Antecedents and consequences of consumers' comparative value assessments across the relationship life cycle

Antecedents and consequences of consumers' comparative value assessments across the relationship life cycle

JBR-07423; No of Pages 7 Journal of Business Research xxx (2011) xxx–xxx Contents lists available at SciVerse ScienceDirect Journal of Business Rese...

244KB Sizes 0 Downloads 9 Views

JBR-07423; No of Pages 7 Journal of Business Research xxx (2011) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Journal of Business Research

Antecedents and consequences of consumers' comparative value assessments across the relationship life cycle John D. Hansen a,⁎, Lauren Skinner Beitelspacher b, George D. Deitz c a b c

Department of Marketing, Economics, and Industrial Distribution, University of Alabama at Birmingham, Birmingham, AL 35294, USA Department of Marketing, Portland State University, Portland, OR 97201, USA Department of Marketing & Supply Chain Management, University of Memphis, Memphis, TN 38152, USA

a r t i c l e

i n f o

Article history: Received 13 July 2010 Accepted 14 October 2011 Available online xxxx Keywords: Customer value Service quality Satisfaction Customer relationship management

a b s t r a c t This study examines the antecedents and consequences of consumers' comparative value assessments across the relationship life cycle. The study positions service quality and the attractiveness of alternatives as value antecedents, and finds that while service quality is more strongly related to value in the exploration, expansion, and commitment life cycle phases, the two variables are of equal importance in the dissolution phase. The study examines the consequential effects of service quality, value, and satisfaction on share-ofcustomer, and finds that the effects associated with service quality and value are much more pronounced. Theoretical and managerial implications are discussed. © 2011 Elsevier Inc. All rights reserved.

1. Introduction Though value is oftentimes examined in conjunction with service quality and satisfaction, knowledge of the interrelationships among the variables and the effects each has on consumers' behaviors remains limited (Lai, Griffin, & Babin, 2009). While some researchers emphasize satisfying consumers (e.g., Fornell, Johnson, Anderson, Jaesung, & Bryant, 1996), others maintain that “Value drives loyalty, not satisfaction” (Neal, 2000, p. 19). Further, though value is conceptualized as the trade-off between benefits received and costs incurred, little is known about the factors capable of influencing the emphasis consumers place on either component. This may contribute to contrasting findings from previous studies. For instance, while Cronin, Brady, and Hult (2000) find that the sacrifice component of value is of minimal importance, Brodie, Whittome, and Brush (2009) report that benefits and costs play an equally important role in driving value assessments. This study addresses these issues and contributes to the literature in the following ways. First, the study advances a model which simultaneously examines the direct and indirect effects of service quality, comparative value, and satisfaction on consumers' behaviors. Though each variable has been researched extensively, most studies have focused only on the bivariate links that exist between the variables and behavioral outcomes (Lai et al., 2009). Second, the study departs from traditional practice by focusing on opportunity costs as opposed to financial ones. Extant research has primarily assessed value in ⁎ Corresponding author. E-mail addresses: [email protected] (J.D. Hansen), [email protected] (L.S. Beitelspacher), [email protected] (G.D. Deitz).

terms of the financial costs incurred; however, the costs associated with foregone opportunities can be significant in competitive markets. Third, the study examines the extent to which these effects are moderated by relationship life cycle phase. Parasuraman (1997) contends that as customer value creation is a dynamic process, the importance of its antecedents is likely to change over time. Through the relationship life cycle the study examines whether this is indeed the case. 2. Theoretical model and hypotheses The theoretical model for the study is presented in Fig. 1. The model holds that consumers' comparative value assessments are cognitively derived through a comparison of service benefits attained against opportunity costs incurred. These assessments facilitate an emotional outcome—the consumer is either satisfied or dissatisfied with the value proposition vis-à-vis others available. Behaviorally, consumers make decisions regarding share-of-purchases allocated to the firm based on their satisfaction with the firm. Consumers' behaviors are also influenced by the availability of alternative offers as a dissatisfied consumer will only be able to allocate a larger percentage of spend to other firms when viable alternatives exist. 2.1. Antecedents of consumers' comparative value assessments Value refers to an “overall assessment of the utility of a product based on the perceptions of what is received and what is given” (Zeithaml, 1988, p. 14). The term comparative value denotes the fact that consumers oftentimes weigh value propositions against one another when making decisions; hence, the focus is on the partner and a comparison level derived through perceptions of what is believed

0148-2963/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jbusres.2011.11.006

Please cite this article as: Hansen JD, et al, Antecedents and consequences of consumers' comparative value assessments across the relationship life cycle, J Bus Res (2011), doi:10.1016/j.jbusres.2011.11.006

2

J.D. Hansen et al. / Journal of Business Research xxx (2011) xxx–xxx

Service Quality

H3

H1a-c

H5 H1 H4

Comparative Value

Relationship Life Cycle Phase

H2a-c

H7

Customer Satisfaction H8

Share-of-Customer

H2

H6

Attractiveness of Alternatives Fig. 1. Antecedents and consequences of consumers' comparative value assessments.

