Service quality and satisfaction in business-to-business services

Richard A. Spreng
Eli Broad College of Business, Michigan State University, East Lansing, Michigan, USA

Linda Hui Shi

Faculty of Business, University of Victoria, Victoria, Canada, and

Thomas J. Page
Eli Broad College of Business, Michigan State University, East Lansing, Michigan, USA
Abstract Purpose The purpose of the paper is to investigate the effects of service quality and service satisfaction on intention in a business-to-business setting. Design/methodology/approach This research addresses three unanswered questions regarding satisfaction and service quality: the distinction between customer satisfaction and perceived service quality; their causal ordering; and their relative impact on intentions. The data were collected using a large survey of buyers in a business setting. Findings The data were analyzed using structural equation modeling. The results show that service quality has a larger impact on intentions than does customer satisfaction. The results also show that the effects of individual transactions on intentions are mediated by corresponding cumulative constructs. Research limitations/implications The primary implications for theory include demonstrating the distinction between satisfaction and service quality; specifying, based on theory and logic, the causal ordering between transaction constructs and cumulative constructs, and between service quality and satisfaction; and assessing their relative impact on behavioral intentions. Originality/value The results show that one negative transaction outcome may not be sufcient to cause the customer to switch if the cumulative levels are sufciently positive. Thus, a negative outcome may be discounted by the user if it is seen as a unique occurrence. However, a series of successive negative transaction outcomes may cause the cumulative constructs to become less positive, resulting in lower intentions to repurchase from the same supplier. Keywords Customer satisfaction, Customer services quality, Business-to-business marketing Paper type Research paper

An executive summary for managers and executive readers can be found at the end of this article.

Introduction
The fundamental proposition in customer satisfaction and service quality research is that high customer satisfaction (CS) and perceived service quality (PSQ) lead to high customer repurchase intention (e.g. Heskett et al., 1994; Rust et al., 1995; Oliver, 1997). Although PSQ and CS to retention links have been studied extensively, the vast majority of PSQ and CS research has been done in the business-to-consumer (B2C) context (Paulssen and Birk, 2007). It is crucial to determine whether the CS and PSQ to intention links are also relevant in business-to-business settings because these linkages may directly inuence economic return (Anderson et al., 1994; Luo and Bhattacharya, 2006). Previous research
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0885-8624.htm

has shown that the ndings in a consumer setting may be able to be translated into an organizational setting (Cooper and Jackson, 1988; Durvasula et al., 1999). In this study, we investigate the relative impacts of CS and PSQ in a businessto-business setting. While we acknowledge that marketing researchers have paid increasing attention to PSQ and CS in the business-to-business setting (Chandrashekaran et al., 2007; Lam et al., 2004; Liu et al., 2003; Gounaris, 2005; Rauyruen and Miller, 2007; Jayawardhena et al., 2007; Patterson and Spreng, 1997; Patterson et al., 1997), the exact nature of the relationship between PSQ, CS, and behavioral intentions in a business-to-business context leaves several questions unanswered. The extensive research on PSQ and CS in a consumer context provides some understanding of these constructs. Research has examined: . the conceptual distinction between CS and PSQ (for a review, see Oliver, 1997); . the causal ordering between them (e.g. Teas, 1993); and . the relative impact of the two on behavioral intentions (e.g. Cronin and Taylor, 1992).
Received: April 2008 Revised: August 2008 Accepted: October 2008

Journal of Business & Industrial Marketing 24/8 (2009) 537 548 q Emerald Group Publishing Limited [ISSN 0885-8624] [DOI 10.1108/08858620910999411]

537

Service quality and satisfaction in business-to-business services Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Journal of Business & Industrial Marketing Volume 24 Number 8 2009 537 548

Furthermore, both PSQ and CS have been conceptualized as episode (transaction) specic and as cumulative (global) evaluations that result from a number of transactions. In spite of these studies, much confusion remains. In addition, this research has not generally been applied to business-tobusiness settings. The purposes of this paper are: . to review the conceptual distinctions and causal ordering between PSQ and CS both at the transaction specic and cumulative levels; . to develop an integrative model that includes behavioral intentions as well as PSQ and CS; and . to test the proposed relationships in a business-to-business context. The paper is organized as follows. First, we review relevant literature in both the service quality and satisfaction areas. Second, the proposed model linking PSQ, CS, and intentions is developed. Next, the method used to test the model is described. Finally, the results of the test of the proposed model are presented, followed by a discussion of the results. Our research possesses both theoretical and managerial implications.

Review of PSQ and CS literature

The rst task is to distinguish between PSQ and CS. To do this, we adopt the accepted proposition that PSQ is primarily a cognitive concept, while CS is more affective in nature (Parasuraman et al., 1988; Bolton and Drew, 1991a, b; Boulding et al., 1993; Yi, 1990). Perceived service quality Parasuraman et al. (1988) dened PSQ as a global judgment or attitude relating to the superiority of the service (p. 16). Similarly, Oliver (1997, p. 27) dened PSQ as a judgment of performance excellence. More specically, Oliver (1994) suggested that quality is largely a performance assessment and it relies on evaluation of performance excellence criteria. Also, de Ruyter and Wetzels (1998) implied that PSQ is a mathematical evaluation or judgment. Thus, it appears that PSQ is conceptualized as either a judgment or an appraisal variable (Cronin and Taylor, 1992; Gotlieb et al., 1994). Customer satisfaction In contrast, CS is usually considered to be a more affective concept. Many customer satisfaction researchers have dened CS as an affective response (e.g. Cadotte et al., 1987; Oliver, 1989; Spreng et al., 1996; Westbrook and Reilly, 1983). For example, Westbrook (1987) argued that both positive and negative affect are directly related to the favorability of customer satisfaction. This belief is further afrmed by the affective-processing mechanisms proposed by Cohen and Areni (1991), in which they propose that emotions will leave strong affective traces in consumers memory which can be retrieved when a consumer evaluates the relevant consumption experience. Finally, Oliver (1989 p. 1) states that CS is an affective, or emotional response. Conceptual distinctions There are empirical ndings supporting the view that PSQ is primarily a cognitive concept while CS is more affective in nature. For example, Oliver (1994) found that PSQ was inuenced by performance (cognitive judgments) but not by affective constructs (negative or positive affect), whereas CS was 538

