---