Four Levels of Loyalty

Journal of Retailing 80 (2004) 139158
The four levels of loyalty and the pivotal role of trust: a study of online service dynamics
Lloyd C. Harris , Mark M.H. Goode
Sir Julian Hodge Chair in Strategic Marketing, Cardiff Business School, Cardiff University, Colum Drive, Cardiff CF10 3EU, UK
Abstract A four-dimension scale of loyalty that reects Olivers [Satisfaction, a Behavioral Perspective on the Consumer, McGraw-Hill, New York, 1997] conceptualization of a sequential loyalty chain is proposed, operationalized, and tested. Further, through both synthesizing and building on existing research into loyalty, trust, satisfaction, value, and service quality, a framework is proposed and tested that positions trust as a pivotal driver of loyalty. Data is collected and analyzed from two surveys of online customers, the rst being purchasers of books and the second being a study of online ight purchasing. Analysis suggests that the hypothesized cognitive-affective-conative-action loyalty sequence is statistically most likely out of all possible variations. Although some differences emerge in the two studies, structural modeling largely supports the hypothesized framework and positions trust as central to service dynamics. 2004 by New York University. Published by Elsevier. All rights reserved.
Keywords: Loyalty; Trust; Satisfaction; Perceived value; Service quality; E-tailing
Introduction While some organizations are content with focusing on simply achieving customer satisfaction, gaining and sustaining customer loyalty constitutes the ultimate aspiration of many contemporary service providers. Loyal customers buy more, are willing to spend more, are easier to reach, and act as enthusiastic advocates for our rms. Predictably, for these and other reasons, ultimate loyalty has been described as an elusive (and arguably unattainable) corporate objective of the new millennium. Given the widespread acceptance of the causal link between loyalty and rm performance, it is perhaps not surprising to nd that customer loyalty for brick-and-mortar rms has been linked to antecedent factors ranging from relationship duration, to service reliability, to servicescape attributes. Although e-commerce1 is expanding exponentially both in the US and elsewhere (up to 49 percent per annum), sec Corresponding author. Tel.: +44 29 2087 5066; fax: +44 29 2087 4419. E-mail address: [email protected] (L.C. Harris). 1 Electronic commerce refers to all electronically assisted business processes that facilitate transaction or exchange (see Barnes-Vieyra & Claycomb, 2001). In contrast, the Internet may be viewed as an electronic medium that facilitates a range of transactions principally through the World Wide Web (WWW). This denition is deliberately generic and is designed to incorporate a variety of electronic channel transactions such as emails, intranet, and other network-assisted transactions although there
tor analysts have observed that only a very small minority of web site visitors (1.33.2 percent) return to make purchases (Boston Consulting Group, 2000; Shop.org, 2001). Nevertheless, recent research indicates that such rare, loyal online customers are highly protable (Nielsen, 1997; Scheraga, 2000). In this regard, it may be claimed that generating loyal customers online is both more difcult and more important than in ofine retailing. These and other ndings lead prominent theorists to argue that establishing online loyalty is dependent on rst generating consumer trust in the procedural rigor and operational abilities of the supplier (e.g., Reichheld, Markey, & Hopton, 2000; Reichheld & Schefter, 2000; Stewart, 2003). In this sense, establishing and maintaining consumer trust with online suppliers is of paramount importance. These insights have led commentators in the eld to argue that too little is known of the nature and drivers of online loyalty (e.g., Gans, 2002; Grewal, Lindsey-Mullikin, & Munger, forthcoming-b; Reichheld, 2001; Shim, Eastlick, Lotz, & Warrington, 2001). Indeed, the Marketing Science Institute rates research into e-commerce and specically into the determinants of online loyalty as its top priority. However, studies consistently demonstrate that online shoppers distrust not only the
is a greater focus on web-based transactions. Electronic commerce is also frequently called e-commerce (see Hamilton, 2001; Kickul & Gundry, 2001; Lee & Whang, 2001).
0022-4359/$ see front matter 2004 by New York University. Published by Elsevier. All rights reserved. doi:10.1016/j.jretai.2004.04.002
140
L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158
e-tailers themselves (e.g., Fukuyama, 1995; Urban, Sultan, & Qualls, 2000) and their payment systems (e.g., Baker, 1999; Hoffman, Novak, & Peralta, 1999a), but also the very nature of the Internet and online shopping (e.g., Hoffman, Novak, & Peralta, 1999b; Schoder & Yin, 2000). The present research project is designed to contribute insights into these issues in three main ways. First, an objective of our research is to develop and extend existing conceptualizations of service dynamics through incorporating the construct of trust and evaluating its role and importance in driving loyalty after accounting for the well-established antecedent links with service quality, perceived value, and satisfaction. Second, we aim to propose, operationalize, and test a four-dimension scale of loyalty that reects Olivers (1997) conceptualization of a sequential loyalty chain. Third, given the centrality of loyalty and trust to (in particular, online) retailing we intend to test our theory in online contexts and thus contribute additional insights into overall service frameworks and dynamics as well as specic insights into the idiosyncrasies of e-tailing. This article is structured in the following way. First, we briey outline existing frameworks that incorporate quality, satisfaction, value, and loyalty. Second, we present a conceptual model of the antecedents of online loyalty that highlights the fundamental and central role of trust in retailing. Thereafter, we describe the research design, methodology, and the approach adopted. After presenting our results and evaluating our framework, we conclude with a discussion.
the issues, through testing four different models of service. Briey, they nd the strongest support for a framework that models behavioral intentions as anteceded by value, quality, and satisfaction, with value and satisfaction also mediating the quality-intention link. The endorsed frameworks reect conceptual and empirical work that support associations between service quality and perceived value (e.g., Parasuraman & Grewal, 2000; Standird, 2001), service quality and satisfaction (e.g., Heskett, Jones, Loveman, Sasser, & Schlesinger, 1994; Lynch & Ariely, 2000), service quality and loyalty (e.g., Cronin et al., 2000; Sirohi, McLaughlin, & Wittink, 1998), perceived value and loyalty (e.g., Peterson, Balasubramanian, & Bronnenberg, 1997; van Riel, Liljander, & Jurriens, 2001), perceived value and satisfaction (e.g., Bolton & Lemon, 1999; Mathwick, Malhotra, & Rigdon, 2002), and satisfaction and loyalty (e.g., Bern, Mu gica, & Yage, 2001; Reibstein, 2002).
A conceptual framework of loyalty, trust, value, satisfaction, and service quality In order to steer the subsequent review and discussion, a gure identifying the key constructs included in the study is provided in Fig. 1. Our framework presents loyalty as a consequence of direct and indirect relationships with trust, perceived value, satisfaction, and service quality. The conceptual framework outlined in Fig. 1 presents a series of associative links (denoted by solid lines) that have been previously described as well-established in extant literature. However, in incorporating the construct of trust, and positioning trust as pivotal to service dynamics, Fig. 1 also depicts a series of novel, hypothesized associations (denoted by broken lines). Our principal contention centers on our proposal that the framework of loyalty and its antecedents supplied in Fig. 1, constitutes a reliable and valid depiction of the drivers of online loyalty. Briey, the rationale for the forwarded conceptual model is generalized, in part, from the theories and ndings of previously-discussed frameworks of service developed in ofine contexts (in particular, Cronin et al., 2000; Dabholkar, Shepherd, & Thorpe, 2000; Oliver, 1997; Parasuraman & Grewal, 2000), as well as research into the nature of specic service constructs in other contexts (e.g., Cronin & Taylor, 1992; Parasuraman et al., 1994), and specic studies of e-commerce (e.g., Anderson & Srinivasan, 2003; Grewal et al., forthcoming-b; Knights, Noble, Vurdubakis, & Willmott, 2001; Szymanski & Hise, 2000). Our rationale for linkages between individual concepts and further details of theory, on which we rely, is outlined below. Customer loyalty With a focus on repeat purchasing, Brown (1952) forwards a taxonomy of customer loyalty that classies loyalty into undivided, divided, unstable, and no loyalty. Although
Established frameworks of service In ongoing attempts to understand the dynamics of service, services researchers continue to devote considerable attention to loyalty intentions, perceived value, satisfaction, and service quality (e.g., Bolton, Kannan, & Bramlett, 2000; Rust & Oliver, 2000; Sirdeshmukh, Singh, & Sabol, 2002; Szymanski & Henard, 2001). Through building on and extending earlier research (notably, Aaker, 1991; Anderson, Fornell, & Lehmann, 1994), Oliver (1997) forwards a framework of service that presents quality which leads to satisfaction which in turn affects loyalty. The conceptual roots of this view are evident in the subsequent framework of Anderson and Mittal (2000) who argue that such issues are best modeled as a sequential chain of attribute performance leading to customer satisfaction, further affecting customer retention, and subsequently producing prot. In contrast, Parasuraman and Grewal (2000) build on their earlier studies (Baker, Grewal, & Parasuraman, 1994; Grewal, Monroe, & Krishnan, 1998) and the work of Heskett, Sasser, and Schlesinger (1997) to forward what they label an expanded model of customer loyalty that presents a sequential chain of quality-valueloyalty. However, it is the work of Cronin, Brady, and Hult (2000) that constitutes the most comprehensive study of
141
Fig. 1. A conceptual framework of loyalty, trust value, satisfaction, and service quality.
such an emphasis on consumer behavior met with initial acceptance (e.g., Lipstein, 1959), later researchers argued that such views concentrated excessively on behavioral elements to the detriment of the more attitudinal facets of loyalty (e.g., Day, 1969; Jacoby, 1971). Indeed, Gremler (1995) argues that a meaningful measure of customer loyalty cannot omit the measurement of either attitude or behavior. Although Jacoby and Chestnut (1978) and Dick and Basu (1994) make seminal contributions to exploring and elaborating the phases of loyalty, it is the work of Oliver (1997) that constitutes the most comprehensive evaluation of the construct. By building upon and extending earlier work in conceptualizing loyalty, Oliver (1997) forwards a detailed framework of loyalty that presents loyalty as comprising four distinct, sequential phases. First, cognitive loyalty refers to the existence of beliefs that (typically) a brand is preferable to others. Second, affective loyalty reects a favorable attitude or liking based on satised usage. Third, conative loyalty constitutes the development of behavioral intentions characterized by a deeper level of commitment (e.g., Hennig-Thurau, Gwinner, & Gremier, 2002; Janda, Trocchia, & Gwinner, 2002; Zeithaml, Berry, & Parasuraman, 1996). Finally, action loyalty relates to the conversion of intentions to action, accompanied by a willingness to overcome impediments to such action. Hence, contemporary researchers appear to support frameworks of loyalty that incorporate and integrate both behavioral and attitudinal components (e.g., Aaker, 1991, 1999). However, the Oliver (1997) view has yet to be empirically evaluated. Thus: H1. There are four sequential levels of loyalty (respectively; cognitive, affective, conative, and action loyalty).