to be available elsewhere. These perceptions provide comparative reference points, and may differ based on context. What is perceived as an acceptable give–get ratio in one context may be unacceptable in another if competing give–get ratios are more attractive in this other context. Service quality and the attractiveness of alternatives are examined as antecedents of consumers' comparative value assessments. Perceptions of service quality are arrived at based on the performance of the service provider (Cronin & Taylor, 1992). Empirical findings have tended to support this performance-based conceptualization and have revealed a positive relationship between service quality and value (e.g., Cronin et al., 2000). Attractive alternatives exist to the extent that viable competing offers are available in the marketplace (Jones, Mothersbaugh, & Beatty, 2000). Research indicates that the likelihood of relationship continuance is enhanced when viable alternatives are not available (Ping, 1993). Thus, while consumers' perceptions of service quality should positively affect their comparative value assessments, the attractiveness of alternative offers should have the opposite effect: H1. Service quality is positively related to comparative value. H2. Attractiveness of alternatives is negatively related to comparative value. 2.2. Moderating Effects of the Relationship Life Cycle Process models of relationship development explicitly recognize and compensate for the fact that marketing relationships are dynamic in nature. These models have drawn from the interpersonal relationship literature in identifying distinct phases relationships progress via a relationship life cycle. Although a number of life cycle models have been presented in the marketing and strategic management literatures (e.g., Dwyer, Schurr, & Oh, 1987; Ford, 1980; Wilson, 1995), the Dwyer et al. (1987) model is utilized for the purposes of this study. The reasons for this choice are three-fold: first, the model provides the most widely recognized and accepted life cycle conceptualization; second, the model focuses specifically on relationships between buyers and sellers; and third, the model is designed for use in consumer and industrial markets alike. This last point is particularly salient given the significant differences that can exist across business and consumer markets. Relative to consumer marketing, the organizational buying process can involve several individuals on the buying side (i.e., the buying center), can be highly complex, may require an extended decision process, involves higher dollar volumes, and focuses on derived over direct demand. Given these differences, utilization of a model deemed appropriate across both contexts is important. As buyers and sellers do not interact in the awareness phase and the sample for the study consists of loyalty program members, this

phase is not included in the analysis. During the exploration phase, the consumer realizes a need and engages in initial prospecting activities to assess the desirability of a long-term relationship with the firm. During the expansion phase, the consumer is receiving increasing benefits and becoming increasingly dependent on the firm. Commitment denotes the most advanced relational phase, as the consumer is receiving a level of benefit that would be difficult to quickly duplicate elsewhere and has made either an implicit or explicit pledge to continue the relationship. The dissolution phase of the relationship, conversely, is characterized by a void in terms of the emotional attachment felt by the consumer. Though repeat patronage behaviors may be exhibited, this lack of attachment suggests that the relationship means little and in some cases is deemed meaningless. The psychological and sociological perspectives provide a foundation for understanding consumers' preferences across the relationship life cycle. Several general theories of relationship development have been advanced within these bodies of literature. Social exchange theory (Thibaut & Kelley, 1959) posits that individuals' initial interactions with others are not only assessed against expectations brought into the relationship (i.e., a comparison level (CL)), but also against expectations of what other relationships might provide (i.e., a comparison level of alternatives (CLALT)). This implies that as consumers initially transact with firms they assess the outcomes of these transactions against their own expectations and their expectations of what alternative firms might be able to provide. Though logical, one of the primary criticisms levied against social exchange theory is that the theory presents a highly rationale approach, one that is cognitively intense and overly economic in nature. Indeed, though individuals assessing others in interpersonal relationships may be willing and able to devote the cognitive resources this process requires, consumers are inundated with hundreds of offers from firms they may potentially enter into relationships with. Given that most consumers have neither the time nor ability to fully assess all of these offers, they must satisfice by balancing the benefits realized through the assessment process against the time and effort the process requires. If this is the case, how do consumers weigh CL against CLALT when making decisions across the relationship life cycle? Do particular points in time exist where consumers focus more on their own expectations vis-à-vis what the firm is providing? At other points in time do consumers focus more on what alternative firms might be able to provide? Social penetration theory (Altman & Taylor, 1973) holds that “interpersonal exchange gradually progresses over time from superficial, nonintimate areas to more intimate, deeper layers of the selves” (p. 10). As relationships evolve to their most intimate points (e.g., across the expansion and commitment life cycle phases), the focus on the partner takes precedent as individuals attempt to gain a deeper understanding of the other party. If this is the case in consumer relationships and, as previously argued, consumers are limited in their ability to simultaneously focus on CL and CLALT, perceptions of

Please cite this article as: Hansen JD, et al, Antecedents and consequences of consumers' comparative value assessments across the relationship life cycle, J Bus Res (2011), doi:10.1016/j.jbusres.2011.11.006

J.D. Hansen et al. / Journal of Business Research xxx (2011) xxx–xxx

the focal firm will take precedent over the possibilities alternative firms present. The relationship between service quality and comparative value will therefore be: H1a. Significantly stronger in the expansion phase relative to the exploration phase. H1b. Not significantly different in the commitment phase relative to the expansion phase. H1c. Significantly weaker in the dissolution phase relative to the commitment phase.

