Journal of Retailing and Consumer Services 26 (2015) 23–31

Contents lists available at ScienceDirect

Journal of Retailing and Consumer Services

journal homepage: www.elsevier.com/locate/jretconser

A two-dimensional model of trust–value–loyalty in service

relationships
Joe Choon Yean Chai a,n, Naresh K. Malhotra b, Frank Alpert c
a
Department of Marketing, University of Otago, Dunedin, New Zealand
b
Scheller College of Business, Georgia Institute of Technology, Atlanta, GA, USA
c
UQ Business School, University of Queensland, St. Lucia, Queensland, Australia

art ic l e i nf o a b s t r a c t

Article history: The study examines the relationships between consumer trust (cognitive and affective), perceived value
Received 7 December 2014 (utilitarian and hedonic), and loyalty behavioral intentions (repurchase and advocacy) in a two-dimen-
Received in revised form sional (affect- and rational-based) consumer–service provider model. It provides an alternative theore-
9 May 2015
tical representation of consumer behavior in service relationships and implications for service providers
Accepted 10 May 2015
about consumers' loyalty intentions. A mail survey analyzed by SEM supports the model of loyalty as a
function of consumers' expressive and instrumental responses resulting from intrinsic and extrinsic
Keywords: perceptions of trust and heuristic and rational-based perceived value. Implications for designing loyalty
Trust programs are explored and future research opportunities recommended.
Perceived value
& 2015 Elsevier Ltd. All rights reserved.
Loyalty behavioral intentions
Service relationships

1. Introduction stream in a very specific way, by disaggregating these highly ab-

stract variables into two dimensions, their cognitive and affective
Service providers have generally exhausted the possibilities of components, and creating a dual pathway model. The present
differentiation based on price, convenience or quality, tried cus- study fills this gap and develops a two-dimensional application of
tomization and changed image or ambiance to provide special consumer trust (cognitive and affective), perceived value (utili-
benefits to their customers. Such strategies work well in the short tarian and hedonic), and loyalty behavioral intentions (repurchase
term but are easily imitated by competitors. Most loyalty studies and advocacy) in a consumer–service provider model that has not
indicate that satisfaction alone is no longer sufficient to bond previously been attempted in the financial services context. The
customers because satisfied customers still switch to competition; research is also important from a practical perspective, as man-
nonetheless the perceived value is deemed to drive loyalty (Hu agers may be confused by the lack of a clear, common-sense and
et al., 2009; Neal, 1999). Thus, a competitive advantage to prevent practical causal flow for understanding and improving behavioral
customer defection and enhance loyalty can be built around cus- loyalty in terms of the key variables that we will address.
tomer perceived value attributes rather than satisfaction or even
service quality per se (Chen, 2015; Parasuraman, 1997; Woodruff,
1997). Although previous customer value studies have examined 2. Theoretical background
the antecedents and consequences of perceived value (Jensen,
2001) and its associations with trust and loyalty outcomes (Jones 2.1. Trust: cognitive and affective
et al., 2006; Sirdeshmukh et al., 2002; Taylor et al., 2014), no
empirical study has investigated the impacts of consumer trust The literature shows that trust is a powerful predictor of loyalty
and perceived value on loyalty intentions in an affect- and ra- in service relationships (Ball et al., 2004; Singh and Sirdeshmukh,
tional-based view of consumer–service provider model. Prior re- 2000). Trust is “a willingness to rely on an exchange partner in
search has largely used highly abstract variables (e.g., trust, per- whom one has confidence” (Moorman et al., 1993, p. 82). Trust is
ceived value). The present study seeks to deepen the research commonly studied in a two-dimensional view based on a rational
evaluation process and an emotional response raised from the
n
interactions (Johnson and Grayson, 2005; Parayitam and Dooley,
Corresponding author.
2009). The first dimension of trust is known as cognitive trust,
E-mail addresses: joe.chai@otago.ac.nz (J.C.Y. Chai),
naresh.malhotra@scheller.gatech.edu (N.K. Malhotra), cognition-based trust, knowledge-based trust, or system trust
f.alpert@business.uq.edu.au (F. Alpert). (Fukuyama, 1995; Lewicki and Stevenson, 1997; McAllister, 1995).

http://dx.doi.org/10.1016/j.jretconser.2015.05.005
0969-6989/& 2015 Elsevier Ltd. All rights reserved.
24 J.C.Y. Chai et al. / Journal of Retailing and Consumer Services 26 (2015) 23–31