affected by both cognitive (disconrmation) and affective antecedents. Other distinctions between PSQ and CS have also been proposed. While these distinctions are still being debated in the literature, they are worth highlighting here. First, some researchers have suggested a difference in temporal focus, with PSQ being primarily a long-term, cumulative concept, while CS is more episode-specic or transactional (Bitner, 1990; Boulding et al., 1993; Cronin and Taylor, 1992). In other words, CS refers to an evaluation of a specic service transaction (transaction CS), while PSQ refers to a judgment made after multiple transactions over time, and is cumulative in nature (cumulative PSQ). However, criticisms of this argument have emerged from both the service quality literature and the customer satisfaction literature. In the service quality literature, Teas (1993, p. 30) challenges this temporal focus view by stating that PSQ can be operationalized as either transaction specic quality or relationship quality. Consistent with Teass (1993) contention, Parasuraman et al. (1994) also argue that service quality can be operationalized as either a transaction or as a cumulative construct. Similarly, in the customer satisfaction literature, researchers maintain that CS can be treated as either a transaction specic concept or a cumulative evaluation after multiple transactions (Anderson et al., 1994; Fornell, 1992). Empirically, CS has been operationalized as transaction CS (Cadotte et al., 1987; Oliver, 1980; Spreng et al., 1996; Tse and Wilton, 1988; Westbrook, 1987) and as cumulative CS (e.g. Anderson et al., 1994; Fornell, 1992). Thus, it is difcult to argue that temporal focus is a means of distinguishing between CS and PSQ. A second distinction that has been suggested is that the standard of comparison is different in each case (Bitner, 1990; Oliver, 1993, 1997; Parasuraman et al., 1988; Zeithaml et al., 1993). This perspective argues that the standard for CS is predictive expectations, while the standard for PSQ is the consumers belief about what a service provider should provide (Parasuraman et al., 1988). However, Spreng and Mackoy (1996) challenged the comparison standard view based on their empirical nding that desires congruency inuences both CS and PSQ. In addition, a great deal of research in customer satisfaction has challenged the view that CS is formed solely through the disconrmation of predictive expectations, and has found support for competitive-based norms (Cadotte et al., 1987) and desires (Spreng and Olshavsky, 1993) as a basis for CS. Thus, as with temporal focus, it is difcult to argue that the standard of comparison can reliably distinguish between PSQ and CS. In summary, we support the argument that PSQ is a judgment of performance, while CS is an affective construct. Specically, transaction PSQ is dened as a judgment of performance excellence based on the customers last experience (Oliver, 1997). Cumulative PSQ is a judgment of performance excellence based on all of the customers experience with a particular product or service (Oliver, 1997). Transaction CS is an affective state that is the emotional reaction to the customers last product or service experience (Oliver, 1980). Cumulative CS is an affective state that is the emotional reaction to all of the customers experience with a particular product or service (Oliver, 1980). As mentioned earlier, there has been little research examining PSQ and CS in a business-to-business setting. There are reasons to believe, however, that the relationships

Service quality and satisfaction in business-to-business services Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Journal of Business & Industrial Marketing Volume 24 Number 8 2009 537 548

among the constructs may be different in a business-tobusiness versus consumer setting, and even between different functional areas within a given rm (Chakraborty et al., 2007). Fern and Brown (1984) advocate the industrial/ consumer dichotomy in that there are signicant differences between organizational versus consumer marketing. Others, however, argue that there are more similarities than differences between organizational versus consumer marketing (Andrus and Norvell, 1990; Coviello and Brodie, 2001; Dawes and Patterson, 1988). Miciak and Desmarais (2001) found support for both propositions, but generally supported the industrial/consumer dichotomy. This difference might manifest itself, for example, in the impact of PSQ. Since organizational buying often involves more face-to-face contact between buyers and sellers, this increased interaction may increase the overall importance of PSQ. However, given the limited guidance from B2B research, the model development below will be based primarily on research in B2C settings, except where relevant B2B research exists.

while Gotlieb et al. (1994) operationalized both as transaction-specic. In each case only the PSQ ! CS link was signicant. Model to be tested and hypotheses Figure 1 shows the model to be tested. After briey describing the overall model, each of the linkages will be discussed. There are two overriding rationales for the proposed model. First, transaction constructs come before the cumulative constructs, because cumulative constructs are based on a series of specic transactions (Holmlund, 2004; Oliver, 1997). Oliver (1997) suggested a model in which transaction constructs lead to cumulative constructs. However, this logic does not require either PSQ or CS to come rst. That is, either PSQ or CS can come rst, but the transaction construct is an antecedent of the cumulative construct. Thus, transaction PSQ has to be an antecedent of cumulative PSQ and transaction CS has to be an antecedent of cumulative CS. Second, based on the arguments above, we specify PSQ as an antecedent of CS. As stated above, research has generally supported a PSQ ! CS order over a CS ! PSQ order. The causal ordering, however, must be based on logical or conceptual grounds rather than empirical ndings. Given our earlier argument that PSQ is a cognitive evaluation while CS is an affective reaction, we adopt the view that PSQ is an antecedent of CS. The basic rationale for this is based on cognitive appraisal theories of emotions. A number of emotion theorists have developed and tested cognitiveappraisal theories of emotions (for reviews, see for example Omdahl, 1995; Nyer, 1997; Bagozzi et al., 1999). These theories claim that emotions are the result of cognitive processing that follows exposure to a stimulus. That is, the cognitive appraisal of the situation evokes an individuals emotional response (Lazarus, 1991). Bagozzi (1992) utilized this perspective and suggested a causal chain of a cognitive appraisal which leads to an emotional response which then leads to behavioral intentions. The Bagozzi (1992) framework can be applied to PSQ/CS context because, as discussed above, PSQ is an appraisal construct (Parasuraman et al., 1988; Bolton and Drew 1991a, b), while CS is an affective construct (e.g. Cadotte et al., 1987; Oliver, 1989; Spreng et al., 1996; Westbrook and Reilly, 1983). Therefore, the sequence proposed by Bagozzi (1992) i.e. a cognitive appraisal ! emotional response ! coping behavior, can be translated into PSQ ! CS ! intentions. The rationale for the remaining linkages is presented below. Antecedents of behavioral intentions Both cumulative PSQ (Path A) and cumulative CS (Path B) are proposed antecedents of behavioral intentions. The more satised the customer is with a series of transactions, and the higher the perceived level of service quality over a series of transactions, the higher the level of intentions to repurchase from the same source. Taylor and Baker (1994) found in four customer service industries that CS had a stronger effect on intentions in each industry, and for two of the industries the effect of PSQ was non-signicant. Cronin and Taylor (1992) operationalized both constructs as cumulative evaluations and in four customer service industries found that cumulative CS had a stronger effect on behavioral intentions than did PSQ. Similarly, Cronin et al. (2000) tested several competing models for CS/PSQ, and found the model t was best when direct paths from both PSQ and CS to intentions were included, 539

Model development
Causal ordering of PSQ and CS There is considerable debate concerning the causal ordering of PSQ and CS. Table I summarizes research since 1992 that has looked at both constructs. Among those studies that have examined the ordering, some researchers contend that CS ! PSQ based on the assumption that CS is an effect of disconrmation of a single transaction and quality perception (PSQ) is an effect of multiple satisfactory/dissatisfactory service transactions (CS) (Bitner, 1990; Bolton and Drew 1991a, b). Note that this argument relies on the temporal order perspective in distinguishing between PSQ and CS. That is, based on the contention that CS is a transaction evaluation and PSQ is a cumulative evaluation, this ordering suggests that the transaction construct (claimed to be CS) leads to the cumulative construct (claimed to be PSQ). As stated above, however, both PSQ and CS can be either a transaction or a cumulative evaluation. Others, however, contend that PSQ ! CS. Oliver (1997, p. 184) stated that quality is one of the key dimensions that are factored into the consumers satisfaction judgment, and based this in part on the denition of quality from the International Organization for Standardization that states: Quality is the totality of features and characteristics of a product or service that bear on its ability to satisfy stated and implied needs. In a conceptual model that integrates PSQ/CS research streams, Oliver (1993) suggests that CS is a function of the disconrmation of predictive expectations regarding both quality and non-quality dimensions, and therefore species PSQ as an antecedent of CS. Spreng and Mackoy (1996) tested this model and found that PSQ was an antecedent of overall satisfaction. As shown in Table I, there are a number of empirical ndings supporting this ordering. For example, Brady and Robertson (2001) empirically tested two models (PSQ ! CS and CS ! PSQ) and found better t and higher explained variance for a model specifying PSQ ! CS. Finally, several studies proposed a reciprocal relationship between CS and PSQ such that PSQ ! CS and CS ! PSQ, but all found only the PSQ ! CS path to be signicant (e.g. Cronin and Taylor, 1992; de Ruyter et al., 1997; Gotlieb et al., 1994). Note that Cronin and Taylor (1992) and de Ruyter et al. (1997) operationalized both constructs as cumulative,

