The Emerald Research Register for this journal is available at The current issue and full text archive

ve of this journal is available at

www.emeraldinsight.com/researchregister www.emeraldinsight.com/0309-0566.htm

A model of
A model of customer loyalty in customer loyalty
the retail banking market
Asunción Beerli, Josefa D. Martı́n and Agustı́n Quintana
University of Las Palmas de Gran Canaria, 253
Facultad Ciencias Económicas y Empresariales,
Las Palmas de Gran Canaria, Canary Islands, Spain Received March 2001
Revised July 2001,
January 2002
Keywords Customer loyalty, Customer satisfaction, Costs, Quality
Abstract On the basis of empirical research carried out in the retail banking market, this paper
proposes a structural equations model enabling us to reach the conclusions that satisfaction
together with personal switching costs are antecedents leading directly to customer loyalty, with the
former exerting the greatest influence; and perceived quality is a consequence of satisfaction. At
the same time, the paper shows that the degree of elaboration in the bank selection process does not
have a moderating influence on the causal relationships between satisfaction/switching costs and
customer loyalty.

Introduction
In the highly competitive, complex and dynamic environment of the banking
industry, the very slight differences which exist in financial services and
products together with an increasingly demanding customer have led to a great
transformation in the industry. The traditional product-oriented bank is
becoming increasingly customer-oriented in accordance with the basic
principles of relational marketing, which focuses on customer loyalty as its
main goal. In this sense, Gilmore (1997) considers that constant
customer-oriented behaviour is a requisite for improving the implementation
of quality in services marketing. Indeed, factors such as financial products and
distribution have attained similar levels of development and technology and
have thus been relegated to a secondary role as reference points for
distinguishing between one bank and another (Rodrı́guez Sánchez and
Rodrı́guez Parada, 1993). In this sense, Barnes and Howlett (1998) argue that,
given that many financial services are parity offerings, it can be stated that a
customer is unlikely to be overly impressed by core product attributes when all
companies are providing similar offerings.
The main objective of this paper is to identify the key factors that influence
the extent to which customers are loyal towards their banks. The customers of
market leaders in retail banking will form the subject of our research. To this
end, we have carried out empirical analyses of the complex interdependence European Journal of Marketing
relationships that exist among the different variables, or factors, which explain Vol. 38 No. 1/2, 2004
pp. 253-275
customer loyalty in the retail banking market, using a methodology based on q Emerald Group Publishing Limited
0309-0566
structural equation models. DOI 10.1108/03090560410511221
EJM Prior to developing the model, we carried out a review of the most important
38,1/2 contributions to academic literature about causal relationships between loyalty
and its antecedents. We reached the conclusion that customer satisfaction
together with barriers or the switching costs were the key factors affecting
loyalty. Similarly, there is a theoretical framework which considers that
perceived quality is one of the indirect antecedents of loyalty that has a direct
254 influence on customer satisfaction, although the nature of the causal
relationship between quality and customer satisfaction is the subject of great
academic debate. While quality researchers claim that satisfaction is an
antecedent of quality (Carman, 1990; Parasuraman et al., 1988; Bitner, 1990),
researchers in the field of customer satisfaction have exactly the opposite point
of view (Woodside et al., 1989; Reidenbach and Sandifer-Smallwood, 1990;
Cronin and Taylor, 1992; Fornell, 1992; Anderson and Sullivan, 1993). In the
light of this, several authors stress the desirability of further comparison of the
causal relationships between these two concepts, all of which has led us to
explore, as an additional objective of this paper, the direction of this
relationship with new empirical evidence.
Finally, and with a view to examining more closely the relationships which
exist between customer loyalty and its antecedents, we decided to analyse the
possible influence on such relationships that the degree of elaboration in the
bank selection process could have.