This type of perceived trustworthiness is objective in nature and is 2.3. Loyalty: behavioral intention
based on “a rational process which determines whether the other
party in the relationship can be trusted” (Zur et al., 2012, p. 74). In The importance of behavioral intentions as predictors of cus-
other words, cognitive trust is related to perceived trustworthi- tomer loyalty is well recognized in many service provider and
ness, which is based on the service provider’s expertise and per- multi-services contexts that cover high and low contact, and ex-
formance, such as competence, credential, and reliability (Johnson perience and credence services (Patterson, 2004), such as mobile
and Grayson, 2005). banking (Luarn and Lin, 2005) and financial consulting (Guenzi
The second dimension of trust is known as affect-based trust, and Georges, 2010). Zeithaml et al. (1996) suggest that behavioral
emotional trust, interpersonal trust, or relational trust (Guenzi and intentions are useful dependent constructs in measures of rela-
Georges, 2010; Lewis and Weigert, 2012; Rousseau et al., 1998). tional and loyalty behavior because they are more closely related
This type of perceived trustworthiness is subjective in nature and to actual behavior than overall service quality or customer sa-
tisfaction constructs.
is based on “the feelings, emotions, and moods of the other” (Zur
Behavioral intentions can be grouped into two categories, as
et al., 2012, p. 75). Simply, affective trust is related to emotionality,
either social or economic behavior (Guenzi and Georges, 2010).
which is raised by the service provider's likeability, such as cour-
Advocacy intention (word of mouth recommendation intention) is
teousness, friendliness, and pleasantness (Nicholson et al., 2001).
widely recognized as a social measure of loyalty (Jones and Taylor,
2007). One important way this can arise in a service context is that
2.2. Perceived value: utilitarian and hedonic a consumer committed to a social relationship with a service
employee will engage in advocacy because of the social benefits
Customer value is widely recognized and defined from either a received from this relationship. The underlying logic of this re-
consumer's perspective or an organization's perspective (Land- ciprocity can be explained by social exchange theory (Blau, 1964),
roguez et al., 2013). Consumer perceived value, the central focus of which suggests an individual endorses reciprocity in return for an
this study, is a customer’s judgment or a valuation based on a obligation (Cropanzano and Mitchell, 2005). Customers who feel
comparison of the perceived benefits and costs received from a comfortable with their service provider relationship may act as
product, service or relationship (Overby and Lee, 2006). Perceived advocates for them, and these recommendations can be influential
value can predict consumer loyalty, influence purchase intention to new customers' decision making, particularly for services,
and prevent consumer switching behavior (Anderson et al., 2014; which are inherently intangible. Research indicates that social ties
Chiu et al., 2014). The consumer perceived value can result from and word-of-mouth referral behavior are related (Lewis and
Weigert, 2012). When customers feel trapped and dependent on
the personal comparison of the benefits gained and the sacrifices
their partners, they are less likely to advocate on behalf of the
made (Overby and Lee, 2006). Research identifies two approaches
partners (Harrison-Walker, 2001). Thus, advocacy intention can be
to the conceptualization and dimensionality of perceived value.
regarded as an expressive response of loyalty which is motivated
(1) Perceived value as a construct configured by benefits received,
by emotional factors or perceived social benefits (Jones et al.,
e.g., social, psychological, economic and customization benefits
2008). Roy's (2013) study also found customer advocacy has direct
(Gwinner et al., 1998), and sacrifices made by the customer, e.g.,
positive impact on customers' behavioral loyalty and positive
price, time, effort and convenience (Cronin et al., 2000; Overby word-of-mouth. In sum, advocacy intention or recommend in-
and Lee, 2006). (2) The conceptualization of perceived value as a tention is the likelihood of recommending a service provider to
construct that incorporates a functional and an affective dimen- others in the future (Wang, 2009).
sion (Roig et al., 2009; Sánchez-Fernández and Iniesta-Bonillo, Anderson and Sullivan's (1993) study indicates that satisfaction
2009). The functional or utilitarian dimension is defined by the is a key determinant of consumers' repurchase intentions and a
rational and economic valuations of individuals. On the other significant association exists between consumer overall satisfac-
hand, the affective or hedonic dimension is defined by emotional tion and intention to repurchase across product categories.
and social valuations of individuals (Boksberger and Melsen, 2011; Woodside et al. (1989) also find that patients tend to choose the
Voss et al., 2003). same hospital again when satisfied with their care and a customer
Utilitarian value is “an overall assessment (i.e., judgment) of evaluation about service experience demonstrated the importance
functional benefits and sacrifices” (Overby and Lee, 2006, p. 1161), of service encounters in the service delivery process. This implies
which is based on the assumption that consumers are rationale. that a consumer may be committed to the service provider be-
Rintamäki et al. (2006) suggest that monetary savings and con- cause of his or her overall satisfaction with the service provided by
venience contribute to utilitarian value. Cronin and Taylor (1992) the service personnel. A consumer committed to an economic
suggest that price, convenience, and availability of product affect exchange relationship with a service employee will continue to
consumer behavioral intentions. Utilitarian value is characterized purchase these services because of the economic benefits received
as extrinsic and instrumental (Chandon et al., 2000) because from this relationship. This is common sense, but the underlying
consumers experience utilitarian value when their functional or logic of this behavior can also be explained in terms of rational
task-related needs are fulfilled. choice theory, which suggests that an individual’s purchase deci-
As opposed to utilitarian value, hedonic value is abstract and sion is based mainly on the economic benefits that he or she can
subjective (Babin et al., 1994). Hedonic value is “an overall as- gain from the exchange (Scott, 2000). Thus, repurchase intention
can be regarded as an instrumental response of loyalty which is
sessment (i.e., judgment) of experiential benefits and sacrifices”
motivated by self-interest and economic factors (Jones et al.,
(Overby and Lee, 2006, p. 1161). Hirschman and Holbrook (1982)
2008). In sum, repurchase intention is an individual's judgment
suggest that hedonic value consists of the experiential view of
about purchasing a designated service from the same company
three F’s: fun, feelings, and fantasies. It can be argued that hedonic
again, taking into account his or her current situation and likely
consumption is related to an individual’s experience of the mul-
circumstances (Hellier et al., 2003).
tisensory, fantasy, and emotive aspects of services or products.
Based on the literature of value, consumer perceived value toward 3. Conceptual model and hypotheses
the service provider can be defined as both intrinsic and extrinsic
to the offering of services discussed in this study. Building on previous research on consumer trust, perceived
 J.C.Y. Chai et al. / Journal of Retailing and Consumer Services 26 (2015) 23–31 25

Fig. 1. A two-dimensional model of trust–value–loyalty.