3

attractiveness will be negatively related to share-of-customer. In support of these arguments, the following hypotheses are offered: H5. Service quality is positively related to share-of-customer. H6. Attractiveness of alternatives is negatively related to share-ofcustomer. H7. Comparative value is positively related to share-of-customer. H8. Customer satisfaction is positively related to share-of-customer. 3. Methods

Dwyer et al. (1987) note that consumers in the expansion and commitment life cycle phases virtually preclude all other alternative offers from consideration. However, as uncertainty reduction is a primary goal during the exploration phase of the relationship (Berger & Calabrese, 1975), consumers in this phase will allocate more cognitive energy to ensuring that the relationship is indeed desirable vis-à-vis others which may be available. They will be keenly aware of alternative offers available in the marketplace. Similarly, in the dissolution phase of the relationship, consumers are dissatisfied to the point that they are exploring alternative offers (indeed, most will remain in the relationship only so long as no viable alternatives can be identified). Hence, the relationship between the attractiveness of alternatives and comparative value will be: H2a. Significantly weaker in the expansion phase relative to the exploration phase. H2b. Not significantly different in the commitment phase relative to the expansion phase. H2c. Significantly stronger in the dissolution phase relative to the commitment phase.

2.3. Consequences of consumers' comparative value assessments Satisfaction refers to “a judgment that a product or service feature, or the product or service itself, provided (or is providing) a pleasurable level of consumption-related fulfillment” (Oliver, 1996, p. 112). Researchers have traditionally positioned service quality as an antecedent of satisfaction and found empirical support for this relationship (Brady, Cronin, & Brand, 2002; Cronin et al., 2000). Similarly, empirical evidence supports the notion that satisfaction is influenced by value (Fornell et al., 1996; Spreng, MacKenzie, & Olshavsky, 1996). Consistent with these findings, the following hypotheses are offered: H3. Service quality is positively related to customer satisfaction. H4. Comparative value is positively related to customer satisfaction. Peppers and Rogers (1995) contend that share-of-customer, or the percentage of spending a consumer dedicates to a particular firm in a category (Reichheld & Sasser, 1990), is the truest barometer of relational success. Though several models have explicated the means through which service quality, value, and satisfaction affect behavioral outcomes (cf. Cronin et al., 2000), Cronin et al. (2000) find that each of the variables directly influences consumers' behaviors. Hence, as relationship evaluations and purchase behavior are significantly related (de Jong & Ruyter, 2004), the combined effects of service quality, value, and satisfaction should become visible through share-ofcustomer. However, in some instances consumers will be constrained to relationships due to a lack of alternative offers (Bendapudi & Berry, 1997). When this is the case consumers may allocate a large percentage of purchases to the service provider even though no psychological attachment is felt. This form of spurious loyalty implies that alternative

3.1. Data collection and sample characteristics Data were gathered through a web-based survey. A marketing consulting firm sponsored the study and provided access to customer contact information through the loyalty program database of a global hotel chain. An initial sample of 2000 American customers was randomly drawn from the database. Sample members were contacted via email and asked to complete an online survey in reference to their relationship with the hotel. A total of 784 responses were received for a response rate of 39.2%. Nineteen respondents were removed from the sample due to response irregularities, leaving a sample of 765 respondents. The average respondent age is 47, the annual income for 60% of the sample is between $50,000 and $150,000, and 75% of the respondents are married. A test for non-response bias across all constructs reveals no significant differences (p b .05) between early and late respondents (Armstrong & Overton, 1977). 3.2. Measures Study measures are provided in the Appendix A. Each measure is drawn from previous research and adapted for use. A five-item measure of service quality is adapted from Hellier, Geursen, Carr, and Rickard (2003). The attractiveness of alternatives measure is drawn from Jones et al. (2000), while the comparative value scale is adapted from Grewal, Monroe, and Krishnan (1998). A three-item measure of global satisfaction is formed from the work of Jap (2001) and Ruekert and Churchill (1984), and share-of-customer is assessed as a percentage via a single item. In order to assess the effects of relationship life cycle phase, Anderson (1995) suggests a cross-sectional design wherein respondents are classified to a particular phase and a multi-sample approach is used to empirically assess differences across samples. This approach is employed through a categorical measure developed by Jap and Ganesan (2000). The measure contains brief descriptions of the phases and asks respondents to choose the description which best describes their relationship. In order to avoid any biases, the life cycle measure was presented last in the survey. In all, 136 respondents (17.8%) are classified to the exploration phase, 291 respondents (38.0%) are classified to the expansion phase, 192 respondents (12.1%) are classified to the commitment phase, and 146 respondents (19.1%) are classified to the dissolution phase. 3.3. Measurement model In specifying the measurement model, a reliability of .90 is assumed for the share-of-customer item, with the factor loading set equal α 2 and the error variance equal 1 − α (Hayduk, 1987). Model results indicate a good fit to the data per the recommendations of Hu and Bentler (1999): χ 2(95) = 300.48, SRMR = .031, RMSEA = .055 (CI90%, .048 to .062), NNFI = .99, CFI = .99, and IFI = .99. Convergent validity is assessed by examining individual item loadings, all of which are significant (p b .01). Reliability is assessed by computing composite reliability (CR) and average variance extracted (AVE)