Table I Review of studies measuring both PSQ and CS

Study and service tested 0.82 0.76 0.94 0.79 0.51 Not tested

CS and PSQ correlation Causal order Comments

Global versus Temporal attribute operationalization measures

Cronin and Taylor (1992) (banking, pest control, dry cleaning, fast food) Spreng and Singh (1993) (banking)

Bitner and Hubbert (1994) (airline travel)

Bitner and Hubbert (1994) (airline travel)

Oliver (1994) (hospital stay)

CS ! PSQ and Only the PSQ ! CS link was signicant PSQ ! CS Not tested CS scale item anchored by very dissatised/very satised loaded on both CS and PSQ constructs Not tested When CS and PSQ are both operationalized as cumulative, there is a lack of distinction between them Not tested When PSQ is cumulative and CS is situational, there is a distinction between them PSQ ! CS CS was a function of positive and negative affect and of PSQ

Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Taylor and Baker (1994) (health care, recreation, airline, 0.64-0.80 telephone) 0.85 Gotlieb et al. (1994) (hospital stay) 0.95 0.92 0.94 0.73 0.73 Not reported

Service quality and satisfaction in business-to-business services

Spreng and Agrawal (1995) (banking)

Spreng and Agrawal (1995) (banking)

Spreng and Agrawal (1995) (student advising)

540
PSQ: cumulative CS: cumulative 0.67 0.53-0.83 0.24-0.46 0.68 0.51 0.66 PSQ: transaction CS: cumulative PSQ: cumulative CS: cumulative PSQ: cumulative CS: cumulative PSQ: cumulative CS: cumulative PSQ: transaction CS: transaction PSQ: cumulative CS: transaction

Spreng and Mackoy (1996) (student advising)

de Ruyter et al. (1997) (chiropractic treatment)

Brady and Robertson (2001) (fast food)

PSQ: cumulative CS: cumulative PSQ: cumulative CS: cumulative PSQ: cumulative CS: cumulative PSQ: cumulative CS: transaction PSQ: transaction CS: transaction PSQ: cumulative CS: cumulative PSQ: transaction CS: transaction PSQ: cumulative CS: transaction PSQ: cumulative CS: cumulative PSQ: cumulative CS: cumulative PSQ: cumulative CS: transaction PSQ: cumulative CS: cumulative PSQ: transaction CS: transaction PSQ: transaction CS: transaction PSQ ! CS

0.39-0.56 Cronin et al. (2000) (spectator sports, participation sports, entertainment, health care, long distance carriers, fast food) Brady et al. (2002) (fast food, banking, pest control, and 0.55-0.62 dry cleaning)

In three out of four settings, both CS and PSQ were signicantly related to intentions CS ! PSQ and Only PSQ ! CS was signicant. In a model with CS ! PSQ, the PSQ ! CS t was poor, but there was a signicant effect (0.88, t 18:45) Not tested In contrast to Bitner and Hubbert (1994), when PSQ is cumulative and CS is situational, there is a lack of distinction Not tested When both constructs are measured as cumulative, there is a lack of distinction between them Not tested When both constructs are measured as cumulative, there is a lack of distinction between them PSQ ! CS This study tested Olivers (1993) model and found support for much of it CS ! PSQ and Only the PSQ ! CS path was signicant PSQ ! CS CS ! PSQ and PSQ ! CS model was superior to. CS ! PSQ model in two PSQ ! CS samples that were from different cultures PSQ ! CS Best tting model was one which specied PSQ ! CS and a direct PSQ ! intentions link

Caruana (2002) (banking)

Journal of Business & Industrial Marketing

Volume 24 Number 8 2009 537 548

Sureshchandar et al. (2002) (banking)

Chumpitaz and Paparoidamis (2004) (business-tobusiness IT service industry) Olorunniwo et al. (2006) (restaurant customers and student subjects) Gonzalez et al. (2006) (spa tourism)

Jayawardhena et al. (2007) (business-to-business nancial service industry)

PSQ: Global CS: global PSQ: global CS: global PSQ: global CS: global PSQ: global CS: global PSQ: global CS: global PSQ: global CS: global PSQ: attribute CS: global PSQ: global CS: global PSQ: global CS: global PSQ: global CS: global PSQ: global CS: global PSQ: attribute CS: global PSQ: global CS: global PSQ: global and attribute CS: global PSQ: global and attribute CS: global PSQ: global CS: global PSQ: attribute CS: attribute PSQ: attribute CS: global PSQ: attribute CS: global PSQ: attribute CS: global PSQ: attribute CS: attribute

The results support that the performance-only approach to the measurement of PSQ is superior to the gap-based scale, and that PSQ is an antecedent of CS PSQ ! CS A mediational model that links PSQ to service loyalty via CS is supported Not tested The results have indicated that the two constructs are indeed independent but are closely related PSQ ! CS The ndings supports that CS is a mediator between PSQ and loyalty PSQ ! CS Both SQ and CS signicantly inuences behavioral intentions, and CS is a mediator on PSQ ! intentions link CS ! PSQ and PSQ ! CS is conrmed. However, the converse relationship PSQ ! CS does not hold CS ! PSQ Transaction CS is signicant on cumulative PSQ

Service quality and satisfaction in business-to-business services Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Journal of Business & Industrial Marketing Volume 24 Number 8 2009 537 548

Figure 1 Proposed Model and Hypothesis

and also found that the effect of CS on intention was stronger. In addition, some theoretical studies suggest that affect will serve as a primary motivator of future behaviors (e.g. Woodruff, 1993; Abelson et al., 1982). For example, Allen et al. (1992) found that emotions serve as a better predictor of intentions than did cognitive evaluations. However, in a recent longitudinal study, Homburg et al. (2006) empirically supported the notion that as experience accumulates, the impact of affective constructs on customer evaluation decreases and the impact of cognitive constructs increases. The studies discussed above were all carried out in a setting in which the individual consumer was the beneciary of the service rather than in a business-to-business setting. The studies that have looked at this relationship in a business-tobusiness setting have found mixed results. Patterson and Spreng (1997) and Patterson et al. (1997) found that CS has a strong positive effect on intention, but did not include PSQ in their model. In a business-to-business setting, it seems logical to expect that the more rational cognitive aspects of the decision may outweigh the affective aspects, particularly in cases where the purchasing agent is not the end user, as is often the case in a business-to business setting (Chakraborty et al., 2007). This would result in cumulative PSQ having a stronger effect on intentions than cumulative CS. This has been conrmed in a business-to-business study done by Wathne et al. (2001). They tested empirically how cognitive and affective constructs inuence choice in the context of business-to-business services. They found that the cognitive construct of service is more important than the affective construct of interpersonal relationship satisfaction in inuencing choice. Since the relative effects of CCS and CPSQ are not totally clear, we will base our hypothesis on the majority of the literature and propose that CCS will have a stronger effect on behavioral intentions than CPSQ. This leads to our rst three hypotheses: H1a. Cumulative PSQ will have a positive effect on behavioral intentions. Path A. H1b. Cumulative CS will have a positive effect on behavioral intentions. Path B. 541