value, and loyalty behavioral intentions, this study suggests that Grayson, 2005, p. 501). As emotional connections deepen, trust
consumer cognitive and affective trust and hedonic and utilitarian in a partner may venture beyond what available knowledge jus-
values are loyalty factors to advocacy and repurchase intentions. tifies. Such emotional exchanges are capable of eliciting a bond of
Consumer trust raised by a service provider's expertise or like- trust in and commitment to the service provider, leading to the
ability affects consumers' extrinsic or intrinsic value perception development of an interpersonal relationship or commercial
and in turn influences consumers' loyalty instrumental or emo- relationship (Grayson, 2007; Price and Arnould, 1999). A service
tional reaction. Based on the review of the theoretical ground and employee's social skills and likeability has a positive association
supporting evidence, a conceptual framework of a two-dimen- with a customer’s interpersonal trust (Guenzi and Georges, 2010;
sional model of trust–value–loyalty is developed (see Fig. 1). Nicholson et al., 2001). Perceptions of the service provider's
The next section develops the model step by step. This model is interpersonal trust closely relate to a customer's affective trust,
a general model but is designed in particular for the industry such that affective trust is an antecedent of perceived hedonic
context it will be tested in, the banking industry. Services are a value.
very broad sector of the economy, encompassing many industries,
so we chose as our focal industry the banking industry. The set of H2. : affective trust has a positive impact on a consumer's per-
variables that we have discussed are important to understand ceived hedonic value to the service provider.
theoretically and also are key variables for this industry. Trust and
Research shows that customer value influences consumer re-
consumer word of mouth are especially important in the banking
purchase intentions (Chiu et al., 2014; Olaru et al., 2008). The
industry.
present study suggests that utilitarian value (instrumental-based)
Cognitive trust is a customer's confidence in or willingness to
is an antecedent of a consumer's repurchase intention (an in-
have faith in a service provider's competence and reliability
strumental response) because of exchange economic benefits re-
(Moorman et al., 1993). This trust arises from “an accumulated
ceived by the consumer. Utilitarian value is characterized as in-
knowledge that allows one to make predictions, with some level of
strumental and extrinsic and provides functional or economic
confidence, regarding the likelihood that a focal partner will live
benefits or perceived value (Babin et al., 1994). Since repurchase
up to his/her obligations” (Johnson and Grayson, 2005, p. 501). A
intention is an instrumental response of loyalty outcome (Jones
customer's perception of a salesperson's expertise reflects the
et al., 2008), it can be argued that utilitarian value is a predictor of
identification of relevant competencies associated with a parti-
cular transaction (Crosby et al., 1990). Expertise is assessed by a economic behavioral intention of repurchase.
service provider's level of knowledge and experience concerning H3. : utilitarian value has a positive influence on a consumer's
the focal service (Johnson and Grayson, 2005). Assessments of repurchase intention.
service personnel's expertise and cognitive trust employ an attri-
bute evaluation process involving the specific identifiable actions Previous research reveals that perceived value has a positive
of the service provider. This notion is consistent with research effect on consumer recommend intentions (Hartline and Jones,
showing that a person's perceived level of expertise enhances his 1996). The present study positions hedonic value (affect-based) as
or her source credibility and thereby trustworthiness. The per- an antecedent of a consumer’s advocacy intention (an expressive
ceived service provider's expertise relates closely to a customer's response) because of exchange social benefits received by the
cognitive trust; thus, cognitive trust is an antecedent of perceived consumer. Hedonic value is characterized as emotional and in-
utilitarian value. trinsic and provides social benefits or perceived value (Babin et al.,
1994). Hedonic value is an outcome related to more subjective
H1. : cognitive trust has a positive impact on a consumer's per-
and personal spontaneous responses. It emphasizes entertain-
ceived utilitarian value to the service provider.
ment, exploration, and self-expression derived from fun and en-
Affective trust is the confidence a person places in a partner joyment rather than from task completion and it is affective, ex-
based on the level of care and concern that partner demonstrates periential, and non-instrumental in nature. Since advocacy inten-
(Johnson-George and Swap, 1982). Characteristics of affective trust tion is an expressive response of loyalty outcome, motivated in
include “feelings of security and perceived strength of the re- part by strong affect toward the service, it can be argued that
lationship” and is “decidedly more confined to personal experi- hedonic value is a predictor of social behavioral intention of
ences with the focal partner than cognitive trust” (Johnson and advocacy.
26 J.C.Y. Chai et al. / Journal of Retailing and Consumer Services 26 (2015) 23–31

H4. : hedonic value has a positive influence on a consumer’s ad- size of 300 cases to ensure appropriate use of maximum likelihood
vocacy intention. (ML) estimation in structural equation modeling (SEM) analysis, to
generate valid model fit measures, and to avoid drawing in-
Research indicates that the cognition-based component of at-
accurate inferences in factor analysis (Norušis, 2006; Tabachnick
titude comprises beliefs, thoughts, and judgments associated with
and Fidell, 2007).
an attitude object, while the affect-based component of attitude
The surveyed respondents were heterogeneous with a wide
consists of feelings, emotions, and drives with an attitude object
variety of characteristics represented throughout the population in
(Chiu et al., 2005). Edwards (1990) suggests that an individual's
New Zealand. The percentage of female participants (56.52%) is
affective component of attitude results from cognition. Ajzen and
higher than male participants (43.48%). This is common in mail
Fishbein's (1980) theory of reasoned action (TRA) suggests that
surveys, which are dominated by female participants (Hair et al.,
cognition is an antecedent factor of affective response. Chiu et al.'s
2006). The group aged 40–49 years (30.44%) is the largest group
(2005) bank study finds that extrinsic utilitarian value is a pre-
among surveyed respondents. More than half of the respondents
dictor of intrinsic hedonic value. Repurchase intention is related to
are married (62.23%). Overall, the sample had a high level of
instrumental, economic, and calculative-based response, whereas
educational background. More than half of the respondents
advocacy intention is primarily experiential, personal, and sub-
(53.53%) have a bachelor's degree, about 18% have a master’s de-
jective (Jones et al., 2008). Olaru et al.'s (2008) findings in the
gree, and 11% have a professional qualification. In terms of annual
research and development industry reveal that customer value is a
income, half of the respondents (50%) earned less than NZ$50,000
key determinant of repurchase and recommendation intentions,
and less than half of the respondents (45%) earned more than NZ
while customers' willingness to recommend to others is a result of
$50,000 at the time of the survey. Non-response bias in the sample
their repurchase intention. These findings support links between
was not evident based on an Armstrong and Overton (1977) test.
utilitarian and hedonic value, and between repurchase and ad-
Preliminary examination of the data resulted in 25 respondents
vocacy intention, in service relationships.
removed due to missing data, and statistical assumptions of uni-
H5. : utilitarian value has a positive impact on hedonic value. variate and multivariate normality of the data were confirmed.
Data factorability (Tabachnick and Fidell, 2007) was confirmed
H6. : repurchase intention has a positive impact on advocacy using the Kaiser–Meyer–Olkin Measure of Sampling Adequacy and
intention. Bartlett's Test of Sphericity (Pallant, 2005).