Please cite this article as: Hansen JD, et al, Antecedents and consequences of consumers' comparative value assessments across the relationship life cycle, J Bus Res (2011), doi:10.1016/j.jbusres.2011.11.006

4

J.D. Hansen et al. / Journal of Business Research xxx (2011) xxx–xxx

scores (Steenkamp & van Trijp, 1991). Each construct is above the minimal score of .60 for CR and .50 for AVE (Bagozzi & Yi, 1988). In support of discriminant validity, the AVE by each underlying construct is greater than the highest shared variance with all other latent constructs (Fornell & Larcker, 1981). Table 1 contains the correlation matrix, means, standard deviations, and reliabilities for all measures. Jap and Ganesan (2000) contend that the efficacy of life cycle descriptors can be assessed based on the mean differences seen across the phases. The Dwyer et al. (1987) conceptualization implies that consumers' perceptions should be most favorable in the expansion and commitment life cycle phases, and least favorable in the exploration and dissolution phases. In order to assess whether this is the case, a MANOVA is conducted with life cycle phase positioned as the independent factor and all other study constructs as dependent variables. The results from this analysis are presented in Table 2. As can be seen in the table, life cycle phase is a significant predictor of differences across all dependent variables. The pattern of results indicates that consumers in the commitment phase hold the most favorable perceptions, followed by consumers in the expansion phase. Further, with the exception of share-of-customer, consumers in the exploration phase of the relationship hold significantly more favorable perceptions that do consumers in the dissolution phase.

Table 2 Mean differences across relationship life cycle phases.

Means Exploration Expansion Commitment Dissolution F-value Scheffe Post Hoc Tests Exploration–expansion Exploration–commitment Exploration–dissolution Expansion–commitment Expansion–dissolution Commitment–dissolution

SQ

AAL

CPV

SAT

SOC

4.80 5.20 5.62 3.66 106.42*

5.25 5.04 4.72 5.94 40.03*

4.06 4.51 4.99 2.82 112.88*

4.78 5.18 5.64 3.42 130.67*

22.75 39.45 52.51 18.27 66.98*

.00 .00 .00 .00 .00 .00

.31 .00 .00 .01 .00 .00

.00 .00 .00 .00 .00 .00

.00 .00 .00 .00 .00 .00

.00 .00 .52 .00 .00 .00

NOTE: *p b .01. SQ = service quality, AAL = attractiveness of alternatives, CPV = comparative value, SAT = satisfaction, SOC = share-of-customer.

not supported, however, as satisfaction is found to have no impact on share-of-customer (SDE = .00, t = .03). This finding highlights the prominent role value plays as a mediating mechanism linking consumers’ service appraisals and behaviors, and reveals that the relationship between consumers' cognitive assessments and their behaviors is not mediated by their attitudes. Examining this further, a competing non-mediating model (minus the path from satisfaction to share-of-customer) is examined. Results from the comparison of the two models reveals that the χ 2 statistic for the non-mediating model is unchanged despite the fact a path is removed and degree of freedom gained: Δχ 2(1) = .00. This finding indicates that inclusion of the path from satisfaction to share-of-customer does not enhance model fit, despite the fact a degree of freedom is lost. The non-mediating model would therefore appear to provide a better representation of reality. The standardized total effects of each cognitive component (i.e., service quality, attractiveness of alternatives, and comparative value) on share-of-customer are next examined. Results from this analysis indicate that each of the three variables is similar in terms of influence: service quality → share-of-customer STE = .31, t = 9.07; attractiveness of alternatives → share-of-customer STE = −.30, t = −8.27; and comparative value → share-of-customer STE = .34, t = 6.73. Thus, though consumers appear to focus more on what they are receiving when arriving at their value assessments, their behaviors appear to be equally affected by what they are receiving and what they are having to give up in return.

4. Findings Standardized path coefficients, t-values, and r-squared values from the structural model analysis are presented in Table 3. Fit statistics reveal a good fit between the model and data: χ 2(96) = 300.68, SRMR = .031, RMSEA = .054 (CI90%, .048 to .061), NNFI = .99, CFI = .99, and IFI = .99. The model explains 55% of the variance in comparative value. In support of H1 and H2, comparative value is positively associated with service quality (standardized direct effect [SDE] = .62, t = 20.28) and negatively associated with the attractiveness of alternatives (SDE = −.26, t = − 8.66). The model explains 73% of the variance in satisfaction. Hypotheses 3 and 4 are supported as service quality (SDE = .54, t = 16.48) and comparative value (SDE = .38, t = 11.80) are positively associated with satisfaction. The fact that the relationship between service quality and satisfaction is significant even when controlling for the effects of comparative value indicates that comparative value only partially mediates the service quality– satisfaction relationship. The standardized total effects (STE) of service quality and the attractiveness of alternatives on customer satisfaction are additionally examined. The effects of service quality on satisfaction (STE = .78, t = 28.01) are more pronounced than are the effects of alternative attractiveness on satisfaction (STE = −.10, t = − 7.02). This finding is not surprising given that the attractiveness of alternatives plays a more prominent role in developing constraint-based as opposed to dedication-based loyalty (Bendapudi & Berry, 1997). The model explains 30% of the variance in share-of-customer. In support of H5 and H6, the relationship between service quality and share-of-customer is significant and positive (SDE = .10, t = 1.72), while the relationship between alternative attractiveness and shareof-customer is significant and negative (SDE = −.21, t = −5.61). Hypothesis 7 is supported as well as comparative value is significantly related to share-of-customer (SDE = .34, t = 6.01). Hypothesis 8 is