Trust and loyalty Gundlach and Murphy (1993, p. 41) contend that the variable most universally accepted as a basis of any human interaction or exchange is trust. This view reects the breadth of perspectives adopted in discussions of trust, including; contractual relations theory (Macneil, 1980), interaction theory (Hkansson, 1982), organizational theory (e.g., Bradach & Eccles, 1989), psychology (e.g., Rushton, 1980), social psychology (e.g., Blau, 1964), transaction cost economics (e.g., Nooteboom, Berger, & Noorderhaven, 1997), trust theory (e.g., Gambetta, 1988), and more recently in the study of online exchange (e.g., Lynch, Kent, & Srinivasan, 2001; Stewart, 2003). The emergence of the relational paradigm has contributed to highlighting the importance of trust in buyer-seller relationships in both industrial (Dwyer, Schurr, & Oh, 1987; Morgan & Hunt, 1994) and consumer markets (e.g., Bennett, 1996; Lau & Lee, 1999). In the study of exchange, the construct of trust is most often portrayed as the outcome of reexive consideration/s of the ability of an actor (e.g., a rm or brand) to meet set obligations (see Chaudhuri & Holbrook, 2001; Doney & Cannon, 1997). As such, trust has been presented as a central attribute in relationship initiation, formation, and maintenance in a variety of exchange contexts (e.g., Sirdeshmukh et al., 2002; Verhoef, Francis, & Hoekstra, 2002). Trust has been linked to a variety of outcomes in both the psychology (e.g., Hennig-Thurau & Klee, 1997) and the marketing literatures (e.g., Chaudhuri & Holbrook, 2001). Hennig-Thurau and Klee (1997) theorize that relational attributes will play important roles in repurchasing decisions. Such arguments are supported by the empirical ndings of
142
Eriksson and Vaghult (2000) who nd a strong relationship between relationship satisfaction and retention, while Sirdeshmukh et al. (2002) positions trust as directly linked to loyalty. Lau and Lee (1999) examine the link between consumers trust in a brand and their brand loyalty and nd a signicant positive association. These insights are extended by Chaudhuri and Holbrook (2001) who nd strong evidence in support of a signicant association between brand trust and both purchase and attitudinal loyalty. Sultan and Mooraj (2001, p. 42) argue that trust is central to exchange whether the business is ofine or online. However, indications are emerging that suggest that trust is more important in the online environment than in conventional ofine contexts. Reichheld and Schefter (2000, p. 107) contend that to gain the loyalty of customers, you must rst gain their trust. Thats always been the case, but on the Web . . . its truer than ever, while Grewal et al. (forthcoming-b, p. 18) conclude that customers are asked to trust Internet rms more than their bricks and mortar counterparts. The high importance of trust online appears intensied by the absence of physical contact with online rms and the lack of touch inherent to online exchange (see Lynch et al., 2001; Reichheld & Schefter, 2000). Indeed, the pivotal nature of online trust is acknowledged not only by buyers (see Grewal, Iyer, & Levy, forthcoming-a) and suppliers (see Benassi, 1999), but also by government regulators (e.g., U.S. Federal Trade Commission, 1998) and those concerned about the erosion of constitutional rights (see Shapiro, 2000). In this sense, the importance of trust in electronic contexts has also been consistently argued (e.g., Grewal et al., forthcoming-b; Knights et al., 2001; Reynolds, 2000; Stewart, 2003). In particular, strong evidence has emerged that consumers are especially concerned about payment security and potential fraud (e.g., Hoffman et al., 1999a; Ratnasingham, 1998). Stewart (2003) nds a strong association between trust and intentions to buy, while Lynch et al. (2001) nds that trust is consistently, signicantly associated with online loyalty in a number of contrasting national contexts. However, it is the work of Reichheld et al. (2000) and Reichheld and Schefter (2000) that has been the most inuential in highlighting the signicance of trust in establishing and maintaining loyalty. These views lead to: H2. Trust has a positive inuence on loyalty. Trust and satisfaction The customer satisfaction literature is rmly grounded in the disconrmation paradigm (see Oliver, 1997). While signicant advances in our understanding of satisfaction have been made, Szymanski and Hise (2000) call for more research into the range of drivers of online satisfaction and in particular the trust-satisfaction link. Indeed, such comments appear pertinent regarding both business-to-business (e.g., Bauer, Grether, & Leach, 2002) and consumer (e.g.,
Sirdeshmukh et al., 2002) relational quality. Certainly, a number of commentators have observed or theorized, in accord with social exchange theory (see Blau, 1964), that trust evaluations will exert a direct inuence on perceptions of satisfaction (e.g., Gwinner, Gremler, & Bitner, 1998; Singh & Sirdeshmukh, 2000). Such claims reect the ndings of Szymanski and Hise (2000) that position online nancial security as the second most important factor driving online satisfaction. However, with some notable exceptions, to date, empirical research that focuses on exploring the trust-satisfaction link have largely been limited to studies of fraud or theft (e.g., Furnell & Karweni, 1999; Ratnasingham, 1998). Thus: H3. Satisfaction has a positive inuence on loyalty. Perceived value and trust Perceived value in brick-and-mortar contexts has received much attention in recent years (e.g., Parasuraman & Grewal, 2000; Sirohi et al., 1998; Sweeney, Soutar, & Johnson, 1999). Indeed, Cronin et al. (2000) argue that the study of perceived value (along with service quality and satisfaction) has dominated research in the services literature. While a number of conceptual models of value have been put forward (e.g., Holbrook, 1994; Sweeney & Soutar, 2001), value is most typically presented as acquisition, transaction, in-use, or redemption-based (see Parasuraman & Grewal, 2000; Woodruff, 1997). Some conceptual support and limited empirical evidence also emerges for an association between perceived value and trust (e.g., Singh & Sirdeshmukh, 2000). Indeed, Anderson and Srinivasan (2003) argue that online perceived value and trust exert similar inuences with regard to linkages between satisfaction and loyalty. In contrast, Grewal et al. (forthcoming-b) deconstruct online value into perceived convenience benets on the one hand and perceived risk, to denote the sacricial element of value, on the other. The importance of trust is explored by Grabner-Kraeuter (2002) who theorizes that through generating trust in online purchasing, retailers add value for consumers through reducing complexity and thus (in part) diminish the uncertainty linked with transactions (a view in accordance with Ratnasingham, 1998). However, it is the work of Sirdeshmukh et al. (2002) that provides the most detailed empirical evaluation of perceived value-trust, to date. Sirdeshmukh et al. (2002) nd some support for direct associations between trust and value, although they are forced to conclude that more research is needed to evaluate the nature of the link between value and trust. Thus: H4. Perceived value has a positive inuence on loyalty. Service quality and trust Through developing the earlier work of Lehtinen and Lehtinen (1982) and Grnroos (1984) presents a disconr-
143
mation model of technical and functional quality intervened by corporate image. However, it is the seminal studies of Parasuraman, Zeithaml, and Berry (1985, 1988, 1991, 1994) that are widely recognized as the most comprehensive evaluations of service quality. Briey, Parasuraman et al. (1988, 1991) identify reliability, responsiveness, assurance, empathy, and tangibles as the core dimensions of service quality. These developments have aided research into the consequences of service quality for both suppliers and buyers (e.g., Cronin et al., 2000; Dabholkar et al., 2000). The issue of service quality appears especially important in online contexts, where it is estimated that online retailers lose nearly $21 billion per annum due to poor service (Datamonitor, 2001) and where service complaints double in number each year (N.A.C.A.A., 2001). The importance of online service quality is, in part, driven by research that associates service quality with trust. A number of studies have attempted to integrate measures of perceptions of trust and security into broad measures of overall e-service quality (e.g., Janda et al., 2002; Kaynama & Black, 2000; Liljander, van Riel, & Pura, 2002; Zeithaml, Parasuraman, & Malhotra, 2000a). However, existing studies have also explored the direct links between quality and trust (e.g., Chen et al., 2002; Sultan & Mooraj, 2001). In the businessto-business context, Sultan and Mooraj (2001) nd links between a number of service quality factors and trust while arguing that service quality issues contribute to each of the 10 major, online trustbusters. In contrast, Chen et al. (2002) nd that service quality is not signicantly linked to trust, in the online pharmacy sector, and suggest that product category effects inuence the link between quality and trust. Nevertheless, a broad range of earlier studies have argued that overall relationship quality is driven by service quality (see Anderson & Sullivan, 1993; Hennig-Thurau & Klee, 1997). In this regard, Hennig-Thurau and Klee (1997) theorize that perceptions of overall quality drive online trust. Thus: H5. Service quality has a positive inuence on trust. Data collection and sampling We tested our framework by surveying customers in two distinct key online markets. This research design reects our desire to pursue a robust testing of hypothesized relationships and is similar in approach to the study of Sirdeshmukh et al. (2002). A review of extant literature leads to the conclusion that the online markets for books and ights are not only the largest and fastest-growing online markets (e.g., Brynjolfsson & Smith, 2000; Forrester Research, 2000; Li & Gery, 2000), but that they also constitute two of the few contexts wherein initial growth expectations have been matched by subsequent, substantive market growth (Listanti, 2001; Wingeld, 2001). Hence, in Study One, we surveyed online purchasers of books while in Study Two we studied online buyers of ights.
While the justication for the study of online books and ights shoppers is grounded in a conceptual rationale, the differences between these two distinct contexts, potentially, generates practical complications for the study of service dynamics. For example, a number of leading commentators have suggested that service dynamics can vary tremendously from industry to industry (e.g., Jones & Sasser, 1995), product category to category (e.g., Wolnbarger & Gilly, 2001), country to country (e.g., Lynch et al., 2001), and especially so in e-commerce (e.g., Chen et al., 2002). In this regard, theorists have argued that the relative importance of particular constructs (such as satisfaction) may vary from one context to another depending on the nature of markets and exchange (e.g., Grewal et al., forthcoming-b; Oliva, Oliver, & MacMillan, 1992). Further, the relationship between service dynamics constructs may also be affected by the idiosyncrasies of the sectors concerned (see, e.g., Szymanski & Hise, 2000; Wolnbarger & Gilly, 2001). Finally, recent structural and operational changes within markets may also produce context-specic ndings. This is especially the case in the study of the ights market, wherein consumers and operators have experience substantial changes in recent years (largely following the tragic events of September 11th). In each of the two studies, a different, single, online supplier was surveyed. In Study One, the company is identied by the pseudonym Books.com while in Study Two, the pseudonym Flights.com is used. Books.com is one of the market leaders for online books purchasing while Flights.com is one of the top three online ight retailers, retailing tickets for a wide variety of airlines. Currently, sales for both rms are growing notably faster than the market average. Our rationale for focusing on the customers of a single online supplier is based on the conclusions of research into customer service and loyalty (e.g., Pedersen & Nysveen, 2001; Sirohi et al., 1998), online consumer behavior (see Rothaermel & Sugiyama, 2001; Standird, 2001), and retailing (e.g., Garretson, Fisher, & Burton, 2002; Sirohi et al., 1998) which indicates that focusing on single suppliers may lead to insights that are missed in surveys of multiple suppliers.