4.2. Measures

4. Methodology The items in measurement were drawn from previously tested

and validated scales with modified wordings to best describe the
4.1. Sample and data collection method banking environment (see Table 1). Trust factors were measured as
cognitive and affective drawing on scales from Johnson and
The banking industry is a good test case for examining the Grayson (2005). Perceived intrinsic and extrinsic values were
drivers of key loyalty types. Contextualizing the research allows for measured as hedonic and utilitarian drawing on scales from Chiu
more precise and concrete questions to consumers. The study et al. (2005) and Liu and Wu (2007). Loyalty behavioral intentions
employs two survey stages. The first pilot survey applied a com- were measured in terms of a consumer's repurchase and advocacy
bination of qualitative and quantitative approaches: (i) a pre-test intentions drawing on scales from Jones et al. (2008). The final
with a panel of experts for content validity, (ii) a pre-test with questionnaire was limited to the key variables and kept short to
banking practitioners to check the relevance and applicability of try to minimize any discouragement to complete the ques-
questions that suit the bank environment, and (iii) a pilot survey tionnaire due to length. All scales applied a seven-point Likert
with a convenience sample of 150 residents living in Dunedin city, Scale with anchors of strongly disagree (1) and strongly agree (7).
New Zealand for construct validity. The second survey randomly
distributed 2000 mailings to New Zealand residents nationwide,
which were drawn from the database of Electoral Rolls of New 5. Analysis and results
Zealand. The advantage of this self-administered method was that
a broader sample could potentially be achieved. The survey Common method variance (CMV) bias can be a potential threat
questionnaire was mailed with a cover letter, an information sheet, to the validity of survey research especially when all self-report
a consent letter, and a return postage paid envelope that allows measures are obtained in a single questionnaire. The data was
respondents to complete the survey questionnaire in their free tested for CMV bias using the methodology proposed by Lindell
time and to send their reply back in the paid envelope. In addition, and Whitney (2001) and further refined by Malhotra et al. (2006)
respondents have an option to fill in their answers online. Thus, and Schaller et al. (2015). All of the observed correlations that
respondents preferring online response could do so, and re- were originally significant remained significant after correcting for
spondents preferring a written and mailed response could do so. CMV as proposed by Lindell and Whitney (2001). These results
The result was 278 mail completed surveys were received and 115 indicate that CMV was not a bias in this study.
online surveys were captured. Mail response preference compared Test statistics support the fit of the model to the data. The
to online response preference was 2.41 to 1. Responses not com- overall fit test result for χ2 (128) ¼ 448.34 was statistically sig-
pleted or invalid questionnaire for mail is 3 and online is 22. Thus, nificant at p value ¼ 0.000. Standard fit indices for the model all fell
the invalid rate of mail response is lower than online response. A inside their recommended thresholds: CFI ¼ 0.95, IFI¼ 0.95,
total net mailed-out of 1839 was valid after deletion of un- NFI ¼0.93, TLI¼0.94, and RMSEA ¼0.08. A confirmatory factor
delivered or returned, refusal and invalid answers. The final valid analysis (CFA) measurement model was used to assess construct
sample was 368, for a 20% response rate (following the formula of reliability and validity (Anderson and Gerbing, 1988). Factor
Brennan (2004)). loadings for the CFA measurement model ranged from 0.63 to
While the response rate is only 20%, this is not uncommon for 0.95, which are above the acceptable level of 0.50. All critical ratio
mail surveys (Fink, 2003). Nevertheless, the sample size of 368 (CR) values were greater than the acceptable level of 1.96, with
cases in this mail survey met the minimum requirement sample po 0.001. Cronbach alphas ranged from 0.83 to 0.95, exceeding the
 J.C.Y. Chai et al. / Journal of Retailing and Consumer Services 26 (2015) 23–31 27

Table 1
Measurement items and descriptive statistics.

Factor Indicator Statement Mean Standard deviation

Cognitive trust (CT) CT1 Given my banking consultant's track record, I have good reason to trust his/her competence. 4.77 1.29
CT2 Given my banking consultant's track record, I have no reservations about acting on his/her advice. 4.36 1.36
CT3 I can rely on my banking consultant to undertake a thorough analysis of the situation before advising 4.74 1.25
me.
Affective trust (AT) AT1 My banking consultant displays a warm and caring attitude towards me. 5.19 1.17
AT2 If I share my problems with my banking consultant, I feel he/she would respond caringly. 4.93 1.26
AT3 I would feel a sense of personal loss if I could no longer use my banking consultant's service. 4.34 1.43
Utilitarian value (UV) UV1 This bank satisfies the majority of my financial needs. 5.75 1.09
UV2 I feel this bank is convenient. 5.78 1.09
UV3 I feel this bank is efficient. 5.45 1.20
Hedonic value HV1 The time I spend in this bank is enjoyable 5.32 1.11
(HV) HV2 I chose this bank not because I had to, but because I wanted to. 5.39 1.39
HV3 I feel that I made the right decision by choosing this bank. 5.54 1.14
Repurchase intention RP1 I will probably use this bank again. 5.25 1.13
(RP) RP2 I intend to purchase services from this bank again in the future. 5.19 1.14
RP3 It is possible that I will use this bank in the future. 5.34 1.17
Advocacy intention (AD) AD1 I will say positive things about this bank to other people. 5.08 1.25
AD2 I will recommend this bank to other people who ask my advice. 5.10 1.30
AD3 I will encourage friends and relatives to do business with this bank. 4.89 1.32

0.70 threshold. Average variance extracted (AVE) values ranged We examined multicollinearity and tested discriminant validity
from 0.76 to 0.89, exceeding the critical level of 0.50. Thus, com- for all measures. The results of regression analyses of the 18
posite reliability and convergent validity are supported. variables included in the model indicate that all variance inflation

Table 2
Results of construct reliability, convergent validity, model fit, assessments of CFA measurement model and structural model.

Construct and path Alpha coefficient (α) Factor loading (β) CR p-value Influence direction Hypothesis result

Reliability
Cognitive trust 0.89
Affective trust 0.85
Utilitarian value 0.90
Hedonic value 0.83
Repurchase intention 0.94
Advocacy intention 0.95
CFA measurement model
CT1 ’ Cognitive trust (CT) 0.89
CT2 ’ Cognitive trust (CT) 0.82 19.85 0.000
CT3 ’ Cognitive trust (CT) 0.85 21.03 0.000
AT1 ’ Affective trust (AT) 0.93
AT2 ’ Affective trust (AT) 0.93 27.40 0.000
AT3 ’ Affective trust (AT) 0.63 14.22 0.000
UV1 ’ Utilitarian value (UV) 0.86
UV2 ’ Utilitarian value (UV) 0.87 21.38 0.000
UV3 ’ Utilitarian value (UV) 0.87 21.49 0.000
HV1 ’ Hedonic value (HV) 0.79
HV2 ’ Hedonic value (HV) 0.68 13.46 0.000
HV3 ’ Hedonic value (HV) 0.89 18.31
RP1 ’ Repurchase intention (RP) 0.92
RP2 ’ Repurchase intention (RP) 0.90 28.02 0.000
RP3 ’ Repurchase intention (RP) 0.93 30.54 0.000
AD1 ’ Advocacy intention (AD) 0.95
AD2 ’ Advocacy intention (AD) 0.94 36.38 0.000
AD3 ’ Advocacy intention (AD) 0.91 31.96 0.000
CFA structural model
CT-UV 0.54 9.99 0.000 Positive Supported
AT-HV 0.32 7.40 0.000 Positive Supported
UV-RP 0.65 12.94 0.000 Positive Supported
HV-AD 0.18 3.70 0.000 Positive Supported
UV-HV 0.69 13.04 0.000 Positive Supported
RP-AD 0.66 13.46 0.000 Positive Supported
Model fit measurement Acceptable level
Chi-square (χ2) 448.34 p 40.05 (at α equals to 0.05 level)
Degree of freedom (df) 128
Probability level 0.00
CFI 0.95 Z 0.90
IFI 0.95 Z 0.90
NFI 0.93 Z 0.90
TLI 0.94 Z 0.90
RMSEA 0.08 r 0.80

Note : CFI ¼comparative-fit index; IFI¼ incremental-fit index; NFI ¼normed-fit index; TLI¼ Tucker–Lewis index; RMSEA ¼root mean square error of approximation.
28 J.C.Y. Chai et al. / Journal of Retailing and Consumer Services 26 (2015) 23–31

Table 3
Average variance extracted (AVE) analysis for discriminant validity.