H1c. Cumulative CS will have a stronger relationship with behavioral intentions than will cumulative PSQ. Antecedents of cumulative CS and cumulative PSQ As noted above, Oliver (1997) proposed a model in which transaction PSQ has an effect on cumulative PSQ, and transaction CS has an effect on cumulative CS. Several other customer satisfaction researchers have suggested a similar set of relationships (e.g. Teas, 1993; Oliver, 1997; Parasuraman et al., 1994). While it may seem tautological to propose that transaction based constructs are antecedents of cumulative based constructs, it is not necessarily clear how this process occurs. Holmlund (2004) suggested that transaction-based constructs (i.e. episodes) are antecedents of cumulative-based construct (i.e. sequence), but it is not necessarily a simple additive process and the cumulative effects will depend heavily on how the individual transactions are aggregated. Holmlund (2004, p. 282) states that [I]nterrelated episodes can in turn be correspondingly grouped into a sequence, which forms a still larger and more extensive entity on a higher interaction level. Thus, some episodes may be weighted much more heavily than others while others may be completely ignored in arriving at cumulative responses. Based on the above empirical ndings and our argument that transaction constructs come before cumulative constructs, and PSQ leads to CS, we propose that: . transaction PSQ is an antecedent of cumulative PSQ (Path D); . transaction CS is an antecedent of cumulative CS (Path E); . cumulative PSQ is an antecedent of cumulative CS (Path C); and . transaction PSQ is an antecedent of transaction CS (Path F). H2a. Transaction PSQ will be positively related to cumulative PSQ. Path D. H2b. Transaction CS will be positively related to cumulative CS. Path E.

Service quality and satisfaction in business-to-business services Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Journal of Business & Industrial Marketing Volume 24 Number 8 2009 537 548

H2c. Cumulative PSQ will be positively related to cumulative CS. Path C. H2d. Transaction PSQ will be positively related to transaction CS. Path F.

Distribution Center is . . .?, using a ve-point scale anchored by poor and excellent (Cronin and Taylor, 1992). To measure transaction CS, respondents were asked How did you feel about the service you received during your last experience with the distribution center? and used two commonly employed scales (Spreng et al., 1996). The rst scale was anchored by dissatised/satised and the second scale was anchored by terrible/delighted. Cumulative CS was assessed by the same anchors (i.e. dissatised/satised and terrible/delighted), but asked the respondents to think about all of their experiences with the distribution center (Bitner, 1990). Behavioral intentions were operationalized with one item: How likely is it that you will use the Distribution Center in the future, with a ve-point scale ranging from denitely not to denitely will. While it is desirable to have multiple measures for most constructs, as Bergkvist and Rossiter (2007) point out, for concrete attributes such as intentions, this is not necessary, and in order to avoid respondent fatigue, every attempt was made to keep the number of measures as low as possible.

Method
Study design Data were collected from internal customers of the distribution center of a large state agency using a written survey. The distribution center is responsible for purchasing, inventory management, and distribution of a wide variety of equipment and supplies for the agency. A major responsibility of the center is that of purchasing, processing nearly 2,000 purchase orders a year and maintaining numerous contracts for the state agency. The data consisted of evaluations by employees of 26 different departments of the state agency concerning their experiences with the distribution center. It should be noted that for some items, purchases were constrained to be made from the distribution center but for other items there were no such constraints. In fact, a primary reason for conducting the study was to enable the distribution center to retain its internal customers in the face of increasing pressure on its customers to reduce purchasing costs by purchasing from other sources. As suggested by Day and Barksdale (2003), organizations and businesses are increasingly seeking external service providers to increase service quality and reduce organizational cost. The list of potential respondents was obtained by job title and classication. A total of 517 surveys were sent out through inter-organizational mail, and 296 usable responses were received, resulting in a 57 percent response rate. Part of the explanation for this high response rate is that the culture of this branch of state government requires obedience to hierarchical authority, and the survey instruction letter came from the head of the organization. Sample characteristics The sample was 77 percent male, and respondents ranged from 22 to 60 years old, with a mean age of 40. About two thirds (64 percent) of the respondents used the distribution center once per year, while 31 percent used it monthly and 5 percent used it weekly. Twenty-four percent of the respondents were high school graduates, 30 percent were technical or community college graduates, 37 percent had a Bachelors degree, and 9 percent had a graduate degree. Measures Transaction PSQ (TPSQ) was measured with two items: 1 The level of service quality I received during my last experience was excellent, using a ve-point scale anchored strongly disagree and strongly agree; and 2 The level of service quality I received from the Distribution Center during my last experience was . . . , using a ve-point scale anchored inferior to superior (Brady and Robertson, 2001). Cumulative PSQ was measured with two items: 1 The distribution center provides excellent overall service quality, using a ve-point scale anchored by strongly disagree and strongly agree; and 2 Thinking about all of your experiences, would you say that the quality of service you receive from the 542

Results
Measurement model We used structural equation modeling to test the hypothesized relationships presented in our model (Bentler, 1995). Table II reports the results of a conrmatory factor analysis of the measures using the covariance matrix. The t was good ( x2 39:96; df 18; NFI 0:99, NNFI 0:99, CFI 0:99, IFI 0:99, GFI 0:96, RMR 0:01, RMSEA 9:06). The measurement parameters and all the coefcients linking the indicators with their latent constructs were signicant (t-values ranged from 14.40 to 17.54). Table II also shows that the construct reliabilities were very good (0.89 to 0.98) and the average variance extracted (AVE) for each construct was far above the 50 percent cut-off suggested by Fornell and Larcker (1981). To demonstrate discriminant validity the factor correlations were constrained (one at a time) to be equal to 1.0. In each case this produced a signicant increase in x2, indicating that the constructs are distinct. As suggested by Fornell and Larcker (1981), the AVE, which shows the amount of the variance that is captured by the construct in relation to the amount of variance due to measurement error, was also used as a test of discriminant validity. The AVE can be compared to the shared variance between any two constructs (the squared correlation between the constructs) in that the AVE should be higher for each construct than the squared correlation between that construct and any other construct. This test holds for all constructs, and thus there is evidence of discriminant validity among the constructs. Model and hypothesis tests We then tested the full model, which consisted of the structural model and the measures of each construct (Figure 1). The t for the proposed model is acceptable ( x2 64:98; df 22; NFI 0:98, NNFI 0:98, CFI 0:99, GFI 0:93, RMR 0:03, RMSEA 0:08). All six paths were signicant at p 0:05. H1a stated that cumulative PSQ will have a positive effect on intentions, and this was supported. H1b predicted that cumulative CS would have a positive effect on intentions, and this was also

Service quality and satisfaction in business-to-business services Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Journal of Business & Industrial Marketing Volume 24 Number 8 2009 537 548