Study One: customers of Books.com In Study One, a sample of 750 potential respondents was obtained from a research agency from its pre-existing database of online book purchasers. Potential respondents were randomly drawn from those customers who had purchased online repeatedly and had purchased at least one book from Books.com within the last six months. Prospective respondents were contacted by telephone to obtain agreement for involvement and to arrange a mutually convenient date, time, and location for a face-to-face meeting to administer the questionnaire. The survey instrument was administered to respondents in a face-to-face meeting with one of a team of four researchers. These researchers were experienced in
144
eld research and were trained by one of the authors over a two-day period. The same author also supervised the rst ve administrations of each researcher. Of the 750 respondents contacted, 342 customers initially agreed to participate. Of the 342, 11 respondents later withdrew their support, 23 were unable to nish the entire questionnaire while 12 questionnaires were spoiled or inappropriately completed. Eventually, 294 completed questionnaires were obtained, yielding a response rate of just over 44 percent (using the calculation method recommended by C.A.S.R.O., 1982).
Study Two: Flights.com To improve the generalizeability of ndings and the validity of conclusions and implications, Study Two adopted a slightly different survey approach, although each used the same research instrument (albeit tailored to either books or ightssee Appendices AC). In Study Two, a randomly selected sample of 1000 recent online purchasers of ights from Flights.com was purchased from a different data brokerage agency from that used in the rst study. The sample for the second study was compared (by name and by zip code) to the sample of the rst study to ensure that the samples did not overlap (three cases were found and excluded from the second sample). Without any prior contact to elicit support, each of the remaining respondents was mailed a personalized, individually signed letter of introduction, the survey instrument, and a pre-paid return envelope. To improve response rates, participation was encouraged with eligibility for inclusion in a prize draw for $150. Reminder cards were mailed to eligible non-respondents 15 days after the initial mailing. Of the questionnaires mailed, 244 questionnaires were received from eligible respondents, of which 40 were unusable due to incomplete or inadequate responses. After accounting for undeliverable packets, this generates a response rate of over 24 percent (C.A.S.R.O., 1982). This compares favorably to response rates of mailed surveys conducted on similar issues (e.g., Jones, Mothersbaugh, & Beatty, 2000; Tax, Brown, & Chandrashekaran, 1998). The differences between the response rates of the two studies is most probably a reection of the data collection processes, with the higher useable response rate for the face-to-face administered survey in Study One reecting the comparative success of a personalized and guided approach.
compared to non-customers of both rms.2 To enable meaningful comparisons to be made, 50 fully-completed surveys were obtained by administering the questionnaire to noncustomers of Books.com, while 53 fully-completed postal surveys were received from non-customers of Flights.com (both randomly selected from a commercial database of online consumers). Subsequently, the socio-demographic characteristics and the developed key constructs (loyalty, satisfaction, perceived value, service quality, and trust) of the samples of customers and non-customers of both companies were compared rst via chi-squared for categorical data, and second by Student t-tests for the continuous data. These tests revealed no statistically signicant differences between these samples at the 1 percent level. Following the recommendations of Armstrong and Overton (1977), in both studies, the responses of early and late respondents were compared to gauge non-response bias and no signicant differences (p < .01) were found among any of the variables or factors used in later analyses. To gauge the authenticity of respondents, a randomly selected sample of useable questionnaires was taken from both samples and evaluated. This evaluation involved re-contacting respondents by telephone and requesting conrmation of socio-demographic data. This evaluation found no signicant discrepancies. Analysis of responses indicates that the demographic characteristics were comparable across the books and the ights samples (although ight purchasers tended to be slightly older and wealthier). For the total sample, the gender split was fairly equally divided between males and females (52.1 percent male). Just under half of the sample were currently married (47.3 percent) while over half of the sample had children (53.2 percent), were over 35 years of age (78.3 percent), and had a family income of over $120,000 per year (72.2 percent). Over 16 percent used their workplace computer to purchase online goods (compared to 76.1 percent at home and 1.4 percent at an Internet caf). Purchasing online was almost exclusively through credit cards (96.7 percent) while average weekly online hours were a surprisingly high 8.5 hr per week, as measured by the median.
Measures and measurement properties Following similar approaches in other studies of multiple industries (e.g., Jones et al., 2000), the items used in the two surveys were identical with the exception of changes of the rm names, industry names, and other minor re2 While this approach to evaluating sample selection bias gauges potential bias from all online customers to customers of the single retailers, bias involved in going from all customers to online customers is not examined. This constitutes a limitation of the study that future studies could overcome through the employment of multiple sample selection bias evaluations.
Sample evaluation Given the research focus on consumers of a single online retailer in each industry, it was deemed necessary to evaluate potential sample selection bias. First, the obtained samples of customers of Books.com and Flights.com were
145
wordings (largely to the anchor questionssee Appendices AC). Although a considerable body of research into loyalty, satisfaction, value, service quality, and trust has been undertaken (see earlier), in order to gauge these concepts in an online context, scale development and adaptation was required. The scales employed were developed using conventional psychometric procedures (Anderson & Gerbing, 1988; Churchill, 1979) and were based on existing scales (largely in ofine contexts) as well as on scale development work conducted during pre-testing. Briey, our scale development procedure involved the development of survey instruments containing measures of loyalty, satisfaction, value, service quality, and trust derived largely from existing measures (see below). First, the questionnaire was subjected to critical review by a panel of 10 senior Internet managers and 5 academicians. After a number of suggested amendments (largely to item wording but also to question order and layout), the modied survey instruments were pretested on consumers. In total, 50 questionnaires were administered regarding Flights.com and 50 for Books.com, with detailed notes taken regarding difculties, criticisms, and suggestions. After a nal critical review, senior faculty members concluded that there were no substantive concerns and the instruments were deemed suitable for nal data collection. As discussed earlier, the concept and construct of service quality has been extensively studied, largely in brick-andmortar contexts (e.g., Cronin et al., 2000; Cronin & Taylor, 1992; Dabholkar et al., 2000), although more recent work has also been conducted regarding online service quality (e.g., Zeithaml, Parasuraman, & Malhotra, 2000b, 2002). A review of such literature supports a focus on perceptions of performance in service quality measures (see Zeithaml et al., 1996). Accordingly, the 22 item scale of Cronin and Taylor (1992) was adapted to the online context, involving extensive rewording and the omission of one item that was inappropriate for this context. The perceived value scale is constructed to measure online purchasers overall assessments of product and service utility based on what is exchanged (a view of perceived value grounded in the conclusions of Zeithaml, 1988). The scale used in the current study is adapted from the scales of Brady and Cronin (2001) and Mathwick et al. (2002). Satisfaction was measured using the scale of Cronin et al. (2000) since it was felt that the two item scale use by Szymanski and Hise (2000) did not sufciently encapsulate the concept of online satisfaction. Thus, an adaptation of the ve item Cronin et al. (2000) scale of satisfaction was adopted. The trust scale is designed to gauge the extent to which customers have condence and faith in the integrity, service, and brands of Books.com or Flights.com. In the current study, the scale of online trust employed is based on the trust measure of Hess (1995) that was originally designed to gauge customers trust in car sales. A review of the avail-
able trust measures3 found that Hesss (1995) measure of perceived brand trust was not only the most applicable to the focus of the studies on specic company brands, but also was explicitly perceptual in nature, designed to gauge consumers trust (and not, e.g., managers trust of another rm), and the most easily transferable and adaptable to an online context. Loyalty has been measured from a variety of perspectives, although recent studies appear biased toward the behavioral intention phase of loyalty (e.g., Cronin et al., 2000; Zeithaml et al., 1996). In this regard, extant measures of customer loyalty are unidimensional, uni-phased/staged, and typically overly-focused on what Oliver (1997) conceives as the early stages of loyalty. As detailed earlier, we nd the arguments and view of Oliver (1997) both compelling and convincing. Thus, guided by the view of Oliver (1997), the items for capturing loyalty mirror this four-stage loyalty framework with four items each designed to evaluate cognitive, affective, conative, and action loyalty (see Appendix C). The precise wording of these items is grounded in our qualitative data and pretesting procedures. Reliability was examined through Conrmatory Factor Analysis (CFA) and the calculation of Cronbach alpha coefcients (Cronbach, 1951). CFA was conducted for both surveys. In both studies, each indicator loaded signicantly on its designated factor (p < .01). Overall, in Study One (of Books.com) CFA produced chi-squared-degrees of freedom ratio well below the criterion of Marsh and Hocevar (1985) with adjusted goodness-of-t (AGFA) signicantly better than a one factor model. Similarly, in Study Two (of Flights.com) CFA generated a chi-squared-degrees of freedom ratio that again was well below the criterion of Marsh and Hocevar (1985) matched by an AGFA signicantly better than a one factor model. Reliability was also gauged via the Cronbach alpha coefcient (Cronbach, 1951) that Nunnally (1978) and Churchill (1979) suggest should be over .7 for a scale to be considered reliable. The Cronbach alpha coefcients range from .69 to .97 (see Appendices AC). Although for cognitive and conative loyalty for the ights survey, the .70 threshold is just missed (at .69), in both cases the variables are very close to the threshold value of .70 and far above the criterion in use level of .60 reported by Peterson (1994) and Slater (1995) and considerably higher than the .50 threshold originally advocated by Nunnally (1967). The assessment of the validity of each scale was conducted via item-total correlation analyses (Nunnally, 1967), the conditions of ac-
3 A review of the trust literature nds a wide variety of trust measures including the Larzelere and Huston (1980) measure of interpersonal trust (adapted by Morgan & Hunt, 1994 to measure interorganizational trust), the Sirdeshmukh et al. (2002) measure of employee and management policy trustworthiness, the Cummings and Bromily (1996) long (62 item) and short (12 item) Organizational Trust Inventory, as well as a host of less comprehensive measures (e.g. Chaudhuri & Holbrook, 2001; Lynch et al., 2001).
146
L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158 Table 1 Correlation table Cognitive loyalty (a) Study One: Books.com Cognitive loyalty 1.00 Affective loyalty .52 Conative loyalty .45 Action loyalty .43 (b) Study Two: Flights.com Cognitive loyalty 1.00 Affective loyalty .57 Conative loyalty .47 Action loyalty .49