Theoretical background
The increasing importance of relational marketing in recent years, particularly
in the servicing and manufacturing industries, has been accompanied by a
bundle of works on customer loyalty. Several authors emphasize the positive
relationship existing between customer loyalty and business performance
(Reichheld and Sasser, 1990; Reichheld, 1993; Sheth and Parvatiyar, 1995).
Loyal customers not only increase the value of the business, but they also
enable it to maintain costs lower than those associated with attracting new
customers (Barroso Castro and Martı́n Armario, 1999).
Generally, loyalty has been, and continues to be, defined as repeat
purchasing frequency or relative volume of same-brand purchasing. Many
definitions in the literature suffer from the problem that they record what the
consumer does, and none taps into the psychological meaning of loyalty
(Oliver, 1999).
According to Jacoby and Kyner (1973), brand loyalty is the biased (i.e.
non-random) behavioural response (i.e. purchase), expressed over time, by
some decision-making unit, either on the part of an individual, family or
organization, with respect to one or more alternative brands out of a set of such
brands, which means that it is necessary to distinguish between exclusivity
and loyalty and a function of psychological processes which involves the
evaluation of different alternatives using specific criteria. Similarly, Oliver A model of
(1999, p. 35) defines loyalty as: customer loyalty
. . . a deeply held commitment to rebuy or repatronize a preferred product/service consistently
in the future, thereby causing repetitive same-brand or same brand-set purchasing, despite
situational influences and marketing efforts having the potential to cause switching behavior.
Jacoby and Chestnut (1978) have explored the psychological meaning of loyalty 255
in an effort to distinguish it from behavioural (i.e. repeat purchase) definitions.
Their analysis concludes that consistent purchasing as an indicator of loyalty
could be invalid because of happenstance buying or a preference for
convenience, and that inconsistent purchasing could mask loyalty if consumers
were multi-brand loyal.
Therefore, loyalty is a concept that goes beyond simple purchase repetition
behaviour since it is a variable which basically consists of one dimension
related to behaviour and another related to attitude, where commitment is the
essential feature (Day, 1969; Jacoby and Kyner, 1973; Berné, 1997). According
to Jacoby and Chestnut (1978), Solomon (1992) and Dick and Basu (1994), the
combination of these two components enables us to distinguish two types of
customer loyalty concepts:
(1) loyalty based on inertia, where a brand is bought out of habit merely
because this takes less effort and the consumer will not hesitate to switch
to another brand if there is some convenient reason to do so; and
(2) true brand loyalty, which is a form of repeat purchasing behaviour
reflecting a conscious decision to continue buying the same brand, and it
must be accompanied by an underlying positive attitude and a high
degree of commitment toward the brand.
Inertia means the consumer is buying the same brand, not because of true
brand loyalty, but because it is not worth the time and trouble to search for an
alternative. A competitor who is trying to change a buying pattern based on
inertia can often do so rather easily, because little resistance to brand switching
will be encountered if some reason to do so is apparent (Solomon, 1992). On the
other hand, under low involvement conditions brand loyalty may reflect only
the convenience inherent in repetitive behaviour rather than commitment to the
brand purchased. Relatively uninvolved consumers are less likely to be brand
loyal and will be more likely to be brand switchers (Traylor, 1981).
In relation to true brand loyalty, Oliver (1999), based on the traditional
consumer attitude structure, considers that all three decision-making phases
must point to a focal brand preference if true brand loyalty exists. Thus:
(1) the brand attribute ratings (beliefs) must be preferable to competitive
offerings;
(2) this information must coincide with an affective preference (attitude) for
the brand; and
EJM (3) the consumer must have a higher intention (conation) to buy the brand
38,1/2 compared with that for alternatives.
Compared to an inertia situation where the consumer passively accepts a
brand, a true brand loyal consumer is actively involved with his or her
favourite.
256 In the retail banking market, the length of relationship between bank and
customer is a common feature. The tradition of the industry has been for banks
and other financial services organisations to engage in long-term customer
relationships. In agreement with Stewart (1998), the reasons for such
relationship longevity are open to interpretation. While genuine preference and
loyalty may have been instrumental, so also could ignorance, inertia and
dependence.
There has been a growing interest in recent years in analysing the factors
influencing customer loyalty. As a result, there are numerous works in
marketing which have attempted to explain the relationships between brand
loyalty and the various variables regarded as antecedents, the most significant
of which are customer satisfaction, and, to a lesser degree, switching costs
(Bearden and Teel, 1983; LaBarbera and Mazursky, 1983; Kasper, 1988;
Bloemer and Lemmink, 1992; Cronin and Taylor, 1992; Fornell, 1992; Oliva
et al., 1992; Anderson and Sullivan, 1993; Bloemer and Kasper, 1993, 1995;
Boulding et al., 1993; Berné, 1997; Oliver, 1999).
Customer satisfaction is a concept that has been widely debated in literature
and for which numerous definitions have been made, but researchers have yet
to develop a consensual definition of this concept. In this sense, Oliver (1997, p.
13) notes that “everyone knows what [satisfaction] is until asked to give a
definition. Then it seems, nobody knows”.
Giese and Cote (2000) suggest in their literature review that consumer
satisfaction comprises three basic components:
(1) the type of response, that is to say, whether the response is cognitive,
affective or conative, and its level of intensity, although those authors
concluded from their validation, carried out by means of group and
personal interview data, that satisfaction is a summary affective
response which varies in intensity;
(2) the centre of interest or the subject on which the response is focused,
which could be based on an evaluation of product-related standards,
product consumption experiences and/or purchase-related attributes
(e.g. salesperson); and
(3) the moment in time at which the evaluation is made, which may be
before choice, after choice, after consumption, after extended experience,
or at just about any other time.
This theoretical framework enables us to develop specific definitions adapted
to the conditions of the context specific to each study, which are conceptually
richer and more empirically useful than a generic definition, making it easier to A model of
interpret and compare empirical results. Along these lines, Halstead et al. (1994) customer loyalty
regard satisfaction as an affective response, focused on product performance
compared to some prepurchase standard during or after consumption. Mano
and Oliver (1993) establish that satisfaction is an attitude or evaluative
judgement varying along the hedonic continuum focused on the product, which
is evaluated after consumption. Fornell (1992) identifies satisfaction as an
257
overall evaluation based on the total purchase and consumption experience
focused on the perceived product or service performance compared with
prepurchase expectations over time. Oliver (1997, 1999) regards satisfaction as
a fulfilment response/judgment, focused on product or service, which is
evaluated for one-time consumption or ongoing consumption.
A concept which is very closely related with satisfaction is perceived
quality, and the differences between these two have not always been very
clearly defined, both having been used on occasion in an indistinct manner. In
an attempt to clarify the distinction between satisfaction and perceived quality,
Anderson et al. (1994) consider that satisfaction requires previous consumption
experience and depends on price, whereas quality can be perceived without
previous consumption experience and does not normally depend on price,
although in circumstances where there is little available information or where
quality evaluation is difficult, price can be an indicator of quality. In this sense,
Spreng and Mackoy (1996), starting from Oliver’s (1997, 1999) conceptual
model of service quality and service satisfaction, concluded that these
constructs are distinct and have different antecedents.
On the other hand, there is a lack of consensus in literature and among
researchers about the causal link between the two constructs. While quality
service literature claims that customer satisfaction is an antecedent of
perceived quality (Parasuraman et al., 1988; Bitner, 1990; Carman, 1990), other
authors regard the relationship as being the other way round, in other words,
perceived quality is considered an antecedent of customer satisfaction
(Woodside et al., 1989; Reidenbach and Sandifer-Smallwood, 1990; Cronin and
Taylor, 1992; Fornell, 1992; Anderson and Sullivan, 1993; Gotlieb et al., 1994;
Spreng and Mackoy, 1996).
According to Parasuraman et al. (1994), this controversy derives from the
type of evaluation that is carried out in terms of quality and satisfaction, and it
is possible to distinguish between a transaction-specific evaluation and an
overall evaluation as the result of cumulative experience. While service quality
researchers start from the premise that satisfaction is a transaction-specific
evaluation and that quality is an overall evaluation made using a whole set of
cumulative evaluations, researchers focusing on the satisfaction topic tend to
have quite the opposite point of view. Parasuraman et al. (1994), following up
the work of Teas (1993), consider that service quality and satisfaction can be
examined from both transaction-specific as well as global perspectives and
EJM they suggest that an interesting direction for further research is to analyse the
38,1/2 causal links of these two constructs from both perspectives. Nevertheless, from
a theoretical point of view these authors establish that perceived quality is an
antecedent of satisfaction when both constructs are measured in the context of
a transaction-specific perspective.
258 Another brand loyalty antecedent is known as switching costs, which can be
defined as the technical, financial or psychological factors which make it
difficult or expensive for a customer to change brand (Selnes, 1993). According
to Alet i Vilaginés (1994), the switching costs can be broken down as follows:
.
the customer’s personal costs, referring to tradition and the client’s habits,
to effort in terms of the time and commitment needed to evaluate new
alternatives, to the economic advantages associated with loyalty, to the
social and psychological risks stemming from making a wrong choice,
and to the established contracts with the supplier company; and
.
costs associated with the product, such as the costs of redesigning the
process of production or consumption, investment in related
equipment, and contractual costs.
When the costs of switching brand are high for the customer, there is a greater
probability that the customer will remain loyal in terms of repeat purchase
behaviour, because of the risk or expense involved in switching and because of
the accompanying decrease in the appeal of other alternatives (Wernerfelt,
1991; Selnes, 1993; Klemperer, 1995; Ruyter et al., 1996; Antón Martı́n et al.,
1998). However, if loyalty is defined as true loyalty, the relationship between
this construct and the switching costs is not so simple. For example, it might
easily be that the customer repurchases, but due to his dissatisfaction, he does
not recommend the product or service to others. Moreover, the effect of
switching costs on loyalty varies with the type of industry, the category of the
product and the characteristics of the customer (Fornell, 1992). In the banking
industry, Sheth and Parvatiyar (1995) found some factors that may inhibit
customer exit in retail banking; for instance, the length of their relationships
with the bank; the fact that they knew, and were known by, the branch staff;
and the perception that closing/transferring accounts was difficult. These
factors, whether real or perceived, act as exit barriers.
Wernerfelt (1991), Selnes (1993) and Klemperer (1995) consider that brand
loyalty increases considerably when switching costs and customer satisfaction
converge, although a competitor will find it more difficult to capture a customer
of a rival brand when the customer’s loyalty is based on satisfaction than when
it is based on switching costs. Along these lines, Fornell (1992) establishes two
disadvantages of switching costs that highlight the greater relative weight of
satisfaction as an antecedent of customer loyalty. These disadvantages are,
first, the greater difficulty of capturing new customers when these are
conscious of the existence of switching costs, and, second, the possibility that A model of
external forces can eliminate switching barriers. customer loyalty
Consistent with Antón Martı́n et al. (1998), in the relationship existing
between customer satisfaction/perceived switching costs and loyalty, the
degree of elaboration which is followed in the decision-making process can
have a moderating influence. Elaboration is a construct based on the
information processing theory (Petty and Cacioppo, 1997) and is determined by
259
the motivation and the ability of a consumer to elaborate on the brand choice
(Bloemer and Ruyter, 1998). Motivation can be operationalised by bank choice
involvement and ability can be operationalised by bank choice deliberation.
Despite the fact that motivation and the ability of a consumer to elaborate on
the choice can be high, if the consumer does not perceive differences among
brands, the degree of elaboration in the decision-making process may be low.
Therefore, to measure elaboration we have also included the dimension related
to the perceived distinction between different brands.
Moreover, when the levels of satisfaction and switching costs are equal in
value, consumers that take a longer time to make a decision tend to be more
loyal to the brand chosen because the decision-making process is more
conscious and deliberate. However, this moderating influence can vary with
respect to product categories. Thus, for example, in the work carried out by
Antón Martı́n et al. (1998), after analysing the moderating influence of the
degree of elaboration in the relationship between satisfaction/switching costs
and loyalty for three categories of products, they were only able to identify a
significant moderating influence in one of the product categories analysed.
Moreover, in the paper on store loyalty by Bloemer and Ruyter (1998), they
concluded that the customer who elaborates more on department store
shopping might take more stores into consideration. This might lead to less
loyalty in the case of a stronger motivation and ability to evaluate a store.
Nevertheless, when combined with the amount of satisfaction and as a
moderator variable, elaboration strengthens the positive effect of store
satisfaction on store loyalty.