Table II CFA results

Construct and measures Transaction PSQ The level of service quality I received during my last experience was excellent (anchored strongly disagree 5 1 and strongly agree 5 5) The level of service quality I received from the Distribution Center during my last experience was . . . (anchored inferior 5 1 and superior 5 5) Transaction CS How did you feel about the service you received during your last experience? (anchored dissatisfied 5 1 and satisfied 5 7) (anchored terrible 5 1 and delighted 5 7) Cumulative PSQ The distribution center provides excellent overall service quality (anchored strongly disagree 5 1 and strongly agree 5 5) Thinking about all of your experience, would you say that the quality of service you receive from the Distribution Center is . . .? (anchored ooor and excellent) Cumulative CS Based on all your experiences, how do you feel about the service you received from the distribution center? (anchored dissatisfied 5 1 and satisfied 5 7) (anchored terrible 5 1 and delighted 5 7) Intentions How likely is it that you will use the Distribution Center in the future? (anchored denitely not 5 1 and denitely will 5 5
a b c

CSPa

ab
0.91

AVEc 0.83

0.90 0.92 0.96 0.98 0.98 0.89 0.88 0.91 0.98 0.98 0.98 0.96 0.80 0.93

1.0

Notes: Completely standardized parameter. Cronbachs a. Average variance extracted (AVE), which is the proportion of variance in the construct that is not P P P due to measurement error (Fornell and Larcker, 1981). AVE is computed as: l2 = l2 var1

supported. H1c stated that cumulative CS would have a stronger relationship with behavioral intentions than would cumulative PSQ. This was not supported. A x2 difference test in which the path between cumulative CS and intentions was constrained to be equal to the path between cumulative PSQ and intentions found that there was a x2 difference of 3.11, p 0:08. This indicates that the effect of cumulative CS on intentions is not signicantly different from the effect of cumulative PSQ on intentions at a 0:05. H2a said that transaction PSQ would be positively related to cumulative PSQ and H2b stated that transaction CS would also be positively related to cumulative CS. Both hypotheses were supported. H2c predicted that cumulative PSQ would have a positive effect on cumulative CS, and H2d predicted that transaction PSQ would have a positive effect on transaction CS, and both of these were supported. Model renements and mediation tests The proposed model makes a number of implicit mediation predictions, such as the cumulative constructs completely mediate the effects of transaction constructs on intentions. These over-identifying restrictions were tested by estimating additional paths between constructs. There are four criteria to demonstrate complete mediation. First, we must show that the two transaction constructs have signicant simple relationships to behavioral intentions, i.e. there are effects to be mediated. To show this, we ran the model with the paths from transaction PSQ to intentions and from transaction CS to intentions, but without the paths from CPSQ to intentions and from CCS to intentions. We found that the path from TPSQ to intentions was signicant (0.78, t 11:38), and the path from TCS to intentions was signicant (0.48, t 11:58). Thus, there are effects to be mediated. Second, we must show 543

that the two transaction constructs are signicantly related to the two cumulative constructs and the results of H2a and H2b show that this is the case. Third, we must show that the mediators (the two cumulative constructs) are signicantly related to intention. The results of H1a and H1b show that this is also the case. Finally, we must show that when direct paths between the two cumulative constructs on intention are present, the effects of the direct paths of the transaction constructs are non-signicant. These tests show that the direct path of TPSQ on intentions is insignicant (0.04, t 0:20) and the direct path of TCS on intentions is also insignicant (0.13, t 1:22). The above tests indicate complete mediation of the effects of the two transaction constructs on intentions by the two cumulative constructs. Table III shows the standardized parameter estimates, tvalues, and t statistics for the nal model.

Discussion
Limitations There are several limitations of our study that should be recognized. First, the setting was in a large unit of state government, and therefore may be different from the private sector. As stated above, however, for most products purchasers were free to seek suppliers from the private sector, so the purchasing and distribution center studied was operating in a competitive environment. In addition, many private-sector rms have constraints in their buying due to long-term agreements, preferred/acceptable supplier lists, and reciprocal buying arraignments. Therefore, we think that the domain studied is similar to competitive, private-sector environments.

Service quality and satisfaction in business-to-business services Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Journal of Business & Industrial Marketing Volume 24 Number 8 2009 537 548

Table III Standardized structural parameters for the nal models

Parameter A. Cumulative PSQ ! intention B. Cumulative CS ! intention C. Cumulative PSQ ! cumulative CS D. Transaction PSQ ! cumulative PSQ E. Transaction CS ! cumulative CS F. Transaction PSQ ! transaction CS Final model Completely standardized coefcient 0.53 0.29 0.70 0.84 0.25 0.90 64.98 22 0.98 0.98 0.99 0.99 0.93 0.03 0.08

t-value
4.18 2.34 9.97 17.93 4.07 17.93

x2 df NFI NNFI CFI IFI GFI RMR RMSEA

Second, as a cross-sectional study we do not look at the initial formation of service quality perceptions and the development, over time, of cumulative PSQ. Similarly, we do not measure the build-up of cumulative satisfaction from initial and subsequent transaction satisfaction experiences. To do so would require manipulated experiences in a laboratory setting. Implications for theory The primary implications for theory include: . demonstrating the distinction between CS and PSQ; . specifying, based on theory and logic, the causal ordering between transaction constructs and cumulative constructs, and between PSQ and CS; and . assessing the relative impact of PSQ and CS on behavioral intentions. First, one of our goals was to address the issue of the distinction between PSQ and CS. A huge literature exists on these two constructs, some of which indicates that they are distinct, while some indicates that they are not. We found empirical discrimination between PSQ and CS at both the transaction and the cumulative levels. While the correlations among the four constructs were high, the discrimination test using average variance extracted, which is generally considered a rather stringent test, indicated that each construct was distinct. Second, based on the logic of transaction constructs being antecedents of cumulative constructs, we specied transaction PSQ as an antecedent of cumulative PSQ, and transaction CS as an antecedent of cumulative CS. In addition, based on cognitive appraisal theories of emotions, we specied that the cognitive evaluation of PSQ would be antecedent to the more emotional construct of CS. While our method of analysis cannot denitively establish the causal ordering of PSQ and CS, our results do lend support to the PSQ ! CS ordering. Finally, we assess the relative impact of PSQ and CS on behavioral intentions. Our results indicate that the four PSQ/ CS constructs can predict behavioral intentions in a single model. It was expected that cumulative CS would have a 544

stronger effect on intention than cumulative PSQ (H1c). However, our results showed this not to be the case in that the effects were not signicantly different from each other even though the effect of CPSQ on intention was almost twice as large as the effect of CCS on intention. This would seem to indicate that in a typical consumer purchase situation, the emotional aspects (CS) of the process may well outweigh the more cognitively based evaluations (PSQ) in inuencing repurchase intentions, but in the business-to-business setting in which this research was carried out, the more cognitively based PSQ plays at least an equal, if not greater, role in determining intentions. Thus, while consumers may repurchase from the same supplier or switch to another supplier based largely on their level of cumulative satisfaction, business decisions are likely to be less emotion based and more cognitively based. Implications for managerial practice and future research From a managerial perspective, the nding that the effects of transaction-specic constructs on intentions are completely mediated by the corresponding cumulative constructs is important. It implies that one negative transaction outcome may not be sufcient to cause the customer to switch if the cumulative levels are sufciently positive. Thus, a negative outcome may be discounted by the user if it is seen as a unique occurrence. However, a series of successive negative transaction outcomes may cause the cumulative constructs to become less positive resulting in lower intentions to repurchase from the same supplier. This may occur even if remedial action, such as repair or replacement, is taken for each of the negative outcomes. A user is likely to tolerate only so many negative experiences before taking some sort of action themselves. Therefore, in the event of a negative transaction outcome, the supplier must act to convince the customer that it was a unique occurrence and is unlikely to happen again. In this situation, two options are worth exploring in future research. One option is for the supplier to offer an excuse for the negative outcome. An excuse is an attempt to show that the factors that caused the negative