Construct AVE Cognitive trust Affective trust Utilitarian value Hedonic value Repurchase intention Advocacy intention

Cognitive trust 0.862 0.928

Affective trust 0.785 0.685 0.886
Utilitarian value 0.892 0.434 0.410 0.944
Hedonic value 0.762 0.488 0.567 0.687 0.873
Repurchase intention 0.863 0.529 0.589 0.571 0.611 0.929
Advocacy intention 0.850 0.467 0.523 0.462 0.552 0.727 0.922

Note: Square root of the average variance extracted compared with the correlations of constructs.
Bold numbers on the diagonal are the square root of the AVE (SQRT AVE) values.

factor (VIF) values ranged from 1.677 to 7.745, which are lower Table 4
than the threshold value of 10 (Aiken and West, 1991; Belsley et al., The result of competing structural models.
1980). This indicates absence of multicollinearity issues in the
Path Proposed model Rival model
data. In addition, the square root of the AVE (SQRT AVE) values
(ranged from 0.87 to 0.94) exceed the inter-construct correlation Factor CR p-value Factor CR p-value
coefficients (Fornell and Larcker, 1981; Kline, 2005), which in- loading loading
dicates discriminant validity for the measures. The results for (β) (β)

construct reliability, convergent validity, model fit, and assess- CT-UV 0.54 9.99 0.000
ments of CFA measurement model and structural model are pre- AT-HV 0.32 7.40 0.000
sented in Table 2, and AVE analysis for discriminant validity is UV-HV 0.69 13.04 0.000
presented in Table 3. RP-AD 0.66 13.46 0.000
UV-RP 0.65 12.94 0.000 0.20 4.27 0.000
HV-AD 0.18 3.70 0.000 0.50 9.17 0.000
5.1. Comparison of overall fit to an alternative model UV-AD 0.50 1.04 0.299
HV-RP 0.48 8.93 0.000
Several studies suggest that as part of testing the proposed CT-AD 0.17 3.57 0.000
model, researchers should include testing it versus an alternative CT-RP 0.20 4.20 0.000
AT-AD 0.22 4.60 0.000
model to further establish its efficacy (Bollen and Long, 1992; AT-RP 0.30 6.25 0.000
Iwasaki and Havitz, 2004; Morgan and Hunt, 1994). An alternative Model fit
model was developed following the alternative model testing measurement
procedure of Morgan and Hunt (1994). As in their classic article, Chi-square (χ2) 448.34 1074.26
Degree of free- 128 127
the theorized model was compared to a model of direct paths to
dom (df)
each dependent variable from all the other variables. That is, paths Probability level 0.00 0.00
determined by the proposed theory, with its mediating flow, are CFI 0.95 0.84
tested for better fit versus a model of “no theory” in the sense of IFI 0.95 0.84
paths not constrained by theory. NFI 0.93 0.83
TLI 0.94 0.81
The comparison shows (Table 4) that the theorized model's PNFI 0.78 0.69
goodness-of-fit indices (CFI ¼0.95, IFI¼0.95, NFI ¼0.93, TLI¼ 0.94), RMSEA 0.08 0.14
which were above common benchmarks of 0.90, are all higher
than the alternative, direct effects model’s goodness-of-fit indices CFI¼ comparative-fit index; IFI ¼incremental-fit index; NFI¼ normed-fit index;
TLI¼ Tucker–Lewis index; PNFI¼ parsimony normed fit index; RMSEA ¼root mean
(CFI¼ 0.84, IFI¼ 0.84, NFI ¼0.83, TLI ¼0.81), which were below
square error of approximation.
common benchmarks of 0.90. Similarly, the theorized model's
RMSEA (0.08) indicates a good fit, whereas the alternative model's
RMSEA (0.14) indicates a poor fit. In addition, the theorized
is supported (β ¼0.18, CR ¼3.70, p o0.001). On the relationship
model's PNFI (0.78) is higher than the rival model's PNFI (0.69). In
between cognitive and affective components of value perceptions,
sum, test statistics show the theorized model is a fit to the data,
hypothesis 5 predicted that the affect-based hedonic value is in-
and a better fit than the alternative model.
fluenced by the cognition-based utilitarian value. On the re-
lationship between economic and social behavioral intentions,
5.2. Hypothesis tests
hypothesis 6 predicted that the affect-based advocacy intention is
Hypothesis 1 predicted that cognitive trust in the service pro- influenced by the instrumental-based repurchase intention. The
vider would positively influence a consumer's perceived utilitarian result showed that the influence of utilitarian value on hedonic
value, whereas hypothesis 2 predicted that affective trust in the value (β ¼0.69, CR ¼13.04, p o0.001), and repurchase intention on
service provider would positively influence a consumer's per- advocacy intention (β ¼0.66, CR ¼13.46, p o0.001) were in the
ceived hedonic value. The results showed that the influence of hypothesized direction and the paths were statistically significant;
cognitive trust on utilitarian value (β ¼0.54, CR ¼9.99, p o0.001) therefore, H5 and H6 were supported.
and affective trust on hedonic value (β ¼0.32, CR¼ 7.40, p o0.001) In sum, the findings support that the overall model of trust–
were in the hypothesized direction and the paths were statistically value–loyalty in banking relationship is valid and that each of the
significant; thus, H1 and H2 were supported. Hypothesis 3, which predicted relationships among these variables are statistically
predicted that perceived utilitarian value resulting from cognitive significant in hypothesized positive directions (see Fig. 2). The
trust would positively influence a consumer's repurchase inten- findings demonstrate the importance of (i) cognitive trust on uti-
tion, is supported (β ¼0.65, CR ¼12.94, p o0.001). Hypothesis 4, litarian value that affects a consumer's repurchase intention, and
which predicted perceived hedonic value resulting from affective (ii) affective trust on hedonic value that affects a consumer's ad-
trust would positively influence a consumer’s advocacy intention, vocacy intention. The findings also confirmed that the hedonic
 J.C.Y. Chai et al. / Journal of Retailing and Consumer Services 26 (2015) 23–31 29

