A Measure of Service Quality for Retail Stores: Scale Development and Validation Pratibha A. Dabholkar Dayle I. Thorpe Joseph 0. Rentz University of Tennessee, Knoxville Current measures of service quality do not adequately capture customers ‘perceptions of service quality for retail stores (i. e. , stores that offer a mix of goods and services).
A hierarchical factor structure is proposed to capture dimensions important to retail customers based on the retail and service quality literatures as well as three separate qualitative studies. Confirmatory factor analysis based on the partial disaggregation technique and cross-validation using a second sample support the validity of the scale as a measure of retail service quality. The implications of this Retail Service Quality Scale for practitioners, as well as for future research, are discussed. INTRODUCTION
The retail environment is changing more rapidly than ever before It is characterized by intensifying competition from both domestic and foreign companies, a spate of mergers and acquisitions, and more sophisticated and demanding customers who have greater expectations related to their consumption experiences (Sellers 1990, Smith 1989). Consequently, retailers today must differentiate themselves by meeting the needs of their customers better than the competition. There is general agreement that a basic retailing strategy for creating competitive advantage is the delivery of high service quality (e. . , Berry 1986; Hummel and Savitt 1988; Reichheld and Sasser 1990). The most widely known and discussed scale for measuring service quality is SERVQUAL, a scale designed to Journal of the Academy of Marketing Science. Volume 24, No. 1, pages 3-16 Copyright © 1996 by Academy of Marketing Science. measure five dimensions of service quality: tangibles, reliability, responsiveness, assurance, and empathy. Although SERVQUAL has been empirically tested in a number of studies involving “pure” service settings (e. g. banking, long-distance telephone service, securities brokerage, and credit card service), it has not been successfully adapted to and validated in a retail store environment. In fact, little research has been conducted in retail settings, defined here as stores that offer a mix of merchandise and service. A retail store experience involves more than a nonretail service experience in terms of customers negotiating their way through the store, finding the merchandise they want, interacting with several store personnel along the way, and returning merchandise, all of which influence customers’ evaluations of service quality.
Thus, although measures of service quality for pure service environments and for retail environments are likely to share some common dimensions, measures of retail service quality must capture additional dimensions. Our purpose is to investigate the dimensions of service quality in a retail environment and to develop and validate a scale to measure retail service quality. EXPLORING POSSIBLE FACTOR STRUCTURES FOR RETAIL SERVICE QUALITY An examination of the retail literature offers little to support a theory-based factor structure of retail service quality.
The retail literature focuses on service quality at either the integrated or the attribute level; there is a lack of discussion of service quality at the factor (or dimension) level. At the integrated level, Westbrook (1981) suggests that two broad categories of retailer-related experiences Copyright © 2001. All Rights Reseved. 4 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE are important to the customer: (1) m-store experiences and (2) experiences related to the merchandise In-store experiences include interactions with store employees as well as the ease of walking around the store.
Experiences related to merchandise include quality and availability of merchandise. Although we agree that these experiences are important to retail customers, both categories appear to encompass more than one factor or dimension. In addition, there may be an overlap between the two categories. For example, merchandise displays could be viewed both as in-store experiences and experiences related to merchandise. Viewing service quality at this integrated level does little to suggest the separate and critical dimensions of retail service quality that would be useful to retailers and researchers.
At the attribute level, researchers suggest that store layout and the quality of merchandise are important to the customer’s perception and evaluation of retail stores (Gutman and Alden 1985; Hummel and Savitt 1988, Mazursky and Jacoby 1985, Oliver 1981) Store layout includes the ease of locating desired merchandise and the ease of moving around m the store. Westbrook (1981) and Mazursky and Jacoby (1985) report that other important criteria on which customers evaluate retail stores are the credit and charge account policies of the store and the ease with which stores refund or exchange merchandise.