Research design
Objectives
This paper has three objectives related to customer loyalty in the retail banking
market. First, using a structural equation model, we analyse empirically
whether customer satisfaction and perceived switching costs are antecedents of
customer loyalty toward different retail banks. Second, we explore the causal
direction between perceived quality and satisfaction, in order to establish
whether perceived quality is an indirect antecedent of customer loyalty, a
consequence of satisfaction or whether there exists a bidirectional relationship
between perceived quality and satisfaction. With this second objective, we
want to add new empirical evidence on the causal relationship between
EJM perceived quality and satisfaction, with the object of shedding some light on
38,1/2 the subject. Finally, we study the moderating influence that the degree of
elaboration might exert on the causal relationships between loyalty and the
aforementioned antecedents.
More specifically, the above-mentioned objectives and the hypotheses
deriving from the same are as follows.
260 Objective 1. To analyse whether satisfaction and switching costs are
antecedents of customer loyalty in the retail banking market.
H1.1. The greater the satisfaction, the greater the customer loyalty.
H1.2. The greater the perceived switching costs, the greater the customer
loyalty.
Objective 2. To explore the causal direction between perceived quality and
satisfaction.
H2.1. The greater the perceived quality, the greater the customer
satisfaction.
H2.2. The greater the customer satisfaction, the greater the perceived
quality.
H2.3. There exists a bidirectional relationship between perceived
satisfaction and customer satisfaction.
Objective 3. To study the moderating influence that the degree of elaboration
can have on the causal relationships between customer satisfaction/perceived
switching costs and loyalty.
H3.1. The greater the degree of elaboration, the greater the positive impact
of satisfaction on loyalty.
H3.2. The greater the degree of elaboration, the greater the positive impact
of perceived switching costs on loyalty.