ceptance being that correlations were signicantly high and in the expected direction (Cook & Campbell, 1979). In all cases, the correlations met these criteria. Hair, Anderson, Tatham, and Black (1998) note that construct validity (convergent and discriminant validity) can be gauged in a variety of ways. Given that our research approach involved two studies using slightly differing methods but with both measured the same multiple traits, construct validity was rst gauged using the multitrait-multimethod approach (see Hair et al., 1998). The results of this analysis provide evidence of convergent validity through nding that the correlations between the key constructs measured are sufciently large, positive and all signicant at the .01 percentage level. Discriminant validity is indicated by (1) the results in the heteromethods block that demonstrate that entries in the validity diagonal are higher than the coefcients in the same column and row of the heteromethod block (Campbell & Fiske, 1959), (2) the nding that the validity coefcients are larger than the correlations in the heterotriat-monomethods triangles, and (3) evidence of discriminant validity is found in the similarity of patterns (rank and direction) of the correlations in the heterotrait triangles (Churchill, 1979; Ford, Walker, & Churchill, 1975). However, more robust evidence of discriminant validity was found through the analysis of chi-squared difference tests, in which the correlations between all possible pairs of constructs are once freely estimated and once set to unity (Gerbing & Anderson, 1988). All chi-square differences for rst-order constructs were signicant at the .05 level, suggesting that these measures are not collinear. The lowest differences were between conative and affective loyalty in both 2 = 13.41, Study Two studies (Study One of Books.com ( 1) 2 = 5.32). These results suggest that, in of Flights.com ( 1) both studies, the constructs under analysis are distinct and discriminately valid. Results H1 states that there are four sequential levels of loyalty (respectively; cognitive, affective, conative, and action loyalty). To test this hypothesis a three-stage approach was adopted in the analysis of both Study One and Study Two. First, preliminary correlation analysis was undertaken to explore potential linear relationships between each loyalty construct (see Table 1a and b). In both studies, the links between the four loyalty measures were positive and statistically signicant. However, while these basic tests provide tentative support for associations between the loyalty constructs, such bivariate analyses often over-estimate links between concepts and further do not gauge sequentiality or causality. Consequently, the second phase utilized a form of path analysis, recommended by both Duncan (1975) and Pendhazur (1982). This approach involves the calculating of a series of regression equations between each phase of loyalty in
Affective loyalty
Conative loyalty
Action loyalty
1.00 .69 .21
1.00 .24
1.00
1.00 .70 .46
1.00 .34
1.00
Statistically signicant at the 5% level: two-tailed test. Statistically signicant at the .1% level: two-tailed test.
both studies. In Study One, the estimated parameters between cognitive-affective loyalty, affective-conative loyalty, and between conative-action loyalty are positive in each case ( .49, .70, .25, respectively) and statistically signicant (p < .01). In Study Two, a similar pattern of results was found ( .59, .68, .26, respectively; p < .01). Such ndings support H1. Although the simple path analysis plotted constitutes a more rigorous evaluation of the relations between the loyalty constructs than simple correlation analysis, to evaluate the sequentiality and causality of loyalty, structural equation modeling was required. James, Mulaik, and Brett (1983) and Jreskog and Srbom (1989) both recommend that to evaluate the sequential order of constructs, structural equation modeling should be undertaken for each potential order sequence. Thus, for both studies, all possible sequential models of loyalty were mapped (a total of 24 structural equations for each study, 48 in total). In Study One (of Books.com), the results of this analysis indicate that the theorized sequence of loyalty phases (cognitive-affective-conative-action loyalty) produced the most robust and valid model of all the 24 models evaluated (through comparative analysis of chi-squared). These results are replicated in Study Two (of Flights.com) where the hypothesized sequential order is the best-t when compared to the other 23 alternative models. The procedure adopted to evaluate H1 provides much support for the hypothesized sequential relationships. In particular, the nding that the theorized loyalty sequences produce the best t with the data in both Study One and in Study Two when compared to the 46 other potential sequences (see James et al., 1983; Jreskog & Srbom, 1989), strongly supports H1. Thus, H1 is found to be supported and is accepted. Service dynamics Given the nature of the associations between the four components of loyalty (see H1) it was deemed appropriate to estimate a second-order factor of loyalty that incorporated cognitive, affective, conative, and action loyalty. CFA generated a robust overall loyalty factor that conformed to
147
reliability and validity requirements. The discriminant validity of the constructs used in structural modeling was found through the analysis of chi-squared difference tests (Gerbing & Anderson, 1988). All chi-square differences were signicant at the .05 level, suggesting that these measures are not collinear. The lowest differences were between trust and loyalty in both studies (Study One of Books.com 2 = 14.33, Study Two of Flights.com 2 = 15.49). ( 1) (1) To facilitate the analysis of both the previously established and the hypothesized relationships, the data generated during Study One (of Books.com) and Study Two (of Flights.com) were rst tentatively explored using correlation matrices and later analyzed using structural equation LISREL models. Data from Study One and Study Two were not pooled, since analyses4 suggested that such an approach may disguise signicant sectoral differences and obscure empirical complexities. Although in some regards this approach may constrain generalizeability, the analysis of two separate models generates deeper and contextually more complete understanding of service dynamics in these two very different sectors. To explore potential associations between constructs, data for Study One (of Books.com) and Study Two (of Flights.com) were rst analyzed via the study of bivariate zero-order correlation coefcients. The results of these equations are shown in Table 2. However, as zero-order correlation analysis can lead to an over-estimation of the importance of the relationships within the model, a structural equation technique, LISREL, was employed to test associations. Structural equation models overcome the limitations of bivariate analyses through the simultaneous analysis of all the complex relationships between the constructs. The results of this approach are presented in tabular form (see Table 3) as well as modeled diagrammatically in Fig. 2. The chi-squared divided by the degrees of freedom ratio for each of the research models were 4.98 for Study One (Books.com) and 2.36 for Study Two (Flights.com). Both these values are acceptable as they are below the recommended level of 5 suggested by the Marsh and Hocevar (1985). A number of goodness of t indices for both models are also reported, namely Bollens (1989) incremental t index (IFI) and Bentlers (1990) comparative t index (CFI).
Table 2 Correlation table Loyalty (a) Study One: Books.com Loyalty 1.00 Service quality .41 Perceived value .66 Satisfaction .69 Trust .72 (b) Study Two: Flights.com Loyalty 1.00 Service quality .28 Perceived value .76 Satisfaction .18 Trust .76