Baker, Grewal, and Parasuraman (1994) suggest that store environment encompassing ambient attributes (eg. , music), design attributes (e. g. , physical facilities), and social attributes (e. g. , customer responsiveness of service providers) are important to customers of retail stores in evaluating service quality Another attribute identified as important to retail shoppers is ease of parking (Oliver 1981). Finally, Westbrook (1981) suggests that customers are sensitive to a service provider’s willingness to promptly attend to problems or complaints.
Although it is useful to review the retail literature to develop a list of attributes that are important to customers in evaluating retail service experiences, there is little support to suggest how these attributes may be combined into a few critical dimensions of retail service quality. Because of the weak theoretical support for a factor structure provided by the retail literature, a review of the service quality literature was conducted to suggest possible factors for retail service quality.
As mentioned earlier, researchers have attempted to test and/or adapt the SERVQUAL instrument in various settings. These settings include a health care setting (Babakus and Mangold 1989), business-to-business services (Brensinger and Lambert 1990), a dental school patient clinic, business school placement center, tire store, and acute care hospital (Carman 1990), a utility company (Babakus and Boiler 1991), department stores (Finn and Lamb 1991), health are (Bowers and Swan 1992), banking, pest control, dry cleaning, and fast food (Cronin and Taylor 1992), department stores (Guiry, Hutchinson, and Weitz 1992), the computer software industry (Pitt, Oosthuizen, and Morris 1992), and banking (Spreng and Singh 1993). In general, these studies do not support the factor structure posited by Parasuraman, WINTER 1996 Zeithaml, and Berry (1988). See Table 1 for a review of the methodology and results from these studies The studies of particular interest to this research are those conducted in a retail setting.
Carman (1990) tested SERVQUAL mainly in pure service settings (dental school patient clinic, business school placement center, acute care hospital); the one exception was a tire store, an example of a retailer offering a mix of merchandise and services. Carman found nine factors of service quality, concluded that the five dimensions identified by Parasuraman et al. (1988) were not generic, and suggested that the instrument be adapted by adding items or factors as pertinent to different situations. Finn and Lamb (1991) tested SERVQUAL in four different types of retail stores, ranging from “stores like Kmart” to “stores like Neiman Marcus. Using confirmatory factor analysis, Finn and Lamb were unable to find a good fit to the proposed five-factor structure and concluded that SERVQUAL, without modification, could not be used as a valid measure of service quality in a retail setting However, they did not offer an alternative acceptable structure or measure In an unpublished paper, Guiry et al (1992) modified the original 22-item SERVQUAL to a 51-item instrument by dropping 7 items and adding 36 new items designed to measure service attributes at the retail store level.
Exploratory factor analysis revealed seven dimensions: (1) personal service during interaction with employees, (2) merchandise assortment, (3) store transaction procedure reliability, employee availability in the store before interaction, tangibles, (6) store service policy reliability, and (7) price. Although this study represents a good start toward the development of a service quality scale for the retail setting, the research was based on exploratory factor analysis and not on a theory-based factor structure. Additionally, the inclusion of price in a service quality construct explication is unusual
Although we recognize that price is an important determinant of store patronage, we view it as distinct from service quality Finally, in a study on customer evaluations of banking, Spreng and Singh (1993) performed a confirmatory factor analysis of SERVQUAL and found a poor fit for the five dimensions They found a lack of discriminant validity between responsiveness and assurance and noted that although modification indices were high, there was no clear indication for ways to improve the model fit. TRIANGULATION OF QUALITATIVE RESEARCH TECHNIQUES
Given the lack of a theory-based factor structure from the retail literature and the fact that SERVQUAL has not been supported or successfully adapted to retailing, it was deemed necessary to conduct further research to gain an understanding of the dimensions of retail service quality. To accomplish this end, qualitative research was conducted using three different qualitative methodologies— phenomenological interviews, exploratory depth Dabholkaretal / RETAIL SERVICE QUALITY 5 TABLE 1 Summary of SERVQUAL Replication Studies Study Instrument Analysis