Methodology
Data were gathered from personal interviews using a structured questionnaire.
The customers of the six banks with the largest market share, in the
geographical zone in which this study was carried out, were selected as the
target population. These are Banco Bilbao-Vizcaya, Banco Central-Hispano,
Banesto, Banco de Santander, Caja de Canarias and La Caixa. The main reason
for selecting the clients of those banks is that those six banks have 71.68 per
cent of all the bank branches in the geographical region of the study, according
to the data of the Internal Report of the Caja de Canarias (2001). None of the
banks whose clients are not included in the sample has more than 2 per cent of
the total number of bank branches. The final sample consists of 576 individuals
who stated that they were habitual customers of one of the aforementioned
banks. Those clients were selected according to age, education and occupation, A model of
which were all proportional to the actual population of the region. The customer loyalty
sampling error was ^4:16 per cent at the 95.5 per cent level. The interviews
were carried out in situ, at the main door of the branches. We carried out the
same number of surveys for each of the banks in order to be able to obtain an
acceptable sample size of the clients of each bank.
The items of the scales and its reliabilities measured with Cronbach’s alpha
261
are reported in Table I.
Loyalty was measured indirectly by means of an attitude scale, which, as
Berné (1997) points out, is the most commonly used system of measurement due
to the difficulty involved in obtaining sequential information about purchase
repetition. To do this we used a seven-point, three-item Likert scale, which
measures resistance to switching bank, individuals’ loyalty attitude and the
degree to which they would recommend the bank that they use. These variables
measure the degree of commitment toward the bank and intention to continue
with the relationship. Therefore, following the properties of true brand loyalty
established by Oliver (1999), the first item attempts to inversely measure the
intention to continue the relationship with the present bank (conation), and the
other two items measure affective preference for the bank (attitude). We have not
evaluated the brand attribute rating (beliefs) because of problems caused by the
length of the questionnaire and because an in-depth study of the clients’ beliefs
was not the main objective of this work. However, attitudes are a direct
consequence of beliefs, and so, when measuring the affective component, we are,
albeit indirectly, considering the cognitive component of loyalty.
In order to evaluate customer satisfaction with their bank we followed
Fornell (1992) and used a seven-point, three-item Likert scale which measures
general satisfaction with the banking entity, the degree to which the bank
confirms customers’ expectations, and the gap which customers consider to
exist between the bank they use and what they regard as being the perfect or
ideal bank. We have used Fornell’s (1992) scale because it incorporates the
three aspects of satisfaction most widely used in the literature. In this way, and
in line with the basic components of satisfaction established by Giese and Cote
(2000), the scale used complies with the following characteristics:
.
the type of response is affective;
.
the centre of interest is based on an evaluation of bank consumption
experiences; and
. the moment of evaluation is after extended experience.
Perceived quality was measured based on an overall evaluation as the result of
cumulative experience of the customer, using a 20-item SERVPERF scale
adapted to the banking industry. Although this scale has a high number of
items, we have not reduced it because we consider that it was important to
maintain the original SERVPERF scale. This scale has five components
EJM
Loyalty Alpha = 0.7829
38,1/2 L1 I do not like to change to another bank because I value the selected bank
L2 I am a customer loyal to my bank
L3 I would always recommend my bank to someone who seeks my advice
Satisfaction Alpha = 0.8106
262 S1 To what extent does this bank live up to your general expectations of it?
S2 Imagine the perfect bank. How far and/or close does this bank come to your ideal?
S3 Given your experience with this bank, how satisfied or dissatisfied are you with it
overall?
Quality Alpha = 0.9373
Q1 X’s physical facilities are attractive and comfortable
Q2 X has modern-looking equipment
Q3 X’s employees are tidy in appearance
Q4 Materials associated with the service are visually appealing at X
Q5 X insists on error-free records
Q6 When you have a problem, X shows a sincere interest in solving it
Q7 Employees of X solve your problems when they promise to do so
Q8 X provides its services at the time it promises to do so
Q9 X performs the service right the first time
Q10 Employees of X give you prompt service
Q11 Employees of X are always willing to help you
Q12 Employees of X are never too busy to respond to your requests
Q13 Employees of X tell you exactly when services will be performed
Q14 The behaviour of employees of X instills confidence in customers
Q15 Employees of X are constantly courteous to you
Q16 Employees of X have the knowledge to answer your questions
Q17 X gives you individual and personal attention
Q18 X has operating hours convenient to all its customers
Q19 X has your best interests at heart
Q20 Employees of X understand your specific needs
Switching costs Alpha = 0.8039
SC1 To change to another bank involves investing time in searching for information about
other banks
SC2 To change to another bank involves much effort in deciding which other bank to use
SC3 To change to another bank involves a risk in choosing another bank which might turn
out not to satisfy me
Degree of elaboration in the bank selection process
Degree of deliberation
DEL 1 When I selected my bank, I made a previous comparison with other bank entities
DEL 2 When I selected my bank, I did a previous analysis of the advantages and disadvantages
of the chosen bank in comparison with other bank entities
Degree of involvement
I1 The first time that I chose a bank, it was a very important decision for me