Fig. 2. The structural equation model of trust–value–loyalty.

value is post-cognition and advocacy intention is a consequence those. The findings suggest managers would think more richly
factor of repurchase intention. about the causes of loyalty intentions if they thought in terms of
cognitive and affective pathways (two dimensions, instead of one).
Furthermore, while the affective pathway should not be neglected,
6. Discussion and implications the cognitive pathway should be emphasized, as its effects are
stronger. Building cognitive trust and delivering utilitarian value
The findings provide both theoretical and practical implica- are demonstrated to be very important in achieving repurchase
tions. Previous studies investigate the interrelationships of trust, intentions, and also in increasing hedonic value and advocacy in-
value, loyalty in one-dimensional ways and the complexity of tention. Affective trust is also important, and not to be seen as
multi-dimensional variables is not captured. The findings suggest trivial. Trust-building strategies should be focused on consumers'
that consumers' loyalty behavioral intentions are essentially the trust in and friendship to the service provider. Hedonic value is
result of intrinsic and extrinsic perceptions of trust and perceived also demonstrated to have a significant effect, on advocacy in-
value of consumers. This provides a model that is both parsimo- tentions, albeit smaller.
nious, with three conceptual variables (trust, perceived value and
loyalty intentions), yet enriched (by dividing up each variable into
two components). The model captures both cognitive and affective 7. Conclusions, limitations and future research
pathways, and also integrates them. The model is strongly sup-
ported by overall fit to the banking industry data that was col- The research has several limitations and offers several avenues
lected, and by statistical significance for each of the hypothesized for future research. First, the study is limited by the banking ser-
relationships in the model. Furthermore, four of the six effects are vices context in which it takes place. Variables had to be included
relatively strong (i.e., β coefficients ranging from 0.54 to 0.69). that were key for this industry, such as cognitive trust, which may
Another finding is that cognitive effects appear stronger than af- be less important in other services. The more affective industries
fective effects. That is, the utilitarian value effect on hedonic value (including entertainment such as concerts) could have a stronger
is very strong (β ¼0.69), and much stronger than the affect trust affective pathway. Similarly, affective trust and hedonic value are
effect on hedonic value (β ¼0.32). Similarly the repurchase in- shown to be important in banking, but may not be so for other
tentional effect on advocacy intention is very strong (β ¼0.66) and services such as for plumbers. The generalizability issue does not
much stronger than the hedonic value effect on advocacy intention eliminate the contribution of the study, but, as with much of
(β ¼0.18). Generally speaking, of the two pathways, cognitive marketing, the model must be adapted to different contexts. The
pathway effects were stronger (βs 0.54, 0.65) than their respective banking industry itself is very large and important, and con-
affective pathway effects (βs 0.32, 0.18). This result may be char- textualizing the research allows for strong findings for this and
acteristic of the more utilitarian industries (such as banking). In similar industries. Further studies could examine different service
sum, researchers should think about loyalty behavioral intentions industry settings so that the generalizability of the findings could
in a more sophisticated way: two dimensional rather than one be assessed, and also suggestions for adapting the model be tested.
dimensional. Furthermore, this is another example to researchers We expect the model generalizes to other people-oriented services
of how understanding cognitive and affective paths adds deeper that share common characteristics with the banking industry, such
insights and a more realistic model. as the health care, consultancy, legal, and real estate sectors. In-
There are several managerial implications. It helps managers depth qualitative studies, such as the focus groups commonly
understand how the key outcome of loyalty behavioral intentions conducted in marketing practice, could investigate consumer
are formed, and therefore potentially how to improve achieving trust–value–loyalty relationships in other service-based industries
30 J.C.Y. Chai et al. / Journal of Retailing and Consumer Services 26 (2015) 23–31