Service quality and satisfaction in business-to-business services Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Journal of Business & Industrial Marketing Volume 24 Number 8 2009 537 548

outcome were beyond the suppliers control (Weiner, 2000). If the excuse is accepted as valid, then no negative repercussions (i.e. switching suppliers) are likely. The other option is to confess that the negative outcome was the suppliers fault. A confession has the effect of reducing the belief that the negative outcome will occur again (Weiner, 2000). Of course, the efcacy of either option will diminish as the number of negative outcomes increases. Both of these options need to be investigated in future research. Future research could also utilize other settings to examine the relative impact of CCS and CPSQ on intentions. In situations in which there is less interpersonal interaction, it may be that CPSQ has less of an effect on intentions. Or, in cases in which failure has extremely negative consequences for the buyer, the emotion created could increase the effect of CCS on intentions. Finally, our ndings that the two cumulative constructs determinate intentions indicate the need to consider history and long-term orientation, which are fundamental in relationship-oriented views (Holmlund, 2004). However, additional research could examine whether transaction constructs, in some circumstances, have direct effects on intention. It is easy to imagine situations in a business-tobusiness setting in which the ramications of even a single negative experience are so severe that the buyer would never again purchase from the supplier. In this case, there might be direct effects of the transaction constructs on intentions that are not mediated by the cumulative constructs.

References
Abelson, R.P., Kinder, D.R., Peters, M.D. and Fiske, S.T. (1982), Affective and semantic components in political person perception, Journal of Personality and Social Psychology, Vol. 42, April, pp. 619-30. Allen, C.T., Machleit, K.A. and Kleine, S.S. (1992), A comparison of attitudes and emotions as predictors of behavior at diverse levels of behavioral experience, Journal of Consumer Research, Vol. 18 No. 4, pp. 493-505. Anderson, E.W., Fornell, C. and Lehmann, D.R. (1994), Customer satisfaction, market share, and protability: ndings from Sweden, Journal of Marketing, Vol. 58, July, pp. 53-66. Andrus, D.M. and Norvell, W. (1990), Marketing differences between large and small rms: a test of the industrial/consumer dichotomy hypothesis, Akron Business and Economic Review, Vol. 21 No. 3, pp. 81-90. Bagozzi, R.P. (1992), The self-regulation of attitudes, intentions, and behavior, Social Psychology Quarterly, Vol. 55, pp. 178-204. Bagozzi, R.P., Gopinath, M. and Nyer, P.U. (1999), The role of emotion in marketing, Journal of the Academy of Marketing Science, Vol. 27 No. 2, pp. 184-206. Bentler, P.M. (1995), EQS: Structural Equations Program Manual, Multivariate Software, Inc., Encino, CA. Bergkvist, L. and Rossiter, J.R. (2007), The predictive validity of multiple-item versus single-item measures of the same constructs, Journal of Marketing Research, Vol. XLIV, May, pp. 175-84. Bitner, M.J. (1990), Evaluating service encounters: the effects of physical surroundings and employee responses, Journal of Marketing, Vol. 54, April, pp. 69-82. 545

Bitner, M.J. and Hubbert, A.R. (1994), Encounter satisfaction versus overall satisfaction versus quality, in Rust, R. and Oliver, R.L. (Eds), Service Quality: New Directions in Theory and Practice, Sage Publications, London, pp. 72-94. Bolton, R.N. and Drew, J.H. (1991a), A longitudinal analysis of the impact of service changes on customer attitudes, Journal of Marketing, Vol. 55, January, pp. 1-9. Bolton, R.N. and Drew, J.H. (1991b), A multistage model of customers assessments of service quality and value, Journal of Consumer Research, Vol. 17, March, pp. 375-84. Boulding, W., Kalra, A., Staelin, R. and Zeithaml, V.A. (1993), A dynamic process model of service quality: from expectations to behavioral intentions, Journal of Marketing Research, Vol. 30, February, pp. 7-27. Brady, M.K. and Robertson, C.J. (2001), Searching for a consensus on the antecedent role of service quality and satisfaction: an exploratory cross-national study, Journal of Business Research, Vol. 51 No. 1, pp. 53-60. Cadotte, E.R., Woodruff, R.B. and Jenkins, R.L. (1987), Expectations and norms in models of consumer satisfaction, Journal of Marketing Research, Vol. 24, August, pp. 305-14. Caruana, A. (2002), Service loyalty: the effects of service quality and the mediating role of customer satisfaction, European Journal of Marketing, Vol. 36 Nos 7/8, pp. 811-28. Chakraborty, G., Srivastava, P. and Marshall, F. (2007), Are drivers of customer satisfaction different for buyers/users from different functional areas?, Journal of Business & Industrial Marketing, Vol. 22 No. 1, pp. 20-8. Chandrashekaran, M., Rotte, K., Tax, S.S. and Grewal, R. (2007), Satisfaction strength and customer loyalty, Journal of Marketing Research, Vol. XLIV, February, pp. 153-63. Chumpitaz, R. and Paparoidamis, N.G. (2004), Service quality and marketing performance in business-to-business markets: exploring the mediating role of client satisfaction, Managing Service Quality, Vol. 14 Nos 2/3, pp. 235-48. Cohen, J.B. and Areni, C.S. (1991), Affect and consumer behavior, in Robertson, T.S. and Kassarjian, H.J. (Eds), Handbook of Consumer Behavior, Prentice-Hall, Englewood Cliffs, NJ, pp. 189-240. Cooper, P.D. and Jackson, R.W. (1988), Applying a service marketing orientation to the industrial services sector, Journal of Services Marketing, Vol. 2 No. 2, pp. 66-70. Coviello, N. and Brodie, R. (2001), Contemporary marketing practices of consumer and business-to-business rms: how different are they?, Journal of Business & Industrial Marketing, Vol. 16 No. 5, pp. 382-400. Cronin, J.J. Jr and Taylor, S.R. (1992), Measuring service quality: a reexamination and extension, Journal of Marketing, Vol. 56, July, pp. 55-68. Cronin, J.J. Jr, Brady, M.K. and Hult, G.T.M. (2000), Assessing the effects of quality, value, and customer satisfaction on consumer behavioral intentions in service environments, Journal of Retailing, Vol. 76 No. 2, pp. 193-218. Dawes, P.L. and Patterson, P.G. (1988), The performance of industrial and consumer product managers, Industrial Marketing Management, Vol. 17 No. 1, pp. 73-84. Day, E. and Barksdale, H.C. Jr (2003), Selecting a professional service provider from the short list, Journal