Table I. Degree of differentiation between bank entities

Scales and reliability DIF 1 I think there are significant differences between different banks
coefficients Note: X denotes the name of the main bank
(tangibles, reliability, responsiveness, assurance and empathy) and was A model of
adopted since it is still the most widely used instrument, having been customer loyalty
developed from data collected across five separate service categories, one of
which was retail banking. Furthermore, some studies show that this scale is a
good interpretation of perceived quality in retail banking (Llorens Montes,
1996; Yavas et al., 1997; Bloemer et al., 1998. In accordance with the codes of the
variables which can be seen in Table I, tangibles correspond with labels Q1 to 263
Q4, reliability with labels Q5 to Q9, responsiveness with labels Q10 to Q13,
assurance with labels Q14 to Q16, and finally, empathy with labels Q17 to Q20.
After carrying out an exploratory factorial analysis of this scale, we were able
to isolate three factors, which represent 59.4 per cent of explained variance. The
first of these factors accounts for some 47.8 per cent of the variance and covers
the dimensions of reliability, responsiveness and assurance with the banking
entity; the second factor explains 6.6 per cent of the variance and deals with the
dimension related to empathy; and finally, the third factor accounts for 5 per
cent of the variance and represents the tangibles of the banking entities. The
differences which exist between the theoretical dimensions of the scale and
those obtained in our research could be due to the fact that the dimensionality
of the service quality scale varies from industry to industry (Babakus and
Boller, 1992). Along these lines, Carman (1990) and Buttle (1995) identified a
number of dimensions that differ from the five included in the SERVPERF
scale. As the number of dimensions of this scale does not coincide with the
theoretical dimensions, and the percentage of explained variance is relatively
reduced, we opt to use all the items of the SERVPERF scale in our structural
equations model.
With respect to switching costs, and using the typology proposed by Alet i
Vilaginés (1994), we have included the personal switching costs that are most
directly related to the industry that is the subject of our research. We have thus
used a seven-point, three-item Likert scale to evaluate the time required to
search for information about other banks, the effort involved in deciding on
another bank and the risk of making a mistake with the switch.
Finally, in order to measure the degree of elaboration in the bank selection
process, we used three seven-point Likert sub-scales in accordance with Antón
Martı́n (1998) to measure the following three dimensions:
(1) the degree of deliberation in the bank selection process;
(2) the degree of involvement in the choice of bank; and
(3) the degree of differentiation which exists between the different bank
entities.

Results and discussion

In order to analyse the antecedents of customer loyalty in the retail banking
market and to explore the direction which exists in the relationship between
perceived quality and customer satisfaction we have applied the structural
EJM equations model, first because it enables us to estimate the multiple and crossed
38,1/2 relationships which exist between dependent and independent variables, and
second, for its capacity to represent constructs not observed in these relationships
and to take into account measurement error in the estimation process.
Before specifying the structural equations model we tested out the reliability of
the scales to be used by means of the Cronbach’s alpha, as can be seen in Table I.
264 The coefficients obtained demonstrate that the indicators used to measure the
different constructs are reliable since their values oscillate between 0.7829 and
0.9373. The high Cronbach’s alpha for the SERVPERF scale is influenced by the
high number of scale items; however, the SERVPERF scale has been extensively
applied in numerous studies because of its good psychometric properties.
With the aid of the literature review, we have proposed a causal model using
the software AMOS 3.6 in which we propose that customer satisfaction and
switching costs are antecedents of customer loyalty and that there exists a
bidirectional relationship between customer satisfaction and perceived quality.
We used a correlation matrix, shown in the Appendix, as the starting point,
since because it is a standardised variance-covariance matrix it enables us to
make direct comparisons of regression coefficients between the variables
included in the model. While many researchers propose a two-stage structural
modelling process in which the measurement and structural models are kept
separate, in our study, and in the line of Hair et al. (1990), we will make a
simultaneous estimation of the measurement and structural models based on
the strength of extensive theoretical support which exists for the use of such a
model and on the high degree of reliability of the measurements used.
Figure 1 displays the resulting causal model with its standardised regression
coefficients, together with the global fit measures for the model which are used
to verify the quality of the fit via the correspondence between the observed
correlation matrix and that which is reproduced using the proposed model. As
measurements of fit we obtained a chi-square value of 414.814 with 375 degrees
of freedom (p ¼ 0:076). This value was obtained for a sample of 170 individuals,
bearing in mind that the recommended sample size for obtaining a valid
contrast with the chi-square statistic using the maximum likelihood criterion is
100-200. This is because, when the sample size is increased above this limit, the
maximum likelihood criterion becomes increasingly sensitive, while the fit of
the model is adversely affected (Hair et al., 1999). In fact, the values of the
remaining measures of the global fit of the model are exactly the same for a
sample size of 576 as for one of 170, with the exception of the standardised
chi-square (CMIN/DF), which obviously increases in value when the sample size
is increased, and the Tucker-Lewis Index (TLI), which changes from 0.982 to
0.879 when the sample size is increased to 576. In our model the values obtained
for the goodness of fit index (GFI ¼ 0:849) and the root mean square residual
(RMR ¼ 0:068) indicate a good absolute fit of the model. Regarding the
incremental fit measurements, which compare the proposed model with the
 A model of
customer loyalty

265

Figure 1.
Regression model
(standardised values)

independence model, we obtained values close to, or superior t,o the 0.9 value
generally considered acceptable. Finally, the values of the last two
measurements, AGFI ¼ 0:825 and CMIN=DF ¼ 1:106, indicate that the model
is parsimonious, since the value of the first is close to 0.9 while the second falls
in the recommended interval between 1 and 2.
Once the global fit has been accepted, we can proceed to evaluate separately
each of the latent variables, analysing their convergent validity by evaluating
the weights of the indicators and their reliability and variance. Thus, in
accordance with Anderson and Gerbing (1988), we assume that convergent
validity exists when the regression weights of a construct’s measurements are
statistically significant, in other words, when the critical ratio (CR) of the
variables observed against their respective latent variables is over 1.96 at the
0.05 level. In our case, as can be observed from the data displayed in Table II,
all the critical ratios of the indicators of constructs satisfy this criterion, and the
convergent validity of the measurements can thus be demonstrated, the
EJM
Standard Critical Standardised
38,1/2 Regression weights Estimates deviation ratio (CR) estimates