4.1. Moderating effects of relationship life cycle phase In order to test the moderating effects of relationship life cycle phase, a multi-group analysis is performed utilizing the three constructs of interest (service quality, attractiveness of alternatives, and comparative value). As an initial step, metric invariance is assessed across all four life cycle groups to ensure that respondents conceptualized the constructs similarly. Results from this analysis indicate that imposing metric invariance against a baseline model produces a slight deterioration in model fit (Δχ 2(27) = 41.71, p = .04). As full metric invariance is not satisfied, partial metric invariance is investigated to

Table 1 Descriptive statistics and correlations. Construct

M

SD

CA

CR

AVE

1

2

3

4

5

1. 2. 3. 4. 5.

4.94 5.17 4.23 4.89 35.71

1.24 1.12 1.35 1.32 28.04

.95 .86 .94 .96

.95 .86 .94 .96

.80 .66 .80 .89

1.00 − .29 .70 .81 .40

1.00 −.44 −.33 −.39

1.00 .76 .50

1.00 .41

1.00

Service Quality Attractiveness of Alternatives Comparative Value Satisfaction Share-of-Customer

NOTE: All correlations significant, p b .05 (two-sided testing). CA = Cronbach's alpha; CR = composite reliability; AVE = average variance extracted.

Please cite this article as: Hansen JD, et al, Antecedents and consequences of consumers' comparative value assessments across the relationship life cycle, J Bus Res (2011), doi:10.1016/j.jbusres.2011.11.006

J.D. Hansen et al. / Journal of Business Research xxx (2011) xxx–xxx Table 3 Structural Model Estimates. Path

Hypothesis β

t-value

Service quality → Comparative value Attractiveness of Alternatives → Comparative value Service quality → Satisfaction Comparative value → Satisfaction Service quality → Share-of-customer Attractiveness of Alternatives → Share-of-customer Comparative value → Share-of-customer Satisfaction → Share-of-customer R2 for comparative value R2 for satisfaction R2 for share-of-customer df χ2 SRMR RMSEA NNFI CFI IFI

H1 H2 H3 H4 H5 H6 H7 H8 .55 .73 .30 96 300.68 .031 .054 .99 .99 .99

20.28 − 8.66 16.48 11.80 1.72 − 5.61 6.01 .03

.62* − .26* .54* .38* .10* −.21* .34* .00

5

significantly across the exploration and expansion relational phases (Δχ2 = 6.75, p = .01), remains constant in strength across the expansion and commitment phases (Δχ2 = .68, p = .41), and becomes significantly stronger across the commitment and dissolution phases (Δχ2 = 9.27, p = .00). Thus, each of the three hypotheses is supported. Attractiveness of alternatives explains 18%, 4%, 3%, and 24% of the variance in comparative value across the exploration, expansion, commitment, and dissolution life cycle phases, respectively. The effects associated with attractiveness of alternatives would therefore appear to be weakest during the expansion and commitment phases of the relationship, and strongest during the exploration and dissolution phases. 5. Discussion

NOTE: Standardized estimates provided; *p b .05 (one-sided testing).

identify those loadings that should be left unconstrained (cf., Byrne, Shavelson, & Muthén, 1989). An examination of the modification indices reveals that the fourth service quality item ([Hotel X's] employees have the knowledge to answer my questions) should be left unconstrained in the dissolution sample. As partial metric invariance is supported (Δχ 2(26) = 34.98, p = .11), the item is left unconstrained for all moderation tests. Results from these tests are presented in Table 4. The table contains completely standardized parameter estimates for each of the paths, the increase in χ 2 resulting from each path being constrained equal, and the r-squared statistic for comparative value for the full model and individual models containing only service quality and attractiveness of alternatives as independent variables. With the exception of the attractiveness of alternatives → comparative value path in the commitment sample, all paths are significant. In testing H1a–c, the service quality → comparative value path is sequentially constrained equal across groups. The corresponding increase in χ 2 is examined for significance (Δχ 2 > 3.84). As can be seen, the difference across the exploration–expansion phases (Δχ 2 = .09, p = 76) and the expansion-commitment phases (Δχ 2 = .07, p = .79) is not significant. Thus, H1a is rejected and H1b is supported. In support of H1c, the relationship between service quality and comparative value is significantly weaker in the dissolution phase of the relationship relative to the commitment phase (Δχ 2 = 4.35, p = .04). Service quality accounts for 27%, 31%, 35%, and 31% of the variance in comparative value across the exploration, expansion, commitment, and dissolution life cycle phases, respectively. Results from the analysis of H2a-c indicate that the relationship between attractiveness of alternatives and comparative value diminishes