Service quality
Perceived value
Satisfaction
Trust
1.00 .34 .51 .44
1.00 .59 .61
1.00 .66
1.00
1.00 .24 .17 .23
1.00 .03 .78
1.00 .04
1.00
Statistically signicant at the 5% level: two-tailed test. Statistically signicant at the .1% level: two-tailed test.
For both the structural equation models these goodness of t indices are close to or above the minimum recommend levels. Therefore, based on these diagnostic tests, both models exceed the recommended validity thresholds and are accepted. Established relationships First, the link between service quality and perceived value has been demonstrated to be consistently signicant
Table 3 Standardized LISREL estimates for Study One and for Study Two Structural links Hypothesis 2 Trust to loyalty Hypothesis 3 Trust to satisfaction Hypothesis 4 Perceived value to trust Hypothesis 5 Service quality to trust Established relationships Perceived value to loyalty Satisfaction to loyalty Service quality to perceived value Service quality to satisfaction Diagnostic statistics Chi-squared Degrees of freedom Probability level (p) Number of observations Incremental t index (IFI) Comparative t index (CFI) Root mean squared residuals Study One: Books.com .36 .54 .51 .27 .28 .29 .34 .27 9.96 2 .007 294 .968 .966 .089 Study Two: Flights.com .39 n/s .75 n/s .48 n/s .35 n/s 4.72 2 .000 204 .989 .994 .011
4 The data for the two e-commerce sectors was evaluated to ascertain whether the responses to the two studies should be pooled, or whether two separate models should be used (Bass & Wittink, 1975; Gujarati, 1988). The major advantage of pooling the data in this way is to improve the precision of the estimated parameters (Gujarati, 1988) and widen generalizeability. However, such an approach, potentially, obscures signicant sectoral differences and disguises empirical complexities. In this regard, as the two studies analyze very different sectors (namely ights and books) it would be very surprising if there were no sector-specic differences. As expected, Student t-tests between the same construct for the two surveys were, in all cases, statistically signicant at (p < .01). These results indicate that the constructs in the two different sectors are not similar and that therefore, as anticipated, two individual structural equation models should be used to model these two sectors.
n/s: positive but not statistically signicant. Statistical signicance at the 1% level. Statistical signicance at the .1% level.
148
Fig. 2. The empirical result for Flights.com and Books.com studies.
in a variety of contexts (e.g., Parasuraman & Grewal, 2000; Rothaermel & Sugiyama, 2001; Standird, 2001; Walsh & Godfrey, 2000). The results of both Study One and of Study Two strongly support this view. Preliminary analysis leads to positive and signicant coefcients between quality and perceived value in the studies of both Books.com and Flights.com (see Table 2). Structural equation modeling corroborates the stability of this link through conrming positive, signicant links (p < .01) (see Table 3) between service quality and perceived value for both Study One and Study Two (see Fig. 2). Similar results also emerge for the linkage between the constructs of perceived value and loyalty that are wellestablished in the existing literature (e.g., Cronin et al., 2000; Peterson et al., 1997; Sirdeshmukh et al., 2002). Correlation analysis for Study One and for Study Two indicates a strong and positive association between perceived value and loyalty (see Table 2). Subsequent multivariate analysis generally supports these ndings with a positive link to loyalty (see Table 3). However, the results are less clear in the cases of the well-established relationships between service quality and satisfaction (see e.g., Heskett et al., 1994; Lynch & Ariely, 2000; Szymanski & Hise, 2000), and between satisfaction and loyalty (see e.g., Bern et al., 2001; Reibstein, 2002; van Riel et al., 2001). In Study One (Books.com), correlation analysis nds signicant coefcients between service quality and satisfaction and between satisfaction and loyalty (see Table 2). Further support emerges during structural equation modeling (see Table 3 and Fig. 2) with signicant linkages found. However, this pattern of results is not found in Study Two (of Flights.com). The analysis of Study Two using both correlation and structural equation modeling nds the links
between service quality and satisfaction and between satisfaction and loyalty to be non-signicant. The lack of statistically signicant links between service quality and satisfaction and between satisfaction and loyalty in Study Two is contrary to a range of extant theories and ndings (e.g., Bern et al., 2001; Lynch & Ariely, 2000; van Riel et al., 2001). However, given the context of Study Two (of Flights.com), such results are both predictable and understandable. Oliva et al. (1992) observe that for rms in the ights sector (such as Flights.com), service quality is not necessarily linked to satisfaction and loyalty since consumers expectations regarding service are capricious, ckle, and unpredictable. More recently, Grewal et al. (forthcoming-b, p. 14) argue that the ights market (including Flights.com) is dominated by loyalty-schemes and is a good example of a market wherein consumers are apt to exhibit loyalty across lower levels of satisfaction. Such ndings and theories provide a coherent and logical explanation for the results of Study Two and the lack of signicant associations between service quality and between satisfaction and loyalty through demonstrating that such ndings can be attributed to the market dynamics and online context of the study of Flights.com. Predicted relationships H2 argues that there is a positive and direct association between trust and loyalty. The provisional evaluation of potential linkages using correlation analysis (see Table 2) indicates strong support for this claim, with trust positively and signicantly correlated with loyalty in both Study One and in Study Two. Structural equation modeling also broadly supports H2, producing signicant parameter estimates. In
149
Study One (of Books.com), some support is found for H2, as trust is signicantly linked to loyalty (see Table 3 and Fig. 2). For Study Two, the structural equation modeling (see Table 3 and Fig. 2) produces unequivocal results in support of H2. Trust is found to be directly associated with loyalty, corroborating the theories that suggest that trust is central to service dynamics (e.g., Hoffman et al., 1999a; Stewart, 2003), is a key driver of customer loyalty (e.g., Lynch et al., 2001; Reichheld & Schefter, 2000; Reynolds, 2000), and is especially important in online contexts (e.g., Grewal et al., forthcoming-b; Sultan & Mooraj, 2001). Hence, sufcient evidence is found to claim that H2 is supported. H3 reasons that trust is positively and directly linked with evaluations of satisfaction. In Study One, this contention is strongly supported. Initial investigations using correlation analysis nd trust to be positively (r = .66) and signicantly (p < .001) linked (see Table 2). Further, structural equation modeling conrms the stability of the association within a broader framework of service dynamics (see Table 3 and Fig. 2), producing a positive signicant relationship ( = .54, p < .001). In this regard, Study One supports H3. However, the results of Study One (of Books.com) are not replicated in Study Two (of Flights.com). In Study Two, trust and satisfaction are not signicantly correlated (see Table 2) and structural links are not found (see Table 3 and Fig. 2). Therefore, given the varied results, H3 is only partially accepted. The nding of associations in the study of Books.com and not in the study of Flights.com deserves further investigation. Our ndings in the study of books conform to early theory that suggests that trust drives satisfaction, especially in online contexts (Reichheld et al., 2000; Szymanski & Hise, 2000). However, in the study of Flights.com (Study Two), trust appears not to be directly linked. There are four main explanations for this nding. First, as mentioned earlier, Grewal et al. (forthcoming-b) theorize that the ights market, in which Flights.com operates, is dominated by loyalty programs that downgrade the relative importance of customer satisfaction (and arguably strengthen trustloyalty linkagessee H2). Second, Szymanski and Hise (2000, p. 319), two of the leading commentators regarding e-satisfaction, conclude that online satisfaction may not be stable in all contexts and that further research is needed to evaluate the efcacy and generalizeability of the concept. In Study Two, our results cast doubt on the importance of satisfaction for Flights.com, in a manner consistent with the earlier comments of Oliva et al. (1992) regarding the broader ights market. Third, in recent years both the government (Power, 2002) and airline and airport operators (Prada, 2002a) have made considerable, high-prole, and widely-publicized efforts to increase consumers overall trust. While such efforts appear to engender trust, many of these procedures erode reexive evaluations of satisfaction (e.g., longer check-in times, additional searches, passenger photographing, etc.) (e.g., Alexander, 2003; Barnes, 2003). Thus, although customers may trust Flights.com, such trust may not drive satisfaction. Finally, the results in Study Two
may reect a trend identied by Oliver (1999), wherein satisfaction is downgraded in relative importance when compared to other service constructs (particularly loyalty but in the current studies, also trust). H4 argues that there is a positive and direct relationship between perceived value and trust. Preliminary correlation analysis appears to support this suggestion in both studies. In Study One, perceived value and trust are positively (r = .61) and signicantly (p < .001) correlated (see Table 2). Similarly, in Study Two (of Flights.com) value and trust are very strongly linked (r = .78, p < .001). These ndings are corroborated in the structural equation models in both studies (see Table 3 and Fig. 2). In Study One and in Study Two perceived value is found to be positively (respectively; = .51, .75) and very signicantly (p < .001) associated with trust. These results directly support the earlier theoretical work by Singh and Sirdeshmukh (2000), Grabner-Kraeuter (2002), and Urban et al. (2000). Hence, overwhelming evidence is found to declare H4 supported. H5 contends that there is a positive and direct relationship between service quality and trust. Study One generates strong support for this view with a positive (r = .44) and signicant (p < .001) bivariate association (see Table 2), conrmed in structural modeling results ( = .27, p < .01). Thus, consistent with a range of existing studies (e.g., Hennig-Thurau & Klee, 1997; Liljander et al., 2002; Sultan & Mooraj, 2001), in the study of Books.com, service quality appears to be an important driver of trust as well as indirectly linked with loyalty (see H2, Fig. 2). However, as in H3, these results are not replicated in Study Two (of Flights.com). In Study Two, although service quality and trust signicantly correlate (r = .23, p < .05), structural modeling (see Table 3 and Fig. 2) suggests that the coefcient may be an over-estimation, since no signicant structural link is found (p > .05). Therefore, given the mixed results, H5 is only partially accepted. The lack of a consistent signicant link between service quality and trust in the study of Flights.com merits further evaluation. Briey, four key reasons appear to contribute to this nding. First, although a number of researchers have theorized that service quality drives trust online (e.g., Hennig-Thurau & Klee, 1997), recent conceptual work has questioned the universality of this link (e.g., Wolnbarger & Gilly, 2001) while empirical work has found contrary results (see Chen et al., 2002). Second, the precise nature of the context of Study Two may explain the differentiated results regarding quality-trust. Flights.com retails a wide range of tickets for a variety of airline operators, some of which focus on very low cost, low service, but perceived high value strategies (Boyeld, 2001; Prada, 2002b). Thus, a proportion of customers of Flights.com are more attentive to value than to service quality. Hence, service quality evaluations are downgraded in relative importance when compared to perceived value (a view supported by the strength of associations between perceived value and both loyalty and trust in Study Two). Third, as mentioned previously, recent changes to the
150
overall ights market (including Flights.com) have seen a variety of service changes that have increased overall trust but eroded service quality (see Barnes, 2003). Finally, stemming in part from September 11th, over-capacity and falling revenue in the ights market has led to widespread cost-cutting, particularly in the form of service personnel cutbacks and reduced ight frequencies, and routes serviced (McCartney, 2002; Prada, 2001). Such strategies have dramatically reduced costs but have also eroded customer choice and overall service quality. In this regard, it would appear that many rms in the airline industry (such as Flights.com) have prioritized trust and value above service quality.
Discussion The review of extant research positions trust as a pivotal driver of loyalty which is central to the understanding and modeling of service encounters. However, while a number of prominent commentators (e.g., Grewal et al., forthcoming-b; Hoffman et al., 1999a; Reichheld & Schefter, 2000; Reynolds, 2000) have repeatedly emphasized the importance of loyalty and trust, empirical research into their dynamics remains in its infancy (e.g., Srinivasan, Anderson, & Ponnavolu, 2002; Stewart, 2003). The current study constitutes one of the rst tentative steps toward a greater appreciation of these issues. In this regard, we forward and test a framework of loyalty and hypothesize a series of associations with trust, as one of the key drivers of loyalty. Analysis of the results of two surveys of online purchasing reveals a variety of positive and statistically signicant relationships. The remainder of the text is devoted to a review of the implications of these ndings as well as to a discussion of the limitations of the study and the implications for future research. The rst contribution of this study stems from the successful operationalization of a four-facet measure of loyalty based in the original perspective of Oliver (1997). In what is already a seminal text, Oliver (1997) develops a conception of loyalty comprising sequential phases of cognitive, affective, conative, and action loyalty. While this view of loyalty has been commended before the current study, no attempt had been made to operationalize this construct in a conventional context let alone in an online environment. Thus, a contribution of the current work is the development and successful testing of measures of cognitive, affective, conative, and action loyalty. Indeed, analysis of the results of our two studies indicates that our four, four item loyalty measures appear both reliable and valid. While it is acknowledged that our research design and LISREL methodology precludes causal claims of sequential phases of loyalty, the results of simple path analysis (Duncan, 1975; Pendhazur, 1982) as well as analysis procedures advocated by James et al. (1983) and Jreskog and Srbom (1989) both strongly