Service quality and satisfaction in business-to-business services Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Journal of Business & Industrial Marketing Volume 24 Number 8 2009 537 548

of Business & Industrial Marketing, Vol. 18 Nos 6/7, pp. 564-79. de Ruyter, K. and Wetzels, M. (1998), On the complex nature of patient evaluations of general practice service, Journal of Economic Psychology, Vol. 19, pp. 565-90. de Ruyter, K., Bloemer, J. and Peeters, P. (1997), Merging service quality and service satisfaction: an empirical test of an integrative model, Journal of Economic Psychology, Vol. 18, pp. 387-406. Durvasula, S., Lysonski, S. and Mehta, S. (1999), Testing the SERVQUAL scale in the B2B section: the case of ocean freight shipping, Journal of Services Marketing, Vol. 13 No. 2, pp. 109-26. Fern, E.F. and Brown, J.R. (1984), Industrial consumer marketing dichotomy: a case of insufcient justication, Journal of Marketing, Vol. 48 No. 2, pp. 68-77. Fornell, C. (1992), A national customer satisfaction barometer: the Swedish experience, Journal of Marketing, Vol. 56, January, pp. 6-21. Fornell, C. and Larcker, D.F. (1981), Evaluating structural equation models with unobservable Variables and measurement error, Journal of Marketing Research, Vol. 18, February, pp. 39-50. Gonzalez, M.E., Comesana, L.R. and Brea, J.A.F. (2006), Assessing tourist behavioral intentions through perceived service quality and customer satisfaction, Journal of Business Research, Vol. 60, pp. 153-60. Gotlieb, J.B., Grewal, D. and Brown, S.W. (1994), Consumer satisfaction and perceived quality: complementary or divergent constructs?, Journal of Applied Psychology, Vol. 79 No. 6, pp. 875-85. Gounaris, S.P. (2005), Trust and commitment inuences on customer retention: insights from business-to-business services, Journal of Business Research, Vol. 58 No. 2, pp. 126-40. Heskett, J.L., Jones, T.O., Loveman, G.W., Sasser, W.E. Jr and Schlesinger, L.A. (1994), Putting the service-prot chain to work, Harvard Business Review, Vol. 72, March/ April, pp. 164-74. Holmlund, M. (2004), Analyzing business relationships and distinguishing different interaction levels, Industrial Marketing Management, Vol. 33 No. 4, pp. 279-87. Homburg, C., Koschate, N. and Hoyer, W.D. (2006), The role of cognition and affect in the formation of customer satisfaction: a dynamic perspective, Journal of Marketing, Vol. 70 No. 3, pp. 21-31. Jayawardhena, C., Souchon, A.L., Farrell, A.M. and Glanville, K. (2007), Outcomes of service encounter quality in a business-to-business context, Industrial Marketing Management, Vol. 36 No. 7, pp. 575-88. Lam, S.Y., Shankar, V., Erramilli, M.K. and Murthy, B. (2004), Customer value, satisfaction, loyalty, and switching costs: an illustration from a business-to-business service context, Journal of the Academy of Marketing Science, Vol. 32 No. 3, pp. 293-311. Lazarus, R.S. (1991), Emotion and Adaptation, Oxford University Press, New York, NY. Liu, A.H., Leach, M.P. and Bernhardt, K.L. (2003), Examining customer value perceptions of organizational buyers when sourcing from multiple vendors, Journal of Business Research, Vol. 58 No. 5, pp. 559-68. 546

Luo, X. and Bhattacharya, C.B. (2006), Corporate social responsibility, customer satisfaction, and market value, Journal of Marketing, Vol. 70 No. 4, pp. 1-18. Miciak, A. and Desmarais, M. (2001), Benchmarking service quality performance at business-to-business and business-to-consumer call centers, Journal of Business & Industrial Marketing, Vol. 16 No. 5, pp. 340-53. Nyer, P.U. (1997), A study of the relationships between cognitive appraisals and consumption emotions, Journal of the Academy of Marketing Science, Vol. 25, Fall, pp. 296-304. Oliver, R.L. (1980), A cognitive model of the antecedents and consequences of satisfaction decisions, Journal of Marketing Research, Vol. 17, November, pp. 460-9. Oliver, R.L. (1989), Processing of the satisfaction response in consumption: a suggested framework and research propositions, Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, Vol. 2, pp. 1-16. Oliver, R.L. (1993), A conceptual model of service quality and service satisfaction: compatible goals, different concepts, Advances in Services Marketing and Management, Vol. 2, pp. 65-85. Oliver, R.L. (1994), Conceptual issues in the structural analysis of consumption emotion, satisfaction, and quality: evidence in a service setting, Advances in Consumer Research, Vol. 21 No. 1, pp. 16-22. Oliver, R.L. (1997), Satisfaction: A Behavioral Perspective on the Consumer, McGraw-Hill, New York, NY. Olorunniwo, F., Hsu, M.K. and Udo, G.J. (2006), Service quality, customer satisfaction, and behavioral intentions in the service factory, Journal of Services Marketing, Vol. 20 No. 1, pp. 59-72. Omdahl, B.L. (1995), Cognitive Appraisal, Emotion, and Empathy, Lawrence Erlbaum Associates, Mahwah, NJ. Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1988), SERVQUAL: a multi-item scale for measuring consumer perceptions of service quality, Journal of Retailing, Vol. 64, Spring, pp. 13-40. Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1994), Reassessment of expectations as a comparison standard in measuring service quality: implications for further research, Journal of Marketing, Vol. 58, January, pp. 111-24. Patterson, P.G. and Spreng, R.A. (1997), Modeling the relationship between perceived value, satisfaction and repurchase intentions in a business-to-business, services context: an empirical examination, International Journal of Service Industry Management, Vol. 8 No. 5, pp. 414-34. Patterson, P.G., Johnson, L.W. and Spreng, R.A. (1997), Modeling the determinants of customers satisfaction for business-to-business professional services, Journal of the Academy of Marketing Science, Vol. 25 No. 1, pp. 4-17. Paulssen, M. and Birk, M.M. (2007), Satisfaction and repurchase behavior in a business-to-business setting: investigating the moderating effect of manufacturer, company and demographic characteristics, Industrial Marketing Management, Vol. 36 No. 7, pp. 983-97. Rauyruen, P. and Miller, K.E. (2007), Relationship quality as a predictor of B2B customer loyalty, Journal of Business Research, Vol. 60 No. 1, pp. 21-31. Rust, R.T., Zahorik, A.J. and Keiningham, T.L. (1995), Return on quality (ROQ): making service quality nancially accountable, Journal of Marketing, Vol. 59 No. 2, pp. 58-70.