Loyalty ˆ Switching 0.145 0.052 2.800 0.182

Loyalty ˆ Satisfaction 0.765 0.079 9.695 0.833
Satisfaction ˆ Quality 0.239 0.542 0.441 0.231
266 Quality ˆ Satisfaction 0.615 0.317 1.938 0.635
SC1 ˆ Switching 0.849 0.072 11.762 0.823
SC2 ˆ Switching 1.000 – – 0.949
S3 ˆ Satisfaction 1.000 – – 0.866
S1 ˆ Satisfaction 0.913 0.078 11.742 0.791
L3 ˆ Loyalty 1.000 – – 0.807
L2 ˆ Loyalty 0.804 0.099 8.122 0.645
Q6 ˆ Quality 0.908 0.077 11.804 0.761
Q7 ˆ Quality 0.992 0.073 13.547 0.831
Q8 ˆ Quality 0.858 0.079 10.869 0.719
Q11 ˆ Quality 1.000 – – 0.838
Q12 ˆ Quality 0.927 0.076 12.182 0.777
Q13 ˆ Quality 0.810 0.081 10.039 0.679
Q14 ˆ Quality 0.960 0.075 12.868 0.805
Q15 ˆ Quality 0.914 0.077 11.915 0.766
Q16 ˆ Quality 0.898 0.077 11.619 0.753
Q10 ˆ Quality 0.941 0.076 12.464 0.789
SC3 ˆ Switching 0.565 0.074 7.582 0.557
L1 ˆ Loyalty 0.914 0.098 9.310 0.735
S2 ˆ Satisfaction 0.724 0.084 8.645 0.627
Q5 ˆ Quality 0.574 0.087 6.566 0.482
Q1 ˆ Quality 0.475 0.090 5.308 0.398
Q4 ˆ Quality 0.589 0.087 6.764 0.494
Q3 ˆ Quality 0.690 0.084 8.169 0.579
Q9 ˆ Quality 0.810 0.081 10.047 0.679
Q18 ˆ Quality 0.588 0.087 6.753 0.493
Q19 ˆ Quality 0.729 0.083 8.748 0.611
Q2 ˆ Quality 0.390 0.091 4.287 0.327
Table II. Q20 ˆ Quality 0.767 0.082 9.339 0.643
Results of regression Q17 ˆ Quality 0.803 0.081 9.919 0.673
2
model Note: Squared multiple correlation (R ) = 72.7 per cent

proposed relationships between indicators and constructs having been verified.

With regard to reliability and variance, the data displayed in Table III
demonstrate that the different indicators are sufficient in the representation of
the constructs, since the reliability values exceed the recommended values of
0.70 and the variances are superior to or close to the 0.5 level.
Starting from the premise that the goodness of fit is high (R 2 ¼ 72:7 per
cent), we will proceed to analyse the hypotheses related to the first two
objectives of this study.
First, and with respect to loyalty antecedents, the relationships between
satisfaction/switching costs and loyalty display positive and statistically
significant regression coefficients. Satisfaction has a greater weight on loyalty A model of
than switching costs (0.833 and 0.182 respectively), which leads us to accept customer loyalty
H1.1 and H1.2.
Second, in order to test out the hypotheses related to the direction of the
relationship between perceived quality and customer satisfaction, we must
previously verify the stability of the non-recursive model between these two
latent variables. In this way, we obtained a stability index of 0.147 which, being 267
lower than 1, indicates that the regression coefficients are stable. The data
displayed in Table II show that there exists a positive and statistically
significant relationship within the satisfaction-perceived quality relationship,
thus verifying H2.2. The perceived quality-satisfaction relationship, although
positive, is not statistically significant, meaning that H2.1 and H2.2 cannot be
verified. These results, which are in contrast to the suggestions of Woodside
et al. (1989), Reidenbach and Sandifer-Smallwood (1990), Cronin and Taylor
(1992), Fornell (1992) and Anderson and Sullivan (1993), indicate that when
perceived quality and satisfaction are measured in a global perspective,
satisfaction is an antecedent of perceived quality and not vice versa, as has
been proposed by Carman (1990), Parasuraman et al. (1988) and Bitner (1990).
In order to address our third objective, which was to analyse the moderating
influence which the degree of elaboration in the bank selection process has on
the relationships between loyalty and customer satisfaction/switching costs,
we carried out a multiple regression analysis using the following equation
L ¼ a þ bSAT þ cSC þ dCLUS þ eCLUS
SAT þ fCLUS
SC
where:
L = Customer loyalty
SAT = Customer satisfaction
SC = Switching costs
CLUS = variable dummy which defines belonging either to the group of
customers with a high degree of elaboration in the decision-making
(1) or to the low level group (0).
In order to classify the customers into two categories according to the degree of
elaboration of their decision making, we applied a K-means cluster analysis to
the four variables that make up the scale used. The results can be observed in

Latent variables Reliability of the constructs Variance of the constructs

Perceived quality 0.940 0.450

Satisfaction 0.809 0.589 Table III.
Switching costs 0.824 0.619 Reliability of latent
Loyalty 0.778 0.541 variables
EJM Table IV. The 288 individuals who make up Cluster 1 display a greater degree
38,1/2 of elaboration in the decision-making process than the 287 individuals who
make up Cluster 2. Similarly, an ANOVA demonstrates that there are
significant differences between the two groups for each of the variables used in
the division process (p # 0:001). From the cluster, we have created dummy
variable CLUS, which has a value of 1 when individuals show a high degree of
268 elaboration and a value of 0 when the degree of elaboration is low.
The results of the multiple regression, which are displayed in Table V,
indicate that the degree of elaboration does not have a moderating influence on
the relationships between satisfaction/switching costs and loyalty, which leads
us to reject H3.1 and H3.2. One possible explanation for these results could be
derived from the industry which we chose as the subject for this research, since
the intangible nature of banking services could have a conditioning effect on the
degree of elaboration in the bank selection process. We could therefore state that
deliberation in bank selection and the perception of differences between rival
entities are far more complex processes than for other categories of products.