support the a priori sequential, causal model presented, as well as the theory of Oliver (1997). Our second contribution centers on our nding that trust plays a pivotal role in service dynamics and, in particular, in directly and indirectly driving loyalty. While a number of sector commentators have previously highlighted the importance of trust online (see Bauer et al., 2002; Reichheld et al., 2000) and found consumers to be deeply concerned about online fraud (see Furnell & Karweni, 1999; Hoffman et al., 1999a), to date, empirical studies of the links between trust and loyalty have been limited in scope (cf. Stewart, 2003). The results of the current studies strongly support the view that trust is a key and central factor during exchange, after accounting for previously established antecedents, namely; perceived value, satisfaction, and service quality (see Table 3). In this sense, not only do these ndings support a number of earlier studies that have argued that trust is central to exchange (e.g., Nooteboom et al., 1997; Verhoef et al., 2002) but also our results support the contentions of a number of theorists who have suggested that trust may be more important online (e.g., Grewal et al., forthcoming-b; Reichheld & Schefter, 2000). This in turn appears supportive of the overall view of online exchange as relational rather than transactional in nature. Indeed, it could be argued (somewhat contentiously) that loyalty is inherently relational in character, especially when contrasted to the (comparatively) more transactional constructs of satisfaction and (to some extent) perceived value. Third, we contribute insights into the idiosyncrasies of online e-tailing service dynamics. Our study nds that, consistent with studies of ofine exchange, service quality exerts an indirect inuence on loyalty, while perceived value exhibits both direct and indirect associations with online loyalty. However, differentiated results are found in relation to satisfaction. In Study One (of Books.com) our results are broadly concordant with earlier brick-and-mortar studies (e.g., Anderson & Mittal, 2000; Dabholkar et al., 2000; Oliver, 1997) and nd a quality-satisfaction-loyalty chain. However, in Study Two, our results do not support this view. These results are attributable to sector differences, in a manner consistent with a number of leading commentators (e.g., Grewal et al., forthcoming-b; Jones & Sasser, 1995; Lynch et al., 2001; Oliva et al., 1992). Such ndings add support to the argument of Oliver (1999) that satisfaction is often over-emphasized and bolster suggestions that service factors (including satisfaction and trust) are both market- and context-sensitive. Our results also provide empirical support to a number of existing studies that have emphasized the importance of examining, concurrently, a range of online service factors, including; perceived value, trust, satisfaction, and service quality (e.g., Mathwick et al., 2002; Szymanski & Hise, 2000; Zeithaml et al., 2002). In this regard, our study contributes a holistic view of service dynamics (see Fig. 1) that incorporates the main service factors into a tested framework of service. In doing so, our study draws on extant
151
studies of online exchange, as well as research into ofine environments, culminating in the (tentative) afrmation of the model. In this sense, general support is found in defense of theories that have drawn parallels between on- and ofine service dynamics. The results also support and build on existing research into loyalty and further extend the generalizeability of such research into the context of online service. Thus, the results corroborate existing frameworks of service that typically present some variation of a service qualityvalue/satisfaction-loyalty chain or model (e.g., Cronin et al., 2000; Dabholkar et al., 2000; Parasuraman & Grewal, 2000). In terms of practice, the results of the study also have implications. In the rst instance, our study indicates that rms should critically evaluate both their e-servicescape (see Koernig, 2003) and their non-visible customer service mechanisms (typically, physical distribution, and return policies). In order to ensure service consistency, online claims and promises need to be matched with physical delivery to develop trust and therefore loyalty. However, a rms e-servicescape constitutes its online presence, and most easily controllable means of inuencing the customer. Thus, rms should continuously maintain, manage, and improve their online servicescapes, to best meet the needs, wants, and demands of both current and potential customers. In this regard, creating and developing websites that generate perceived value and satisfaction, while maintaining appropriate service quality would appear to be fruitful avenues for those rms interested in enhancing factors linked with customer loyalty. In this respect, it would appear that online customers are similar to their ofine counterparts. Further, our study suggests that the conveyance of trustworthiness is of paramount importance. Indeed, such is the importance of online trust in the current studies, that it seems prudent for rms to develop strategies, systems, and sites that recognize customers concerns and that further build and maintain trust (a view also strongly advocated in Urban et al., 2000). Nevertheless, it is crucial that rms recognize that, in addition to the controllable factors to which we have already alluded, other less controllable factors are also critical. Such ndings are concordant with our emphasis on trust and, in our view, promote the relative importance of e-trust. Thus, practitioners should develop policies, procedures, and, systems that are designed to generate trust while recognizing that loyalty is not easily achievable. In this regard, it is arguable that the integration of online and traditional ofine channels may well prove a productive and complementary approach. As with all research projects, the studies presented are characterized by limitations that restrict the extent to which the results and implications of the research can be reliably generalized. In particular, three main limitations deserve highlighting. First, the LISREL methodology should be viewed with appropriate caution. While this methodology provides for the simultaneous analysis of multiple variables,
the results per se, should not be viewed as denitive evidence of causal links. In contrast, the results of the study (largely) support the a priori causal model (presented in Fig. 1). In this regard, causal effects cannot be inferred, although support for a causal model is found. Second, while the developed framework of service features a number of constructs, the developed model focuses on the incorporation of the main factors that have been theorized to be (principally) directly related to loyalty. As such, the studies are limited in the breadth of their focus. Future studies should incorporate a wider range of covariates that inuences one construct but not another (thus broadening the scope of the model and assisting in conceptually and empirically distinguishing associations between constructs). Third, care should be taken when generalizing the results of the study. Although data was collected in two contexts, enhancing generalizeability, any generalization must be undertaken with extreme caution and in a manner that, as has been observed previously, respects the complexities of differences and similarities in domain (see Chen et al., 2002), product category (see Wolnbarger & Gilly, 2001), industry (see Jones & Sasser, 1995), and national context (see Lynch et al., 2001). The implications and limitations of this study also highlight a number of potentially interesting future research projects. First, additional research is needed into the nature and dynamics of trust. While trust has long been viewed as a core concept, with notable exceptions, the centrality of the concept within service dynamics has previously been understated, overlooked, or ignored. Future studies could prot from a re-evaluation of the concept and its position within extant frameworks, models, and theories. Second, although this research has incorporated a variety of constructs into the developed framework of service presented in Fig. 1, it seems probable that other factors may also exert an inuence, particularly in different markets (e.g., computer hardware and grocery shopping online). As such, the exploration of differentiated service dynamics in alternative contexts seems a potentially fruitful avenue for research. Future studies might also prot from focusing on a wider range of variables, possibly exploring the effect of online servicescape, privacy and security issues, customer location, environmental contingencies, as well as switching inducements and deterrents. Third, this research also illuminates the need for longitudinal, causal research into the antecedents and consequences of service both on- and ofine. To date, research is largely limited to cross-sectional studies that, while analyzed with increasing complexity and depth, are restricted to supporting a priori models rather than conrming causal links. Fourth, this study has demonstrated that scales, originally developed to explore service in an ofine or conventional context, can be successfully adapted and employed in alternative contexts. This is not to suggest that such scales can be universally generalized across all contexts, but rather indicates that after careful adaptation, such scales may be
152
judiciously employed. In this regard, in the current study, the adaptation of existing trust, perceived value, customer satisfaction, and service quality scales to an online context provides a successful illustration of such an approach and also supplies a tentative starting point from which subsequent studies can explore online service dynamics. Finally, while additional research into electronic service is clearly needed, this study also suggests that additional benet could be derived from further exploring and applying the developed four-stage loyalty scale originally conceived in Oliver (1997). Indeed, we believe that the operationalized scale, with relatively minor adaptations, could protably be applied to a wide range of service contexts.
2. Conclusions The three main aims of this research project were (1) to develop and extend existing conceptualizations of service dynamics through incorporating the construct of trust, (2) to propose, operationalize, and test a specicallydeveloped four-dimension scale of loyalty that reects Olivers (1997) conception of a sequential loyalty chain, and (3) to test our theory in online contexts and thus contribute additional insights into overall service frameworks and dynamics. A review of existing literature regarding service encounters and online consumer behavior led to
the proposal of a framework of service that positions trust as the pivotal factor in driving loyalty. Analysis of the results of two studies of online purchasing supports the hypothesized relationships, presenting loyalty as a sequential, four-stage framework that is directly and indirectly related to trust, perceived value, satisfaction, and service quality. Overall, we provide evidence that supports a view and a framework of service dynamics that positions trust as the central driver of loyalty, concurrent with the direct and indirect forces of perceived value, satisfaction, and service quality. While the research design of the study precludes causal claims, the empirical results conrm associations and largely support the conceptualized model. In this regard, the evidence forwarded constitutes a preliminary attempt to understand online service dynamics. The results highlight the importance of each of these constructs, demonstrating that not only is each fundamental to understanding service but also that each must be considered simultaneously for a clearer understanding to emerge. Thus, a central conclusion must be that future evaluations of on- and ofine service should incorporate not only a broad range of constructs and variables but also an array of covariates to contextualize, illuminate, and elucidate the complexities of service dynamics. Indeed, without such holistic research, our understanding of service would be doomed to sporadic, subjectand context-specic insights that raise more questions than answers.
Appendix A. Exogenous variables used in the models tested Service quality (21 item scale using 9-point rating scales) Study One: Books.com .97a Study Two: Flights.com .95a
Please indicate how important each of the following is when buying books/ights online from Books.com/Flights.com:b 1. Providing services as promised 2. Dependability in handling customers service problems 3. Performing services right the rst time 4. Providing services at the promised time 5. Keeping customers informed about when services will be performed 6. Prompt service to customers 7. Willingness to help customers 8. Readiness to respond to customers requests 9. Web site should instill condence in its customers 10. Making customers feel safe when buying online 11. The web site is polite and courteous 12. Web site provides useful information to answer customer questions 13. Individual attention is given 14. That the web site is designed to provide a caring service 15. Having the customers best interest at heart 16. The web page should understand the needs of their customers 17. Modern design and graphics are used
153
Appendix A (Continued ) Study One: Books.com .97a 18. 19. 20. 21. Study Two: Flights.com .95a
Visually appealing pages are used The web site is well designed and has a professional appearance The site is visually appealing The web site is easily accessible
This scale was adapted from the article by Cronin and Taylor (1992). a Cronbach alpha. b The terms Books.com and books included only for the survey of book buyers and Flights.com and ights included only for the survey of ight buyers. Appendix B. Intermediate variables used in the models tested Study One: Books.com .72a Study Two: Flights.com .73a