Service quality and satisfaction in business-to-business services Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Journal of Business & Industrial Marketing Volume 24 Number 8 2009 537 548

Spreng, R.A. and Agrawal, M. (1995), A conceptual and empirical examination of the distinction between perceived service quality and satisfaction, Proceedings of the Society for Consumer Psychology, pp. 103-4. Spreng, R.A. and Mackoy, R.D. (1996), An empirical examination of the antecedents of perceived service quality and satisfaction, Journal of Retailing, Vol. 72 No. 2, pp. 201-14. Spreng, R.A. and Olshavsky, R.W. (1993), A desire congruency model of consumer satisfaction, Journal of the Academy of Marketing Science, Vol. 21 No. 3, pp. 169-77. Spreng, R.A. and Singh, A.K. (1993), An empirical assessment of the SERVQUAL scale, and the relationship between service quality and satisfaction, in Cravens, D.W. and Dickson, P.R. (Eds), Enhancing Knowledge Development in Marketing, Vol. 4, American Marketing Association, Chicago, IL, pp. 1-6. Spreng, R.A., MacKenzie, S.B. and Olshavsky, R.W. (1996), A re-examination of the determinants of consumer satisfaction, Journal of Marketing, Vol. 60, July, pp. 15-32. Sureshchandar, G.S., Chandrasekharan, R. and Anantharaman, R.N. (2002), The relationship between service quality and customer satisfaction a factor specic approach, Journal of Services Marketing, Vol. 16 No. 4, pp. 363-79. Taylor, S.A. and Baker, T.L. (1994), An assessment of the relationship between service quality and customer satisfaction in the formation of consumers purchase intentions, Journal of Retailing, Vol. 70 No. 2, pp. 163-78. Teas, R.K. (1993), Expectations, performance evaluation, and consumers perceptions of quality, Journal of Marketing, Vol. 57, October, pp. 18-34. Tse, D.K. and Wilton, P.C. (1988), Models of consumer satisfaction formation: an extension, Journal of Marketing Research, Vol. 25, May, pp. 204-12. Wathne, K.H., Biong, H. and Heide, J.B. (2001), Choice of supplier in embedded markets: relationship and marketing program effects, Journal of Marketing, Vol. 65 No. 2, pp. 54-66. Weiner, B. (2000), Attributional thoughts about consumer behavior, Journal of Consumer Research, Vol. 27 No. 3, pp. 382-8. Westbrook, R.A. (1987), Product/consumption-based affective responses and postpurchase processes, Journal of Marketing Research, Vol. 24, August, pp. 258-70. Westbrook, R.A. and Reilly, M.D. (1983), Value-percept disparity: an alternative to the disconrmation of expectations theory of consumer satisfaction, in Bagozzi, R.P. and Tybout, A.M. (Eds), Advances in Consumer Research, Vol. 10, Association for Consumer Research, Ann Arbor, MI, pp. 256-61. Woodruff, R.B. (1993), Developing and applying consumer satisfaction research: implications for future research, Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, Vol. 6, pp. 1-11. Yi, Y. (1990), A critical review of consumer satisfaction, in Zeithaml, V.A. (Ed.), Review of Marketing 1990, American Marketing Association, Chicago, IL, pp. 68-123. Zeithaml, V.A., Berry, L.L. and Parasuraman, A. (1993), The nature and determinants of customer expectations of service, Journal of the Academy of Marketing Science, Vol. 21, Winter, pp. 1-12. 547

Further reading
Anderson, E.W., Fornell, C. and Mazvancheryl, S.K. (2004), Customer satisfaction and shareholder value, Journal of Marketing, Vol. 68, October, pp. 172-85. Spreng, R.A. (1999), Perceived performance in satisfaction research, Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, Vol. 12, pp. 100-8.

Corresponding author
Thomas J. Page is the corresponding author and can be contacted at: tpage@msu.edu

Executive summary and implications for managers and executives

This summary has been provided to allow managers and executives a rapid appreciation of the content of the article. Those with a particular interest in the topic covered may then read the article in toto to take advantage of the more comprehensive description of the research undertaken and its results to get the full benet of the material present. While the principle Treat them well because they wont give us a second chance when responding to customers needs might be commendable, the fact is you are more than likely to get a second chance if something goes wrong. But a third or fourth chance? Well, maybe not. Better to always get it right, but if something does go wrong make sure you make your excuses known or, if theres no excuse or no one else to blame, confess, and in either event convince them it will never happen again. In Service quality and satisfaction in business-to-business services, Richard A. Spreng et al. nd that one negative transaction outcome may not be sufcient to cause the customer to switch if the cumulative levels are sufciently positive. Thus, a negative outcome may be discounted by the user if it is seen as a unique occurrence. However, a series of successive negative transaction outcomes may cause the cumulative constructs to become less positive, resulting in lower intentions to repurchase from the same supplier. This may occur even if remedial action, such as repair or replacement, is taken for each of the negative outcomes. A user is likely to tolerate only so many negative experiences before taking some sort of action. Therefore, in the event of a negative transaction outcome, the supplier must act to convince the customer that it is unlikely to happen again. One option is for the supplier to offer an excuse an attempt to show that the factors that caused the negative outcome were beyond the suppliers control. If the excuse is accepted as valid, then no negative repercussions (i.e. switching suppliers) are likely. The other option is to confess that it was the suppliers fault. A confession has the effect of reducing the belief that the negative outcome will occur again. Of course, the efcacy of either option will diminish as the number of negative outcomes increases. When investigating the effects of service quality and service satisfaction in a business-to-business setting, Spreng et al. rst address the issue of the distinction between perceived service quality (PSQ) and customer satisfaction (CS). The majority of PSQ and CS research has been done in the business-toconsumer (B2C) context, and even in research in a business-

Service quality and satisfaction in business-to-business services Richard A. Spreng, Linda Hui Shi and Thomas J. Page

Journal of Business & Industrial Marketing Volume 24 Number 8 2009 537 548

to-business setting the exact nature of the relationship between PSQ, CS, and behavioral intentions leaves unanswered questions. Relationships among constructs may be different in a business-to-business versus a consumer setting, and even between different functional areas within a given rm. This difference might manifest itself, for example, in the impact of PSQ. Since organizational buying often involves more face-toface contact between buyers and sellers, this increased interaction may increase the overall importance of PSQ. The authors support the argument that PSQ is a judgment of performance, while CS is an affective construct. Specically, transaction PSQ is dened as a judgment of performance excellence based on the customers last experience. Cumulative PSQ is a judgment of performance excellence based on all of the customers experience with a particular product or service. Transaction CS is an affective state that is the emotional reaction to the customers last product or service experience. Cumulative CS is an affective state that is the emotional reaction to all of the customers experience with a particular product or service. Disagreement exists about the causal ordering which comes rst, PSQ or CS? Based on the logic of transaction constructs being antecedents of cumulative constructs, Spreng et al. specify transaction PSQ as an antecedent of cumulative PSQ, and transaction CS as an antecedent of

cumulative CS. In addition, based on cognitive appraisal theories of emotions, they specify that the cognitive evaluation of PSQ would be antecedent to the more emotional construct of CS. The results also indicate that the four PSQ/CS constructs can predict behavioral intentions in a single model. It was expected that cumulative CS would have a stronger effect on intention than Cumulative PSQ. However, the results showed this not to be the case in that the effects were not signicantly different from each other, even though the effect of CPSQ on intention was almost twice as large as the effect of CCS. This would seem to indicate that in a typical consumer purchase situation, the emotional aspects (CS) of the process may well outweigh the more cognitively based evaluations (PSQ) in inuencing repurchase intentions, but in the business-tobusiness setting in which this research was carried out, the more cognitively based PSQ plays at least an equal, if not greater, role in determining intentions. So, while consumers may repurchase from the same supplier or switch to another supplier based largely on their level of cumulative satisfaction, business decisions are likely to be less emotionally based and more cognitively based. (A precis of the article Service quality and satisfaction in business-to-business services. Supplied by Marketing Consultants for Emerald.)

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit our web site for further details: www.emeraldinsight.com/reprints

548

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.