Conclusions
The aim of this research, which is based on the retail banking market, was to
carry out an empirical analysis of the factors determining customer loyalty,
using a structural equation modeling. We have tried to contribute to the
academic literature by considering customer loyalty not only as a frequency of
repeated purchase, but also including the psychological meaning of loyalty. In
this sense, we have differentiated between the two types of concept of customer
loyalty (loyalty based on inertia and true brand loyalty). Moreover, along these
lines, we analysed the influence on customer loyalty, not only of satisfaction,
which is the most widely studied factor in academic literature, but also of the

Cluster DIF1 I1 DEL1 DEL2 n

Table IV.
Size and profiles of the Cluster 1 3.715 4.535 5.576 5.670 288
clusters Cluster 2 3.199 3.014 2.056 2.216 287

Independence variables Betas Standard deviations Standardised betas t-value p-value

SAT 0.762 0.054 0.594 14.065 0.000

SC 0.164 0.046 0.156 3.557 0.000
CLUS 20.591 0.486 2 0.181 21.216 0.224
CLUS*SAT 0.030 0.085 0.047 0.355 0.723
Table V. CLUS*SC 0.084 0.068 0.135 1.239 0.216
Results of multiple Constant 0.156 0.297 – 0.525 0.599
regression Notes: F ¼86.75 ( p ¼0.000); Squared multiple Correlation ðR 2 Þ ¼ 43:47 per cent
switching costs, which are most directly related to the banking industry. A model of
Similarly, and taking into account the great controversy which exists about customer loyalty
whether perceived quality is an antecedent exerting an indirect influence on
loyalty and a direct influence on satisfaction, or whether, to the contrary, it is a
consequence of satisfaction, we have attempted to explain the causal direction
of the relationship existing between perceived quality and satisfaction.
The results of the proposed model, whose validity has been demonstrated by
269
the relatively acceptable levels of the indicators of the absolute, incremental
and parsimony fits and by its high explanatory value, demonstrate that both
satisfaction and switching costs can be regarded as loyalty antecedents.
Nevertheless, the influence exerted by satisfaction is far greater than that of
switching costs.
With respect to the direction of the relationship between satisfaction and
perceived quality, our model demonstrates that there is only a positive and
statistically significant relationship in this link. Thus, when these two constructs
are measured in a global perspective, satisfaction is an antecedent of perceived
quality in the retail banking market, and not vice versa. A possible theoretical
explanation of this result is that the satisfaction construct supposes an evaluative
judgement of the value received by the customer. Perhaps a customer perceives a
high level of quality and is not satisfied because of the economic sacrifice made
to obtain the service. Moreover, a high level of perceived quality may not
implicate a high level of satisfaction if the quality does not meet customer needs.
Furthermore, if a customer is satisfied, he will tend to value the perceived quality
more positively in order to be congruent with himself and to avoid dissonance.
In this study, we also explored the possible moderating influence which the
degree of elaboration in the bank selection process could exert on the
relationship between customer loyalty and its antecedents, using a multiple
regression analysis, and we reached the conclusion that, in the retail banking
market, the degree of elaboration does not exert any moderating influence.