This study addresses two gaps in the literature. First, though most research conceptualizes value in terms of the trade-off between benefits received and costs incurred, knowledge of the contextual factors capable of influencing the emphasis consumers place on either of these two components is limited. Second, though research examining the consequences of value typically does so in conjunction with service quality and satisfaction, little is known about the interrelationships that exist among the three variables as well as the direct and indirect effects each has on consumers' behaviors. Study results indicate that while service quality is more capable of shaping consumers' comparative value assessments in the exploration, expansion, and commitment life cycle phases, the attractiveness of alternatives is focused on heavily as well in the dissolution phase. Study results additionally reveal that the effects associated with service quality and value are much more pronounced than those associated with satisfaction. The theoretical and managerial implications arising from these results are discussed in the section that follows. 5.1. Theoretical and managerial implications Study findings bring some clarity to contrasting findings from previous studies regarding the level of emphasis consumers place on benefits received versus costs incurred when assessing value. As noted, Brodie et al. (2009) report that each of the components plays an equally important role while Cronin et al. (2000) find that consumers focus on benefits received to the virtual exclusion of costs incurred. The fact that relationship life cycle phase emerges in this study as a contextual variable capable of differentially affecting consumers' value assessments might partially explain these contrasting results. Specifically, results indicate that service quality is most capable of influencing value in the expansion and commitment life cycle phases, while the attractiveness of alternatives plays an equally important role in the exploration and dissolution phases. This finding is in accordance with research from Flint, Woodruff, and Gardial (2002), who note that as customers' value assessments are dynamic in nature firms

Table 4 Tests for moderation. Hypothesis

Exploration

Expansion

Commitment

Dissolution

H1 H2

.60* −.41*

.57* − .17*

.59* − .10

.40* − .41*

.09 6.75*

.07 .68

4.35* 9.27*

.33 .31 .04

.36 .35 .03

.44 .31 .24

β Service quality → Comparative value Attractiveness of alternatives → Comparative value Δχ2 Service quality → Comparative value Attractiveness of alternatives → Comparative value R2 statistics Full model a Service quality b Attractiveness of alternatives c

H1a-c H2a-c .45 .27 .18

NOTE: Common metric standardized estimates provided; *p b .05. a Service quality and attractiveness of alternatives modeled as independent variables; modeled as independent variable; c attractiveness of alternatives alone modeled as independent variable.

b

Service quality alone

Please cite this article as: Hansen JD, et al, Antecedents and consequences of consumers' comparative value assessments across the relationship life cycle, J Bus Res (2011), doi:10.1016/j.jbusres.2011.11.006

6

J.D. Hansen et al. / Journal of Business Research xxx (2011) xxx–xxx

must differentially shape value propositions depending on how developed the relationship is. The finding carries significant managerial and theoretical implications. Managerially, as service firms communicate with their consumers, the extent to which competitive communications are utilized may partly depend on which segment of consumers the communication is being targeted to. Researchers have long since debated the merits of comparative communications, and have noted that the effectiveness of these type communications depends on factors such as whether the focal brand has a dominant or follower market position (Murphy & Amundsen, 1981) and whether the communication incorporates two-sided claims (Muehling, Stoltman, & Grossbart, 1990). Study results from this study highlight the fact that relationship life cycle phase may be a variable managers should consider as well. Specifically, early and late in the relationship, as consumers' value assessments are influenced by the availability and attractiveness of alternative offers, service managers would be well-advised to focus on communications that highlight value propositions relative to others available. Conversely, when the relationship advances across the expansion and commitment life cycle phases, communications that focus only on the focal firm may be in order. Theoretically, the finding supports the traditional relationship life cycle conceptualization of Dwyer et al. (1987). In a recent study, Blut et al. (2011) empirically contrast this traditional life cycle trajectory against a u-shaped trajectory in an examination of interorganizational franchise relationships. The authors find that these relationships are characterized by four distinct phases—a honeymoon phase, a routine phase, a crossroad phase, and a stabilization phase—which do indeed display an inverted u-shaped trajectory. Relationship properties are at their highest in the honeymoon phase, diminish across the routine and crossroads phases, before increasing again in the stabilization phase. While differences in trajectory may be attributable to the fact that the current study is conducted in the consumer context, previous empirical research in the business-to-business context also supports the traditional trajectory (Jap, 2001). Nonetheless, the emergence of a u-shaped relationship trajectory is an interesting finding warranting additional investigation across a variety of contexts. The impact service quality, value, and satisfaction have on behavioral outcomes and the interrelationships that exist among the variables has been a source of considerable debate. Results from this study strongly support the notion that consumers' behaviors are largely a function of their cognitions. In a recent examination of consumers of a Chinese mobile communications company, Lai et al. (2009) similarly report that value affects loyalty more strongly than does satisfaction (.48 vs. .39, respectively)—satisfaction only partially attenuates the value-loyalty link. In unison, these findings lend credence to those who argue that value is the dominant variable capable of influencing behaviors, not satisfaction. 5.2. Study limitations and future research directions Study limitations must be considered when interpreting the findings uncovered. First, the design of the study is not longitudinal in nature. Despite the difficulties longitudinal research studies present, the process dynamics and cumulative effects present in developing relationships are difficult to uncover when collecting data at only one point in time. Second, the variables examined in the study are by no means exhaustive. Aside from those examined here, many other variables capable of influencing desirable relational outcomes aside from those examined in the study. However, the intent of the study was not to provide an exhaustive of all such factors but rather to provide a focused examination of the variables examined across the various life cycle phases. Third, the behaviorally-oriented relationship outcome variable share-of-customer is measured via survey as no database information was available. Though examining true behaviors would have been more optimal, firms have a difficult time