Factor Structure Babakus and Mangold (1989) 15 of original 22 items Carman (1990)Modifications of SERVQUAL (using 12 to 21 of the original items in each case) Brensinger and LambertAll of original 22 items (1990) Finn and Lamb (1991)All of original 22 items Babakus and Boiler (1991) All of original 22 items Pitt, Oosthuizen, and Moms All of original 22 items (1992) Guiry, Hutchinson, and Weitz (1992) Cronin and Taylor (1992) Bowers and Swan (1992) Spreng and Singh (1993) 51 items (15 from the original 22 items plus 36 added items) All of original 22 items
Focus groups All of the original 22 items Exploratory factor analysis Principal axis factor analysis followed by oblique rotation Principal axis factor analysis followed by oblique rotation Confirmatory factor analysis Principal axis factor analysis followed by oblique rotation as well as confirmatory factor analysis (CFA) Principal components factor analysis followed by promax rotation Exploratory factor analysis (1) Confirmatory factor analysis (CFA), (2) OBLIMIN oblique factor rotation Qualitative methodologies Confirmatory factor analysis
One meaningful factor was found Could not identify the a prion five-factor structure Five to nine factors were identified A five-factor structure was found, but only four had eigenvalues > 1 The five-factor structure had a poor fit No other structures were analyzed The five-factor structure was not supported and no other structures were analyzed CFA resulted in a two-factor structure. The only study to obtain a five-factor structure However, the factors were different from SERVQUAL with reliability and responsiveness loading on one factor A seven-factor structure was found. ) CFA resulted in a unidimensional factor, 2) The five-factor structure had a poor fit. Identified the five original factors, plus four others The five-factor structure had a poor fit. Combimng responsiveness and assurance into one factor did not significantly improve the fit interviews, and tracking the customer through the store to monitor evaluations of the shopping experience. The objective was to use a triangulation of research techniques to gain further insights into factors important to customers of retail stores in evaluating service quality.
The first study consisted of three phenomenological interviews conducted to assign meaning to the shopping experience as the participant sees it, not as the researcher perceives it. We used the participants’ own words in follow-up questions to probe for important issues. This study found that customers were most concerned about store layout as it pertained to walking around the store and finding what they want, treatment by service employees in terms of being helped and feeling comfortable, ease of conducting exchanges and refunds and of resolving problems, and store policies regarding merchandise quality, parking, and credit card acceptance.
A second study employed six depth interviews to discover relevant determinants of the shopping experience not yet identified. Some of the findings of this study were similar to the first one. For example, customers were concerned about the ease of moving through the store, ease of finding what they were looking for, helpfulness of store employees, ease of returning merchandise, and store policies regarding credit In addition, the depth interviews revealed that the appearance of the store and its facilities (e. . , restrooms and fitting rooms) was important to customers. These participants also mentioned that they expected the service to be good and anything that was promised to be delivered. Finally, the third qualitative study used a “tracking” method to monitor the thought processes of three customers as they evaluated a specific shopping experience. A portable tape recorder with a lavalier microphone was inconspicuously attached to the participant’s clothing.
As they moved through the store, interacted with merchandise and store employees, and made a purchase decision, the participants spoke aloud their thoughts, observations, and reactions, which were tape-recorded. By unobtrusively “monitoring” the customer’s experience in the store, information regarding identified components of the shopping experience and the customers’ interactions with these elements was collected without significantly altering the natural flow of the experience.
These participants commented on store appearance, store layout, helpfulness of service employees, availability of service personnel at the cash registers, and quality of merchandise. Because there is general agreement in the literature about the conceptual definition of service quality, in our qualitative studies we asked general questions about expe- Copyright © 2001. All Rights Reseved. 6 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE WINTER 1996 FIGURE 1 Proposed Hierarchical Structure for Retail Service Quality [pic] f300 [???|?”1?”|11’1 fy*