Satisfaction (ve item scale using 7-point rating scales) Please indicate your level of agreement with the following statements:b 1. When purchasing products from Books.com/Flights.com I feel surprised, amazed or astonished 2. When purchasing products from Books.com/Flights.com I sometimes feel angry, enraged or annoyedc 3. I continue to use Books.com/Flights.com because other rms arent as good 4. My choice to purchase from Books.com/Flights.com was a wise one 5. I think I did the wrong thing when I purchased book/ights from Books.com/Flights.com This scale was adapted from the article by Cronin et al. (2000). Study One: Books.com .83a Study Two: Flights.com .83a
Perceived value (ve item scale using 7-point rating scales) Please indicate your level of agreement with the following statements:b 1. Books.com/Flights.com products are excellent value for money 2. Books.com/Flights.com services are excellent value 3. I am happy with the value for money I get at Books.com/Flights.com 4. The goods I purchase from Books.com/Flights.com are worth every cent This scale was adapted from the articles by Brady and Cronin (2001) and Mathwick et al. (2002). Study One: Books.com .77a Study Two: Flights.com .80a
Trust (eight item scale using 7-point rating scales) Please indicate your level of agreement with the following statements:b 1. Books.com/Flights.com is interested in more than just selling me goods and making a prot 2. There are no limits to how far Books.com/Flights.com will go to solve a service problem I may have 3. Books.com/Flights.com is genuinely committed to my satisfaction 4. Most of what Books.com/Flights.com says about its products is true 5. I think some of Books.com/Flights.coms claims about its service are exaggeratedc 6. If Books.com/Flights.com makes a claim or promise about its product, its probably true 7. In my experience Books.com/Flights.com is very reliable 8. I feel I know what to expect from Books.com/Flights.com This scale was adapted from the article by Hess (1995). Cronbach alpha. The terms Books.com included only for the survey of book buyers and Flights.com included only for the survey of ight buyers. c These questions are reverse coded.
b a
154
Appendix C. Endogenous variables used in the models tested Study One: Books.com Study Two: Flights.com .83a .69a Cognitive loyalty (four item scale using 7-point rating scales) Please indicate your level of agreement with the following statements:b 1. I believe that using Books.com/Flights.com is preferable to other companies 2. I believe that Books.com/Flights.com has the best offers at the moment 3. I believe that the features of Books.com/Flights.com are badly suited to what I likec 4. I prefer the service of Books.com/Flights.com to the service of competitors Study One: Books.com Study Two: Flights.com .75a .73a Affective loyalty (four item scale using 7-point rating scales) Please indicate your level of agreement with the following statements:b 1. I have a negative attitude to Books.com/Flights.comc 2. I dislike the Books.com/Flights.com offeringc 3. I like the features of Books.com/Flights.com services and offers 4. I like the performance and services of the Books.com/Flights.com Study One: Books.com Study Two: Flights.com .72a .69a Conative loyalty (four item scale using 7-point rating scales) Please indicate your level of agreement with the following statements:b 1. I have repeatedly found Books.com/Flights.com is better than others 2. I nearly always nd the offer of Books.com/Flights.com inferiorc 3. I have repeatedly found the features of Books.com/Flights.com inferiorc 4. Repeatedly, the performance of Books.com/Flights.com is superior to that of competitor rms Study One: Books.com Study Two: Flights.com .74a .78a Action loyalty (four item scale using 7-point rating scales) Please indicate your level of agreement with the following statements:b 1. I would always continue to choose Books.com/Flights.com before others 2. I will always continue to choose the features of Books.com/Flights.com before others 3. I would always continue to favor the offerings of Books.com/Flights.com before others 4. I will always choose to use Books.com/Flights.com in preference to competitor rms Study One: Books.com Study Two: Flights.com .87a .88a Overall loyalty (16 item scale using 7-point rating scales) The Overall loyalty scale is a second-order factor and includes all 16 items from the cognitive, affective, conative, action loyalty scales The above scales were adapted from the article by Oliver (1997). a Cronbach alpha. b The terms Books.com included only for the survey of book buyers and Flights.com included only for the survey of ight. c These questions are reverse coded.
155
References
Aaker, D. A. (1991). Managing brand equity: Capitalizing on the value of a brand name. New York: The Free Press. Alexander, K. L. (2003, March 11). Seeing red on security system. The Washington Post, E1. Anderson, Erin W., & Sullivan, Mary W. (1993). The antecedents and consequences of customer satisfaction for rms. Marketing Science, 12(2), 125143. Anderson, Eugene U., & Mittal, Vikas. (2000). Strengthening the satisfaction-prot chain. Journal of Service Research, 3(2), 107120. Anderson, Eugene W., Fornell, Claes, & Lehmann, Donald R. (1994, July). Customer satisfaction, market share, and protability: Findings from Sweden. Journal of Marketing, 58, 5366. Anderson, James C., & Gerbing, David W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103, 411423. Anderson, Rolph E., & Srinivasan, Srini S. (2003). E-satisfaction and Eloyalty: A contingency framework. Psychology and Marketing, 20(2), 99121. Armstrong, J. Scott, & Overton, T. Terry. (1977, August). Estimating non-response bias in mail surveys. Journal of Marketing Research, 14, 396403. Baker, Julie, Grewal, Dhruv, & Parasuraman, A. (1994). The inuence of store environment on quality inferences and store image. Journal of the Academy of Marketing Science, 22(4), 328339. Baker, Richard C. (1999). An analysis of Fraud on the internet. Internet Research: Electronic Networking Applications and Policy, 9(5), 348 359. Barnes, Brooks. (2003, January 10). The new face of air rage. The Wall Street Journal, W1. Barnes-Vieyra, Pamela & Claycomb, Cindy. (2001, MayJune). Businessto-business E-commerce: Models and managerial decisions. Business Horizons, 1320. Bass, Frank M., & Wittink, Dick R. (1975, November). Pooling issues and methods in regression analysis with examples in marketing research. Journal of Marketing Research, 12, 414425. Bauer, Hans H., Grether, Mark, & Leach, Mark. (2002). Building customer relations over the internet. Industrial Marketing Management, 31, 155163. Benassi, Paola. (1999). TRUSTe: An Online Privacy Seal Program. Communications of the ACM, 42(2), 5659. Bennett, Roger. (1996). Relationship formation and governance in consumer markets: Transactional analysis versus the behaviourist approach. Journal of Marketing Management, 12, 417436. Bentler, Peter M. (1990). Comparative t indexes in structural models. Psychological Bulletin, 107, 238246. Bern, Carmen, Mu gica, Jos M., & Yage, M. Jesu s. (2001). The effect of variety-seeking on customer retention in services. Journal of Retailing and Consumer Services, 8, 335345. Blau, Peter M. (1964). Exchange and power in social life. New York: Wiley. Bollen, Kenneth A. (1989). A new incremental t index for general structural equation models. Sociological Methods and Research, 17, 303316. Bolton, Ruth N., & Lemon, Katherine N. (1999, May). A dynamic model of customers usage of services: Usage as an antecedent and consequence of satisfaction. Journal of Marketing Research, 36, 171186. Bolton, Ruth N., Kannan, P. K., & Bramlett, Matthew D. (2000). Implications of loyalty program membership and service experiences for customer retention and value. Journal of the Academy of Marketing Science, 28(1), 95108. Boyeld, Keith. (2001, October 22). Business Europe: Europes no-frills airlines have all the fun. The Wall Street Journal, 7. Bradach, Jeffery, & Eccles, Robert G. (1989). Price, authority, trust: From ideal types to plural forms. Annual Review of Sociology, 15, 97118.
Brady, Michael K., & Cronin, J. Joseph, Jr. (2001). Customer orientation: Effects on customer service perceptions and outcome behaviors. Journal of Service Research, 3(3), 241251. Brown, George H. (1952, June 9). Brand loyaltyFact or ction? Advertising Age, 23, 5355. Brynjolfsson, Erik, & Smith, Michael D. (2000). Frictionless commerce? A comparison of internet and conventional retailers. Management Science, 46(4), 563585. Campbell, Donald R., & Fiske, Donald W. (1959). Convergent and discriminant validation by the Multitrait-Multimethod matrix. Psychological Bulletin, 56, 81105. Chaudhuri, Arjun, & Holbrook, Morris B. (2001, April). The chain of effects from brand trust and brand affect to brand performance: The role of brand loyalty. Journal of Marketing, 65, 8193. Churchill, Gilbert A., Jr. (1979, February). A paradigm for developing better measures of marketing constructs. Journal of Marketing Research, 14, 6473. Cook, Thomas D., & Campbell, Donald T. (1979). Quasi-experimentation: Design and analysis issues for eld settings. Chicago, IL: Rand McNally. Council of American Survey Research Organizations. (1982). On the denition of response rates, C.A.S.R.O. special report. Port Jefferson, NY: The Council of American Survey Research Organizations. Cronbach, Lee J. (1951). Coefcient alpha and the internal structure of tests. Psychometrica, 16, 297334. Cronin, J. Joseph, Jr., Brady, Michael K., & Hult, G. Tomas M. (2000). Assessing the effects of quality, value, and customer satisfaction on consumer behavioral intentions in service environments. Journal of Retailing, 76(2), 193218. Cronin, J. Joseph, Jr., & Taylor, Steven A. (1992, July). Measuring service quality: A reexamination and extension. Journal of Marketing, 56, 5568. Cummings, L. L., & Bromily, Philip. (1996). The Organizational Trust Inventory (OTI). In Roderick M. Kramer & Tom R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research. London: Sage Publications. Dabholkar, Pratibha A., Shepherd, C. David, & Thorpe, Dayle I. (2000). A comprehensive framework for service quality: An investigation of critical conceptual and measurement issues through a longitudinal study. Journal of Retailing, 76(2), 139173. Datamonitor. (2001). E-commerce: Missing out through poor CRM, www.datamonitor.com. Day, George S. (1969). A two-dimensional concept of brand loyalty. Journal of Advertising Research, 9(3), 2935. Dick, Alan S., & Basu, Kunal. (1994). Customer loyalty: Toward an integrated conceptual framework. Journal of the Academy of Marketing Science, 22(2), 99113. Doney, Patricia M., & Cannon, Joseph P. (1997). An examination of the nature of trust in buyer-seller relationships. Journal of Marketing, 61(2), 3551. Duncan, Otis D. (1975). Introduction to structural equation models. New York: Academic Press. Dwyer, F. Robert, Schurr, Paul H., & Oh, Sejo. (1987, April). Developing buyer-seller relationships. Journal of Marketing, 51, 1127. Eriksson, Kent, & Vaghult, Anna Lfmarck. (2000). Customer retention, purchasing behavior and relationship substance in professional services. Industrial Marketing Management, 29, 363372. Federal Trade Commission. (1998, June 4). Privacy online: A report to Congress, www.ftc.gov/reports/privacy3/index.htm. Ford, Neil M., Walker, Orville C., Jr., & Churchill, Gilbert A., Jr. (1975, April). Expectation-specic measures of the intersender conict and role ambiguity experienced by industrial salesmen. Journal of Business Research, 3, 95112. Forrester Research. (2000). www.forrester.com. Fukuyama, Francis. (1995). Trust: The social virtues and the creation of prosperity. New York: Free Press.
156
L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158 cial uses of anonymity on the web. The Information Society, 15(2), 129140. Holbrook, Morris B. (1994). The nature of customer value: An axiology of services in the consumption experience. In Ronald Russ & Richard L. Oliver (Eds.), Service Quality: New Directions in Theory and Practice (pp. 2171). Newbury Park, CA: Sage. Jacoby, Jacob. (1971). Brand loyalty: A conceptual denition. In Proceedings of the American Psychological Association (Vol. 6, pp. 655656). Washington, DC: American Psychological Association. Jacoby, Jacob, & Chestnut, Robert W. (1978). Brand loyalty: Measurement and management. New York: Wiley. James, Lawrence R., Mulaik, Stanley A., & Brett, Jeanne M. (1983). Causal analysis: Assumptions, models and data. Newbury Park, CA: Sage. Janda, Swinder, Trocchia, Philip J., & Gwinner, Kevin P. (2002). Consumer perceptions of internet retail service quality. International Journal of Service Industry Management, 13(5), 412431. Jones, Michael A., Mothersbaugh, David L., & Beatty, Sharon E. (2000). Switching barriers and repurchase intentions in services. Journal of Retailing, 76(2), 259274. Jones, Thomas O., & Sasser, E. Earl, Jr. (1995, November/December). Why satised customers defect. Harvard Business Review, 8899. Jreskog, Karl G., & Srbom, Dag. (1989). LISREL 7 Users reference guide. USA: Scientic Software, Inc. Kaynama, Shohreh A., & Black, Christine J. (2000). A proposal to assess the service quality of online travel agencies: An exploratory study. Journal of Professional Services Marketing, 21(1), 6388. Kickul, Jill, & Gundry, Lisa. (2001). Breaking through boundaries for organizational innovation: New managerial roles and practices in ecommerce rms. Journal of Management, 27, 347361. Knights, David, Noble, Faith, Vurdubakis, Theo, & Willmott, Hugh. (2001). Chasing shadows: Control, virtuality and the producer of trust. Organization Studies, 22(2), 311336. Koernig, Stephen K. (2003). E-scapes: The electronic physical environment and service tangibility. Psychology and Marketing, 20(2), 151 167. Larzelere, Robert E., & Huston, Ted L. (1980, August). The dyadic trust scale: Toward understanding interpersonal trust in close relationships. Journal of Marriage and the Family, 42, 595604. Lau, Geok Theng, & Lee, Sook Han. (1999). Consumers trust in a brand and the link to brand loyalty. Journal of Market-Focused Management, 4, 341370. Lee, Hau L., & Whang, Seungjin. (2001). Winning the last mile of Ecommerce. Sloan Management Review, 42(4), 5462. Lehtinen, Uolevi, & Lehtinen, Jarmo R. (1982). Service quality: A study of quality dimensions (Working paper). Helsinki: Service Management Institute. Li, Zhan G., & Gery, Nurit. (2000, November/December). E-tailing for all product? Business Horizons, 4954. Liljander, Veronica, van Riel, Allard C. R., & Pura, M. (2002). Customer satisfaction with e-services the case of an on-line recruitment portal. In M. Bruhn & B. Stauss (Eds.), Jahrbuch Dienstleistungsmanagement (pp. 407432). Wiesbaden: Electronic Services, Gabler. Lipstein, Benjamin. (1959). The dynamics of brand loyalty and brand switching. In Proceedings of the fth annual conference of the advertising research foundation (pp. 101108). New York: Advertising Research Foundation. Listanti, Tony. (2001). Whos who in retailA preview. DSN Retailing Today, 40(1), 14. Lynch, John, & Ariely, Dan. (2000). Wine online: Search costs and competition on price, quality and distribution. Marketing Science, 19(1), 83103. Lynch, Patrick, Kent, Robert, & Srinivasan, Srini S. (2001). The global internet shopper: Evidence from shopping tasks in twelve countries. Journal of Advertising Research, 41(3), 1523. Macneil, Ian R. (1980). The new social contract: An inquiry into modern contractual relations. New Haven, CT: Yale University Press.