assessing share-of-customer without first asking customers about overall category purchases. Lastly, the sample for the study is limited in scope to hotel customers and all findings should be interpreted within this context. Mindful of this, future research should first seek to replicate studies of this sort across a broader range of companies and industries. This will allow for a better understanding of whether study findings are consistent across other contexts or for some reason particular to the hotel context. Second, the efficacy of different life cycle classificatory tools should be examined in future research. Of particular relevance would be the use of secondary data to determine which variables are capable of accurately predicting consumer placement in the life cycle through discriminant analysis (i.e., if secondary data were available, one could compare the extent to which the secondary data utilized is accurately able to predict how respondents actually classify themselves to the differing life cycle phases). This is an issue of critical importance as managers' ability to act upon research of this type is dependent on their ability to classify consumers in their database to the correct relational phase. Third, although the difficulties associated with longitudinal research have been noted, research in this area would benefit from a longitudinal design. The ability to track individual consumers through the relationship life cycle, as opposed to comparing different consumers at varying points in the relationship, would appear to be worth the considerable effort such research would require. Appendix A. Construct Scale Items

Construct Service quality—Hellier et al. (2003) [Hotel X's] employees are consistently courteous to me. [Hotel X's] employees provide services at the time they promise to do so. [Hotel X's] employees are always willing to help me. [Hotel X's] employees have the knowledge to answer my questions. The behavior of [Hotel X's] employees instills confidence in me. Attractiveness of alternatives—Jones et al. (2000) Compared to [Hotel X], there are many other good hotels from which I can choose. Compared to [Hotel X], I am as happy or happier with the service other hotels provide me. Compared to [Hotel X], there are other hotels with which I am equally or more satisfied. Comparative value—Grewal et al. (1998) Compared to alternatives, the benefits that I receive from [Hotel X] are consistently fair for the effort and money I spend. I am willing to forgo the benefits that other hotels might offer because of the advantages that [Hotel X] provides. Time after time, the value I receive from [Hotel X] is better than the value I receive from other hotels. After evaluating [Hotel X's] offerings against competitors', I am confident that I am receiving quality service for the price. Satisfaction—Ruekert and Churchill (1984) I am typically satisfied when I stay at [Hotel X]. I am typically pleased when I stay at [Hotel X]. I typically have my expectations fulfilled when I stay at [Hotel X]. Share-of-customer As a percentage, how often do you stay at [Hotel X] compared to other hotels? Relationship Life Cycle Phase—Jap and Ganesan (2000) Relationships between customers and companies typically evolve through a number of relationship phases over time. Which of the following best describes your current relationship with [Hotel X]? (Please check only one) Exploration—I am a new [Hotel X] customer and am discovering the benefits and obligations a long-term relationship with [Hotel X] entails. Expansion—I am receiving increasing levels of benefits and satisfaction from [Hotel X] such that I am more willing to become committed to a relationship with [Hotel X] on a long-term basis. Commitment—I am receiving high levels of benefits and am highly satisfied with [Hotel X] such that I stay at [Hotel X] if at all possible. Dissolution—I am receiving decreasing benefits and am dissatisfied with [Hotel X] such that I am contemplating relationship termination and considering other hotels for my lodging needs. NOTE: All loadings significant, p b .01. χ2 = 412.51(95), SRMR = .030, RMSEA = .051, NNFI = .99, IFI = .99, CFI = .99. With the exception of share-of-customer, all constructs measured on seven-point scale (1 = strongly disagree, 7 = strongly agree).

Please cite this article as: Hansen JD, et al, Antecedents and consequences of consumers' comparative value assessments across the relationship life cycle, J Bus Res (2011), doi:10.1016/j.jbusres.2011.11.006