and provide more confidence to transfer or adapt the results of the sales promotion effectiveness. J. Mark. 64 (4), 65–81.
model. Chen, S.-C., 2015. Customer value and customer loyalty: is competition a missing
link? J. Retail. Consum. Serv. 22, 107–116.
Second, this study, like many questionnaires, is limited to Chiu, C.M., Wang, E.T., Fang, Y.H., Huang, H.Y., 2014. Understanding customers' re-
consumer self-report data. Future studies could consider actual peat purchase intentions in B2C e‐commerce: the roles of utilitarian value,
behavioral measures to calibrate how the self-report data predicts hedonic value and perceived risk. Inf. Syst. J. 24 (1), 85–114.
Chiu, H.-C., Hsieh, Y.-C., Li, Y.-C., Lee, M., 2005. Relationship marketing and con-
actual behavior for this model. sumer switching behavior. J. Bus. Res. 58 (12), 1681–1689.
Third, the findings are also limited as based on cross-sectional Cronin, J.J., Taylor, S.A., 1992. Measuring service quality: a reexamination and ex-
data, whereas loyalty is a dynamic process; thus longitudinal re- tension. J. Mark. 56 (3), 55–68.
Cronin, J.J., Brady, M.K., Hult, G.T.M., 2000. Assessing the effects of quality, value,
search could be considered to fully understand how loyalty arises and customer satisfaction on consumer behavioral intentions in service en-
and evolves. vironments. J. Retail. 76 (2), 193–218.
Fourth, the study examines the banking relationship in the Cropanzano, R., Mitchell, M.S., 2005. Social exchange theory: an interdisciplinary
review. J. Manag. 31 (6), 874–900.
traditional banking environment. The traditional banking en-
Crosby, L.A., Evans, K.R., Cowles, D., 1990. Relationship quality in services selling: an
vironment is still very important, but e-banking is growing and interpersonal influence perspective. J. Mark. 54 (3), 68–81.
how the model works in the e-banking environment could be Edwards, K., 1990. The interplay of affect and cognition in attitude formation and
change. J. Personal. Soc. Psychol. 59 (2), 202–216.
assessed. Perhaps cognitively connected consumers are using the
Fink, A., 2003. The Survey Handbook. Sage, Thousand Oaks, CA.
e-banking channels more than those affectively connected to their Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with un-
banking service provider. observable variables and measurement error. J. Mark. Res. 18 (1), 39–50.
Fukuyama, F., 1995. Trust: The Social Virtues and the Creation of Prosperity. Free
Finally, the current conceptual framework concentrates on six
Press, New York, NY.
key variables and their pathways. For future research, propositions Grayson, K., 2007. Friendship versus business in marketing relationships. J. Mark.
related to the moderating effects of additional variables could also 71 (4), 121–139.
be addressed, for example the moderating role of culture value in Guenzi, P., Georges, L., 2010. Interpersonal trust in commercial relationships:
antecedents and consequences of customer trust in the salesperson. Eur. J.
the model that captures culture differences in emotion and ra- Mark. 44 (1/2), 114–138.
tionality. The influence of culture in consumption has become Gwinner, K.P., Gremler, D.D., Bitner, M.J., 1998. Relational benefits in service in-
ubiquitous in today's globalization markets, especially for poten- dustries: the customer's perspecti. J. Acad. Mark. Sci. 26 (2), 101–114.
Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E., Tatham, R.L., 2006. Multivariate
tially large cultural differences such as East vs West, thus the in- Data Analysis. Prentice Hall, Upper Saddle River, NJ.
clusion of the moderator role of culture value would add to the Harrison-Walker, L.J., 2001. The measurement of word-of-mouth communication
development of a more complete theoretical framework. and an investigation of service quality and customer commitment as potential
antecedents. J. Serv. Res. 4 (1), 60–75.
Notwithstanding these limitations, our research makes a Hartline, M.D., Jones, K.C., 1996. Employee performance cues in a hotel service
number of significant contributions. It examines a two dimen- environment: influence on perceived service quality, value, and word-of-mouth
sional view of consumer trust, perceived value, and loyalty beha- intentions. J. Bus. Res. 35 (3), 207–215.
Hellier, P.K., Geursen, G.M., Carr, R.A., Rickard, J.A., 2003. Customer repurchase in-
vioral intentions in a service relationships framework. Thus, we tention: a general structural equation model. Eur. J. Mark. 37 (11/12),
theoretically extend the previous research in this area. We derive 1762–1800.
specific hypotheses that are grounded in theory. Moreover, we test Hirschman, E.C., Holbrook, M.B., 1982. Hedonic consumption: emerging concepts,
methods and propositions. J. Mark. 46 (3), 92–101.
these hypotheses based on a national sample. We establish the Hu, H.-H., Kandampully, J., Juwaheer, T.D., 2009. Relationships and impacts of
measurement properties of all our constructs and test these hy- service quality, perceived value, customer satisfaction, and image: an empirical
potheses using appropriate structural equation modeling metho- study. Serv. Ind. J. 29 (2), 111–125.
Iwasaki, Y., Havitz, M.E., 2004. Examining relationships between leisure involve-
dology. All of our hypotheses are supported and have substantial
ment, psychological commitment and loyalty to a recreation agency. J. Leis. Res.
theoretical and managerial implications. We hope that our effort 36 (1), 45–72.
will inspire more research in this area. Jensen, H.R., 2001. Antecedents and consequences of consumer value assessments:
implications for marketing strategy and future research. J. Retail. Consum. Serv.
8 (6), 299–310.
Johnson, D., Grayson, K., 2005. Cognitive and affective trust in service relationships.
References J. Bus. Res. 58 (4), 500–507.
Johnson-George, C., Swap, W.C., 1982. Measurement of specific interpersonal trust:
construction and validation of a scale to assess trust in a specific other. J. Per-
Aiken, L.S., West, S.G., 1991. Multiple Regression: Testing and Interpreting Inter- sonal. Soc. Psychol. 43 (6), 1306–1317.
actions. Sage, Thousand Oaks, CA. Jones, M.A., Reynolds, K.E., Arnold, M.J., 2006. Hedonic and utilitarian shopping
Anderson, K.C., Knight, D.K., Pookulangara, S., Josiam, B., 2014. Influence of hedonic value: investigating differential effects on retail outcomes. J. Bus. Res. 59 (9),
and utilitarian motivations on retailer loyalty and purchase intention: a face- 974–981.
book perspective. J. Retail. Consum. Serv. 21 (5), 773–779. Jones, T., Taylor, S.F., 2007. The conceptual domain of service loyalty: how many
Ajzen, I., Fishbein, M., 1980. Understanding Attitudes and Predicting Social Beha- dimensions? J. Serv. Mark. 21 (1), 36–51.
vior. Prentice-Hall, Englewood Cliffs, NJ. Jones, T., Taylor, S.F., Bansal, H.S., 2008. Commitment to a friend, a service provider,
Anderson, E.W., Sullivan, M.W., 1993. The antecedents and consequences of cus- or a service company – are they distinctions worth making? J. Acad. Mark. Sci.
tomer satisfaction for firms. Mark. Sci. 12 (2), 125–143. 36 (4), 473–487.
Anderson, J.C., Gerbing, D.W., 1988. Structural equation modeling in practice: a Kline, R.B., 2005. Principles and Practice of Structural Equation Modeling. Guilford
review and recommended two-step approach. Psychol. Bull. 103 (3), 411–423. Press, New York, NY.
Armstrong, J.S., Overton, T.S., 1977. Estimating nonresponse bias in mail surveys. J. Landroguez, S.M., Castro, C.B., Cepeda-Carrión, G., 2013. Developing an integrated
Mark. Res. 14 (3), 396–402. vision of customer value. J. Serv. Mark. 27 (3), 234–244.
Babin, B.J., Darden, W.R., Griffin, M., 1994. Work and/or fun: measuring hedonic and Lewis, J.D., Weigert, A.J., 2012. The social dynamics of trust: theoretical and em-
utilitarian shopping value. J. Consum. Res. 20 (4), 644–656. pirical research, 1985-2012. Soc. Forces 91 (1), 25–31.
Ball, D., Coelho, P.S., Machás, A., 2004. The role of communication and trust in Lewicki, R.J., Stevenson, M.A., 1997. Trust development in negotiation: proposed
explaining customer loyalty. Eur. J. Mark. 38 (9/10), 1272–1287. actions and a research agenda. Bus. Prof. Ethics J. 16 (1/3), 99–132.
Belsley, D.A., Kuh, E., Welsch, R.E., 1980. Regression Diagnostics: Identifying Influ- Lindell, M.K., Whitney, D.J., 2001. Accounting for common method variance in
ential Data and Sources of Collinearity. John Wiley & Sons, New York, NY. cross-sectional research designs. J. Appl. Psychol. 86 (1), 114–121.
Blau, P.M., 1964. Exchange and Power in Social Life. John Wiley & Sons, New York, Liu, T.C., Wu, L.W., 2007. Customer retention and cross-buying in the banking in-
NY. dustry: an integration of service attributes, satisfaction and trust. J. Financ. Serv.
Boksberger, P.E., Melsen, L., 2011. Perceived value: a critical examination of defi- Mark. 12 (2), 132–145.
nitions,concepts and measures for the service industry. J. Serv. Mark. 25 (3), Luarn, P., Lin, H.-H., 2005. Toward an understanding of the behavioral intention to
229–240. use mobile banking. Comput. Hum. Beh. 21 (6), 873–891.
Bollen, K.A., Long, J.S., 1992. Tests for structural equation models: introduction. Malhotra, N.K., Kim, S.S., Patil, A., 2006. Common method variance in IS research: a
Sociol. Methods Res. 21 (2), 123–131. comparison of alternative approaches and a reanalysis of past research. Manag.
Brennan, M., 2004. A test of two procedures for increasing responses to mail survey. Sci. 52 (12), 1865–1883.
Mark. Bull. 15, 1–9. McAllister, D.J., 1995. Affect- and cognition-based trust as foundations for inter-
Chandon, P., Wansink, B., Laurent, G., 2000. A benefit congruency framework of personal cooperation in organizations. Acad. Manag. J. 38 (1), 24–59.
 J.C.Y. Chai et al. / Journal of Retailing and Consumer Services 26 (2015) 23–31 31