Furnell, Steven M., & Karweni, Titis. (1999). Security implications of electronic commerce: A survey of consumers and businesses. Internet Research: Electronic Networking Applications and Policy, 9(5), 372 382. Gambetta, Diego. (1988). Can we trust trust? In Diego Gambetta (Ed.), Trust: Making and breaking co-operative relations. Oxford: Basil Blackwell. Gans, Noah. (2002). Customer loyalty and supplier quality competition. Management Science, 48(2), 207221. Garretson, Judith A., Fisher, Dan, & Burton, Scot. (2002). Antecedents of private label attitudes and national brand promotion attitudes: Similarities and differences. Journal of Retailing, 78, 9199. Gerbing, David W., & Anderson, James C. (1988). An updated paradigm for scale development incorporating unidimensionality and its assessment. Journal of Marketing Research, 25(2), 186192. Grabner-Kraeuter, Sonja. (2002). The role of consumers trust in onlineshopping. Journal of Business Ethics, 39, 4350. Gremler, David D. (1995). The effect of satisfaction, switching costs, and interpersonal bonds on service loyalty. Unpublished doctoral dissertation, Arizona State University. Grewal, Dhruv, Iyer, Gopalkrishnan R., & Levy, Michael. (forthcoming-a). Internet retailing: Enablers, limiters and market consequences. Journal of Business Research. Grewal, Dhruv, Lindsey-Mullikin, Joan, & Munger, Jeanne. (forthcomingb). Loyalty in e-tailing: A conceptual framework. Journal of Relationship Marketing. Grewal, Dhruv, Monroe, Kent B., & Krishnan, R. (1998, April). The effects of price comparison advertising on buyers perceptions of acquisition value and transaction value. Journal of Marketing, 62, 4659. Grnroos, Christian. (1984). A service quality model and its marketing implications. European Journal of Marketing, 18(4), 3644. Gujarati, Damodar N. (1988). Basic econometrics. New York: McGrawHill. Gundlach, Gregory T., & Murphy, Patrick E. (1993, October). Ethical and legal foundations of relational marketing exchanges. Journal of Marketing, 57, 3546. Gwinner, Kevin P., Gremler, Dwayne D., & Bitner, Mary Jo. (1998). Relational benets in service industries: The customers perspective. Journal of the Academy of Marketing Science, 26(2), 101114. Hair, Joseph F., Anderson, Rolph E., Tatham, Ronald L., & Black, William C. (1998). Multivariate data analysis (5th ed.). USA: Prentice-Hall International Inc. Hkansson, Hakan. (1982). International marketing and purchasing of industrial goods. New York: Wiley. Hamilton, Ralph M. (2001). E-commerce new venture performance: How funding impacts performance. Internet Research: Electronic Applications and Policy, 11(4), 277285. Hennig-Thurau, Thorsten, & Klee, Alexander. (1997). The impact of customer satisfaction and relationship quality on customer retention: A critical reassessment and model development. Psychology and Marketing, 14(8), 737764. Hennig-Thurau, Thorsten, Gwinner, Kevin P., & Gremier, Dwayne D. (2002). Understanding relationship marketing outcomes. Journal of Service Research, 4(3), 230247. Heskett, James L., Jones, Thomas O., Loveman, Gary W., Sasser, W. Earl, Jr., & Schlesinger, Leonard A. (1994, MarchApril). Putting the service-prot chain to work. Harvard Business Review, 72, 164174. Heskett, James L., Sasser, W. Earl, Jr., & Schlesinger, Leonard A. (1997). The service prot chain. New York: Free Press. Hess, Jeffery S. (1995). Construction and assessment of a scale to measure consumer trust. In Barbara Stern & George M. Zinkhans (Eds.), AMA winters educators proceedings (Vol. 6, pp. 2026). Chicago, IL. Hoffman, Donna L., Novak, Thomas P., & Peralta, Marcos A. (1999a). Building consumer trust online environments: The case for information privacy. Communications of the ACM, 42(4), 8085. Hoffman, Donna L., Novak, Thomas P., & Peralta, Marcos A. (1999b). Information privacy in the marketspace: Implications for the commer-
L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158 Marsh, Herbert W., & Hocevar, Dennis. (1985). Applications of conrmatory factor analysis to the study of self concept: First- and higher order factor models and their invariance across groups. Psychological Bulletin, 97, 562582. Mathwick, Charla, Malhotra, Naresh K., & Rigdon, Edward. (2002). The effect of dynamic retail experiences on experiential perceptions of value: An internet and catalog comparison. Journal of Retailing, 78(1), 5051. McCartney, Scott. (2002, August 14). American to cut capacity, jobs Worlds biggest airline aims to offset the impact of industrywide downturn. The Wall Street Journal, A4. Morgan, Robert M., & Hunt, Shelby D. (1994). The commitment-trust theory of relationship marketing. Journal of Marketing, 58, 2038. N.A.C.A.A. (2001). N.A.C.A.A./C.F.As Annual Consumer Complaint List, National Association of Consumer Agency Administration, available at: www.cosumerfed.org. Nielsen, Jakob. (1997). Loyalty on the web. Alertbox, August, 13. Nooteboom, Bart, Berger, Hans, & Noorderhaven, Niels G. (1997). Effects of trust and governance on relational risk. Academy of Management Journal, 40(2), 308338. Nunnally, Jum C. (1967). Psychometric theory. New York: McGraw-Hill. Nunnally, Jum C. (1978). Psychometric theory. New York: McGraw-Hill. Oliva, Terence A., Oliver, Richard, & MacMillan, Ian C. (1992, July). A catastrophe model for developing service satisfaction strategies. Journal of Marketing, 56, 8395. Oliver, Richard L. (1997). Satisfaction a behavioral perspective on the consumer. New York: McGraw-Hill. Oliver, Richard L. (1999). Whence customer loyalty. Journal of Marketing, 63, 3344. Parasuraman, A., & Grewal, Dhruv. (2000). The impact of technology on the quality-value-loyalty chain: A research agenda. Journal of the Academy of Marketing Science, 28(1), 168174. Parasuraman, A., Zeithaml, Valerie A., & Berry, Leonard L. (1985, Fall). A conceptual model of service quality and its implications for further research. Journal of Marketing, 49, 4150. Parasuraman, A., Zeithaml, Valerie A., & Berry, Leonard L. (1988). SERVQUAL: A multiple item scale for measuring consumer perceptions of service quality. Journal of Retailing, 64, 1240. Parasuraman, A., Zeithaml, Valerie H., & Berry, Leonard L. (1991). Renement and reassessment of the SERVQUAL scale. Journal of Retailing, 67(4), 420450. Parasuraman, A., Zeithaml, Valerie H., & Berry, Leonard L. (1994). Reassessment of expectations as a comparison standard in measuring service quality: Implications for further research. Journal of Marketing, 58(1), 111124. Pedersen, Per E., & Nysveen, Herbjrn. (2001). Shopbot banking: An exploratory study of customer loyalty effects. International Journal of Bank Management, 19(4), 146155. Pendhazur, Elazar J. (1982). Multiple regression in behavioural research. Fort Worth: Holt, Rinehart and Winston, Inc. Peterson, Robert. (1994). A meta-analysis of Cronbachs coefcient alpha. Journal of Consumer Research, 21(2), 381391. Peterson, Robert A., Balasubramanian, Sridhar, & Bronnenberg, Bart J. (1997). Exploring the implications of the internet for consumer marketing. Journal of the Academy of Marketing Science, 25(4), 329 346. Power, Stephen. (2002, February 20). Security takeover at U.S. airports begins smoothly. The Wall Street Journal, A3. Prada, Paulo. (2001, September 19). Shock waves hit European airlines. The Wall Street Journal, 23. Prada, Paulo. (2002a, January 18). Slowdown in the States: Security moves to lengthen delays. The Wall Street Journal, 36. Prada, Paulo. (2002b, June 11). Tour operators y into a low-cost zone. The Wall Street Journal, A7. Ratnasingham, Pauline. (1998). The importance of trust in electronic commerce. Internet Research: Electronic Networking Applications and Policy, 8(4), 131321.
157
Reibstein, David J. (2002). What attracts customers to online stores, and what keeps them coming back? Journal of the Academy of Marketing Science, 30(4), 465473. Reichheld, Frederick F., Markey, Robert G., & Hopton, Christopher. (2000). E-customer loyalty: Applying the traditional rules of business for online success. European Business Journal, 12(4), 173180. Reichheld, Frederick F., & Schefter, Phil. (2000, July/August). E-loyalty your secret weapon on the Web. Harvard Business Review, 105113. Reynolds, Jonathan. (2000). E-commerce: A critical review. International Journal of Retail Management, 28(10), 417444. Rothaermel, Frank T., & Sugiyama, Stephen. (2001). Virtual internet communities and commercial success: Individual and community-level theory grounded in the atypical case of TimeZone.com. Journal of Management, 27, 297312. Rushton, J. Philippe. (1980). Altruism, socialisation, and society. New York: Prentice-Hall. Rust, Roland T., & Oliver, Richard L. (2000, Winter). Should we delight the customer? Journal of the Academy of Marketing Science, 28, 6 94. Scheraga, Dan. (2000, September). Penneys net advantage. Chain Store Age, 114118. Schoder, Detlef, & Yin, Pai-Ling. (2000, December). Building rm trust online. Communications of the ACM, 43(12), 7379. Shapiro, Carl. (2000, MayJune). Will E-commerce erode liberty? Harvard Business Review, 189196. Shim, Soyeon, Eastlick, Mary Ann, Lotz, Sherry L., & Warrington, Patricia. (2001). An online intentions model: The role of intention to search. Journal of Retailing, 77, 397416. Shop.org. (2001). Shop.org press room. Washington, DC: National Retail Federation. [http://www.shop.org]. Shop.org & Boston Consultancy Group. (2000). State of online retailing 3.0. Washington, DC: National Retail Federation [http://www.shop.org]. Singh, Jagdip, & Sirdeshmukh, Deepak. (2000). Agency and trust mechanisms in consumer satisfaction and loyalty judgments. Journal of the Academy of Marketing Science, 28(1), 150168. Sirdeshmukh, Deepak, Singh, Jagdip, & Sabol, Barry. (2002). Consumer trust, value, and loyalty in relational exchanges. Journal of Marketing, 66, 1537. Sirohi, Niren, McLaughlin, Edward W., & Wittink, Dick R. (1998). A model of consumer perceptions an store loyalty intentions for a supermarket retailer. Journal of Retailing, 74(2), 223245. Slater, Stanley F. (1995). Issues in conducting marketing strategy research. Journal of Strategic Marketing, 3(4), 257270. Srinivasan, Srini S., Anderson, Rolph, & Ponnavolu, Kishore. (2002). Customer loyalty in e-commerce: An exploration of its antecedents and consequences. Journal of Retailing, 78, 4150. Standird, Stephen S. (2001). Reputation and E-commerce: eBay auctions and the asymmetrical impact of positive and negative ratings. Journal of Management, 27, 279295. Stewart, Katherine J. (2003). Trust transfer on the World Wide Web. Organization Science, 14(1), 517. Sultan, Fareena, & Mooraj, Hussain, A. (2001, November/December). Designing a trust-based E-business strategy. Marketing Management, 4045. Sweeney, Jillian C., & Soutar, Geoffrey N. (2001). Consumer perceived value: The development of a multiple item scale. Journal of Retailing, 77, 203220. Sweeney, Jilian C., Soutar, Geofrrey N., & Johnson, Lester W. (1999). The role of perceived risk in the quality-value relationship: A study in a retail environment. Journal of Retailing, 75(1), 77105. Szymanski, David M., & Henard, David H. (2001). Customer satisfaction: A meta-analysis of the empirical evidence. Journal of the Academy of Marketing Science, 29(1), 1635. Szymanski, David M., & Hise, Richard T. (2000). e-Satisfaction: An initial examination. Journal of Retailing, 76(3), 309322.
158
L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158 Woodruff, Robert B. (1997, Spring). Customer value: The next source of competitive advantage. Journal of the Academy of Marketing Science, 25, 139153. Zeithaml, Valerie A. (1988, July). Consumer perceptions of price, quality, and value: A means-end model and synthesis of evidence. Journal of Marketing, 52, 222. Zeithaml, Valerie A., Berry, Leonard L., & Parasuraman, A. (1996, April). The behavioral consequences of service quality. Journal of Marketing, 60, 3146. Zeithaml, Valerie A., Parasuraman, A., & Malhotra, Arvind. (2000a). A conceptual framework for understanding E-service quality: Implications for future research and managerial practice (Report No. 00-115). Cambridge, MA: Marketing Science Institute. Zeithaml, Valerie A., Parasuraman, A., & Malhotra, Arvind. (2000b). Eservice quality: Denitions, dimensions and conceptual model (Working Paper). Cambridge, MA: Marketing Science Institute. Zeithaml, Valerie A., Parasuraman, A., & Malhotra, Arvind. (2002). Service quality delivery through Web sites: A critical review of extant knowledge. Journal of the Academy of Marketing Science, 30(4), 362 375.
Tax, Stephen S., Brown, Stephen W., & Chandrashekaran, Murali. (1998, April). Customer evaluations of service complaint experiences: Implications for relationship marketing. Journal of Marketing, 62, 6076. Urban, Glen L., Sultan, Fareena, & Qualls, Fareena. (2000). Placing trust at the center of your internet strategy. Sloan Management Review, Fall, 3948. van Riel, Allard C. R., Liljander, Veronica, & Jurriens, Peter. (2001). Exploring consumer evaluations of e-service: A portal site. International Journal of Service Industry Management, 12(4), 359377. Verhoef, Peter C., Francis, Philip Hans, & Hoekstra, Janny C. (2002). The effect of relational constructs on customer referrals and number of services purchased from multiservice provider: Does age of relationship matter? Academy of Marketing Science, 30, 3. Walsh, John, & Godfrey, Sue. (2000). The internet: A new era in customer service. European Management Journal, 18(1), 8592. Wingeld, Nick. (2001, January 25). Cash supply shrinks while Webvan losses continue. The Wall Street Journal, B1, B4. Wolnbarger, Mary, & Gilly, Mary C. (2001). Shopping online for freedom, control and fun. California Management Review, 43(2), 34 55.

Four Levels of Loyalty

Uploaded by

Copyright:

Available Formats

Four Levels of Loyalty

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Four Levels of Loyalty

Uploaded by

Copyright:

Available Formats

Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

1.00 .69 .21

1.00 .70 .46

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

1.00 .34 .51 .44

1.00 .59 .61

1.00 .24 .17 .23

1.00 .03 .78

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

Fig. 2. The empirical result for Flights.com and Books.com studies.

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

L.C. Harris, M.M.H. Goode / Journal of Retailing 80 (2004) 139158

You might also like