J.D. Hansen et al. / Journal of Business Research xxx (2011) xxx–xxx

References Altman I, Taylor DA. Social penetration: the development of interpersonal relationships. New York: Holt, Rhinehart, and Winston; 1973. Anderson JC. Relationships in business markets: exchange episodes, value creation, and their empirical assessment. Journal of the Academy of Marketing Science 1995;23 (4):346–50. Armstrong JS, Overton TS. Estimating nonresponse bias in mail surveys. Journal of Marketing Research 1977;14(3):396–402. Bagozzi RP, Yi Y. On the evaluation of structural equation models. Journal of the Academy of Marketing Science 1988;16(2):74–94. Bendapudi N, Berry LL. Customers' motivations for maintaining relationships with service providers. Journal of Retailing 1997;73(1):15–37. Berger CR, Calabrese RJ. Some explorations in initial interaction and beyond: toward a developmental theory of interpersonal communication. Human Communication Research 1975;1:99-112. Blut M, Backhaus C, Heussler T, Woisetschläger DM, Evanschitzky H, Ahlert D. What to expect after the honeymoon: testing a lifecycle theory of franchise relationships. Journal of Retailing 2011;87(3):306–19. Brady MK, Cronin Jr JJ, Brand RR. Performance-only measurement of service quality: a replication and extension. Journal of Business Research 2002;55(1):17–31. Brodie RJ, Whittome JRM, Brush GJ. Investigating the service brand: a customer value perspective. Journal of Business Research 2009;62(3):345–55. Byrne BM, Shavelson RJ, Muthén B. Testing for the equivalence of factor covariance and mean structures: the issue of partial measurement invariance. Psychological Bulletin 1989;105(3):456–66. Cronin Jr JJ, Brady MK, Hult GTM. Assessing the effects of quality, value, and customer satisfaction on consumer behavioral intentions in service environments. Journal of Retailing 2000;76(2):193–216. Cronin Jr JJ, Taylor SA. Measuring service quality: a reexamination and extension. Journal of Marketing 1992;56(3):55–68. de Jong A, Ruyter KD. Adaptive versus proactive behavior in service recovery: the role of self-managing teams. Decision Sciences 2004;35(3):457–91. Dwyer FR, Schurr PH, Oh S. Developing buyer–seller relationships. The Journal of Marketing 1987;51(2):11–28. Flint DJ, Woodruff RB, Gardial SF. Exploring the phenomenon of customers' desired value change in a business-to-business context. Journal of Marketing 2002;66(4): 102–17. Ford D. The development of buyer–seller relationships in industrial markets. European Journal of Marketing 1980;14(5/6):339–53. Fornell C, Johnson MD, Anderson EW, Jaesung C, Bryant BE. The American customer satisfaction index: nature, purpose, and findings. Journal of Marketing 1996;60 (4):7-18. Fornell C, Larcker DF. Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research 1981;18(3):39–50. Grewal D, Monroe KB, Krishnan R. The effects of price-comparison advertising on buyers' perceptions of acquisition value, transaction value, and behavioral intentions. Journal of Marketing 1998;62(2):46–59.

7

Hayduk LA. Structural equation modeling with LISREL: essentials and advances. Baltimore, MD: Johns Hopkins University Press; 1987. Hellier PK, Geursen GM, Carr RA, Rickard JA. Customer repurchase intention: a general structural equation model. European Journal of Marketing 2003;37(11/12): 1762–800. Hu L-T, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Structural Equation Modeling 1999;6 (1):1-55. Jap SD. The strategic role of the salesforce in developing customer satisfaction across the relationship lifecycle. Journal of Personal Selling & Sales Management 2001;21(2):95-108. Jap SD, Ganesan S. Control mechanisms and the relationship lifecycle: implications for safeguarding specific investments and developing commitment. Journal of Marketing Research 2000;37(2):227–45. Jones MA, Mothersbaugh DL, Beatty SE. Switching barriers and repurchase intentions in services. Journal of Retailing 2000;76(2):259–74. Lai F, Griffin M, Babin BJ. How quality, value, image, and satisfaction create loyalty at a Chinese telecom. Journal of Business Research 2009;62(10):980–6. Muehling DD, Stoltman JJ, Grossbart S. The impact of comparative advertising on levels of message involvement. Journal of Advertising 1990;19(4):41–50. Murphy JH, Amundsen MS. The communications-effectiveness of comparative advertising for a new brand on users of the dominant brand. Journal of Advertising 1981;10(1):14–48. Neal W. When measuring loyalty satisfactorily, don't measure CS. Mark News 2000;34 (13):19. Oliver RL. Satisfaction: a behavioral perspective on the consumer. New York: McGrawHill; 1996. Parasuraman A. Reflections on gaining competitive advantage through customer value. Journal of the Academy of Marketing Science 1997;25(2):154–61. Peppers D, Rogers M. A new marketing paradigm: share of customer, not market share. Planning Review 1995;23(2):14–8. Ping RA. The effects of satisfaction and structural constraints on retailer exiting, voice, loyalty, opportunism, and neglect. Journal of Retailing 1993;69(4):320–52. Reichheld FF, Sasser Jr WE. Zero defections: quality comes to service. Harvard Business Review 1990;69(1):105–11. Ruekert RW, Churchill GA. Reliability and validity of alternative measures of channel member satisfaction. Journal of Marketing Research 1984;21(2):226–33. Spreng RA, MacKenzie SB, Olshavsky RW. A reexamination of the determinants of consumer satisfaction. Journal of Marketing 1996;60(3):15–32. Steenkamp J-BEM, van Trijp HCM. The use of LISREL in validating marketing constructs. International Journal of Research in Marketing 1991;8(4):283–99. Thibaut JW, Kelley HH. The social psychology of groups. New York: John Wiley & Sons; 1959. Wilson DT. An integrated model of buyer–seller relationships. Journal of the Academy of Marketing Science 1995;23(4):335–45. Zeithaml VA. Consumer perceptions of price, quality, and value: a means-end model and synthesis of evidence. Journal of Marketing 1988;52(3):2-22.

Please cite this article as: Hansen JD, et al, Antecedents and consequences of consumers' comparative value assessments across the relationship life cycle, J Bus Res (2011), doi:10.1016/j.jbusres.2011.11.006