Moorman, C., Deshpandé, R., Zaltman, G., 1993. Factors affecting trust in market 260–276.
research relationships. J. Mark. 57 (1), 81–101. Sánchez-Fernández, R., Iniesta-Bonillo, M.Á., 2009. Efficiency and quality as eco-
Morgan, R.M., Hunt, S.D., 1994. The commitment-trust theory of relationship nomic dimensions of perceived value: conceptualization, measurement, and
marketing. J. Mark. 58 (3), 20–38. effect on satisfaction. J. Retail. Consum. Serv. 16 (6), 425–433.
Neal, W.D., 1999. Satisfaction is nice, but value drives loyalty. Mark. Res. 11 (1), Schaller, T.K., Patil, A., Malhotra, N.K., 2015. Alternative techniques for assessing
20–23. common method variance: an analysis of the theory of planned behavior re-
Nicholson, C.Y., Compeau, L.D., Sethi, R., 2001. The role of interpersonal liking in search. Organ. Res. Methods 18 (2), 177–206.
building trust in long-term channel relationships. J. Acad. Mark. Sci. 29 (1), Scott, J., 2000. Rational choice theory. In: Browning, G., Halcli, A., Webster, F. (Eds.),
3–15. Understanding Contemporary Society: Theories of the Present. Sage, London,
Norušis, M.J., 2006. SPSS 14.0 Guide to Data Analysis. Prentice Hall, Upper Saddle pp. 126–138.
River, NJ. Singh, J., Sirdeshmukh, D., 2000. Agency and trust mechanisms in consumer sa-
Olaru, D., Purchase, S., Peterson, N., 2008. From customer value to repurchase in- tisfaction and loyalty judgments. J. Acad. Mark. Sci. 28 (1), 150–167.
tentions and recommendations. J. Bus. Ind. Mark. 23 (8), 554–565. Sirdeshmukh, D., Singh, J., Sabol, B., 2002. Consumer trust, value, and loyalty in
Overby, J.W., Lee, E.-J., 2006. The effects of utilitarian and hedonic online shopping relational exchanges. J. Mark. 66 (1), 15–37.
value on consumer preference and intentions. J. Bus. Res. 59 (10/11), 1160–1166. Tabachnick, B.G., Fidell, L.S., 2007. Using Multivariate Statistics. Harper Collins, New
Pallant, J., 2005. SPSS Survival Manual. Open University Press, Berkshire, England. York, NY.
Parasuraman, A., 1997. Reflections on gaining competitive advantage through Taylor, S.A., Donovan, L.A.N., Ishida, C., 2014. Consumer trust and satisfaction in the
customer value. J. Acad. Mark. Sci. 25 (2), 154–161. formation of consumer loyalty intentions in transactional exchange: the case of
Parayitam, S., Dooley, R.S., 2009. The interplay between cognitive- and affective a mass discount retailer. J. Relatsh. Mark. 13 (2), 125–154.
conflict and cognition- and affect-based trust in influencing decision outcomes. Voss, K.E., Spangenberg, E.R., Grohmann, B., 2003. Measuring the hedonic and
J. Bus. Res. 62 (8), 789–796. utilitarian dimensions of consumer attitude. J. Mark. Res. 40 (3), 310–320.
Patterson, P.G., 2004. A contingency model of behavioural intentions in a services Wang, C.-Y., 2009. Investigating antecedents of consumers’ recommend intentions
context. Eur. J. Mark. 38 (9/10), 1304–1315. and the moderating effect of switching barriers. Serv. Ind. J. 29 (9), 1231–1241.
Price, L.L., Arnould, E.J., 1999. Commercial friendships: service provider–client re- Woodruff, R.B., 1997. Customer value: the next source for competitive advantage. J.
lationships in context. J. Mark. 63 (4), 38–56. Acad. Mark. Sci. 25 (2), 139–153.
Rintamäki, T., Kanto, A., Kuusela, H., Spence, M.T., 2006. Decomposing the value of Woodside, A.G., Frey, L.L., Daly, R.T., 1989. Linking service quality, customer sa-
department store shopping into utilitarian, hedonic and social dimensions: tisfaction, and behavioral intention. J. Healthc. Mark. 9 (4), 5–17.
evidence from Finland. Int. J. Retail Distrib. Manag. 34 (1), 6–24. Zeithaml, V.A., Berry, L.L., Parasuraman, A., 1996. The behavioral consequences of
Roig, J.C.F., García, J.S., Moliner Tena, M.A., 2009. Perceived value and customer service quality. J. Mark. 60 (2), 31–46.
loyalty in financial services. Serv. Ind. J. 29 (6), 775–789. Zur, A., Leckie, C., Webster, C.M., 2012. Cognitive and affective trust between Aus-
Rousseau, D.M., Sitkin, S.B., Burt, R.S., Camerer, C., 1998. Not so different after all: a tralian exporters and their overseas buyers. Australas. Mark. J. 20 (1), 73–79.
cross-discipline view of trust. Acad. Manag. Rev. 23 (3), 393–404.
Roy, S.K., 2013. Consequences of customer advocacy. J. Strateg. Mark. 21 (3),