Limitations and managerial implications

The results of this research are limited to and conditioned by the context in which
the empirical work was carried out, and we would therefore recommend future
research to study not only the direct influence of satisfaction and switching costs
on loyalty, but also the relationship between perceived quality and satisfaction at
both the global and specific perspectives in other industries and for other
categories of products. In the same way, an interesting line of enquiry would be to
replicate the research across the corporate sector of the banking industry.
Similarly, we would recommend studying other possible antecedents which could
have an influence on bank customer loyalty, for example, some of the dimensions
which constitute brand equity, such as brand image, reputation and awareness; as
well as the market orientation strategy implemented by the bank entities. Such
factors are important to the process of adding value to service offerings and, hence,
EJM achieving competitive advantage in retail financial services markets (Devlin, 2000).
38,1/2 Gardener et al. (1999) argue that, in the new financial environment, there is also a
greater focus on banks’ achieving added value and improving their public image.
Issues relating to the elements of the offering that should be emphasised when
adding value may be particularly important in the case of services. For offerings
which are highly intangible, either mentally, or cognitively, as well as physically,
270 options for adding value, and hence, achieving competitive advantage, may be
limited due to customer reliance on experience and credence qualities during the
purchase decision. Financial services are arguably highly typical of service
offerings in general, with more complicated financial services in particular being
highly intangible as well as problematic in terms of consumer cognition (Devlin,
2000). According to Nguyen and LeBlanc (1998), in financial services, future
research might consider a better knowledge of how customer satisfaction, service
quality, and value interact to influence image assessments and loyalty, which
promises to provide useful insights for formulating competitive strategy.
Finally, we would draw attention to the possibility of carrying out further
research to analyse the effect of other possible moderating variables in the
relationship between loyalty and its antecedents, such as, for example, the
psycho-demographic characteristics of individuals like the degree of financial
knowledge on the part of customers in the retail banking market.
Our findings have several managerial implications. The impact of
satisfaction on loyalty is considerably stronger than the cost of switching.
This implies that banks should place greater emphasis on achieving high levels
of customer satisfaction than on creating switching barriers. This is because,
on the one hand, loyalty is based mainly on satisfaction, and on the other,
switching costs present the additional disadvantage of the difficulty of
attracting new customers when these are aware of the existence of such costs,
and the possibility that outside forces may eliminate the barriers erected by
switching costs. Nevertheless, the direct positive relationship between
switching costs and loyalty may imply that banks could undertake actions
that increase switching costs for their customers, such as establishing preferred
customer programmes, which can also contribute to increasing customer
satisfaction. In this sense, Barnes and Howlett (1998) argue that the loyalty
programmes would be customer-focused and the companies would examine:
.
the manner in which the customer defines a relationship;
.
whether the conditions under which the company interacts with
customers are conducive to forming relationships; and
.
the factors which contribute most to quality relationships.
On the other hand, while there may not be a direct relationship between overall
service quality and satisfaction response, the banks should not overlook the
importance of quality, whenever the quality improvement efforts are oriented
to meet the customers’ needs.
References A model of
Alet i Vilaginés, J. (2000), Marketing Relacional: Cómo Obtener Clientes Leales y Rentables, Ed., customer loyalty
Gestión S.A., Barcelona.
Anderson, E.W. and Sullivan, M. (1993), “The antecedents and consequences of consumer
satisfaction for firms”, Marketing Science, Vol. 12, Spring, pp. 125-43.
Anderson, E.W., Fornell, C. and Lehmann, D.R. (1994), “Customer satisfaction, market share, and
profitability: findings from Sweden”, Journal of Marketing, Vol. 58, July, pp. 53-66. 271
Anderson, J.C. and Gerbing, D.W. (1988), “Structural modeling in practice: a review and
recommended two-step approach”, Psychologycal Bulletin, Vol. 103 No. 3, pp. 411-23.
Antón Martı́n, C., Garrido Samaniego, M.J. and Rodrı́guez Escudero, A.I. (1998), “El Grado de
Elaboración del Proceso de Elección de Compra como Refuerzo de la Lealtad de Marca”, X
Encuentro de Profesores Universitarios de Marketing, Santander, pp. 435-48.
Babakus, E. and Boller, G.W. (1992), “An empirical assesment of the SERVQUAL scale”, Journal
of Business Research, Vol. 24, pp. 253-68.
Barnes, J.G. and Howlett, D.M. (1998), “Predictors of equity in relationships between financial
services providers and retail customers”, International Journal of Bank Marketing, Vol. 16
No. 1, pp. 15-23.
Barroso Castro, C. and Martı́n Armario, E. (1999), “Nivel de Servicio y Retención de Clientes: El
Caso de la Banca en España”, Revista Española de Investigación de Marketing ESIC, Vol. 3
No. 1, pp. 9-33.
Bearden, W.O. and Teel, J.E. (1983), “Selected determinants of consumer satisfaction and
complaint reports”, Journal of Marketing Research, Vol. 20, February, pp. 21-8.
Berné, C. (1997), “Modelización de la Poscompra: Satisfacción y Lealtad”, in Múgica Grijalva, J.M.
and Ruiz de Maya, S. (Eds), El Comportamiento del Consumidor, Cap. 5, Ariel Economı́a,
Barcelona, pp. 163-80.
Bitner, M.J. (1990), “Evaluating service encounters: the effects of physical surrounding and
employee response”, Journal of Marketing, Vol. 54, April, pp. 69-82.
Bloemer, J.M. and Kasper, H. (1993), “Brand loyalty and brand satisfaction: the case of buying
audio cassettes anew in The Netherlands”, in Proceeding of the 22nd European Academy
Conference, Barcelona.
Bloemer, J.M. and Kasper, H. (1995), “The complex relationship between consumer satisfaction
and brand loyalty”, Journal of Economic Psychology, Vol. 16, pp. 311-29.
Bloemer, J.M. and Lemmink, J.G. (1992), “The importance of customer satisfaction in explaining
brand and dealer loyalty”, Journal of Marketing Management, Vol. 8, pp. 351-64.
Bloemer, J. and Ruyter, K. (1998), “On the relationship between store image, store satisfaction and
store loyalty”, European Journal of Marketing, Vol. 32 No. 5/6, pp. 499-513.
Bloemer, J., Ruyter, K. and Peeters, P. (1998), “Investigating drivers of bank loyalty: the complex
relationship between image, service quality and satisfaction”, International Journal of
Bank Marketing, Vol. 16 No. 7, pp. 276-86.
Boulding, W., Kalra, A., Staelin, R. and Zeithaml, V.A. (1993), “A dynamic process model of
service quality: from expectations to behavioral intentions”, Journal of Marketing
Research, Vol. 30 February, pp. 7-27.
Buttle, F.A. (1995), “What future for SERVQUAL?”, in Proceedings of the 24th European
Academy Conference, Paris, pp. 211-30.
Caja de Canarias (2001), Informe Gran Canaria: Oficinas Bancarias, Población y Capacidad de
Consumo por Municipios, Caja de Canarias.
EJM Carman, J.M. (1990), “Consumer perceptions of service quality: an assesment of the SERVQUAL
dimensions”, Journal of Retailing, Vol. 66 No. 1, pp. 33-5.
38,1/2
Cronin, J.J. Jr and Taylor, S.A. (1992), “Measuring service quality: a reexamination and
extension”, Journal of Marketing, Vol. 56, July, pp. 55-68.
Day, G.S. (1969), “A two-dimensional concept of brand loyalty”, Journal of Advertising Research,
Vol. 9, September, pp. 29-35.
272 Devlin, J.F. (2000), “Adding value to retail financial services”, The International Journal of Bank
Marketing, Vol. 18 No. 5, pp. 222-32.
Dick, A.S. and Basu, K. (1994), “Customer loyalty: toward and integrated conceptual framework”,
Journal of the Academy of Marketing Science, Vol. 22 No. 2, pp. 99-103.
Fornell, C. (1992), “National satisfaction barometer: the Swedish experience”, Journal of
Marketing, Vol. 56, January, pp. 6-21.
Giese, J.L. and Cote, J.A. (2000), “Defining consumer satisfaction”, Academy of Marketing Science
Review (on line).
Gardener, E., Molyneux, P. and Williams, J. (1999), European Savings Banks. Coming of Age?,
Lafferty Publication Ltd, Dublin.
Gilmore, A. (1997), “Implementing quality in a services marketing contex”, Marketing Intelligence
and Planning, Vol. 15 No. 4, pp. 185-9.
Gotlieb, J., Grewal, D. and Brown, S. (1994), “Consumer satisfaction and perceived quality:
complementary or divergent constructs”, Journal of Applied Psychology, Vol. 79 No. 6,
pp. 875-85.
Hair, J.F. Jr, Anderson, R.E., Tatham, R.L. and Black, W.C. (1999), Análisis Multivariante, 5th ed.,
Prentice-Hall, Madrid.
Halstead, D., Hartman, D. and Schmidt, S.L. (1994), “Multisource effects on the satisfaction
formation process”, Journal of the Academy of Marketing Science, Vol. 22, Spring,
pp. 114-29.
Jacoby, J. and Chestnut, R. (1978), Brand Loyalty: Measurement and Management, John Wiley &
Sons, New York, NY.
Jacoby, J. and Kyner, D.B. (1973), “Brand loyalty vs repeat purchasing behavior”, Journal of
Marketing Research, Vol. 10, February, pp. 1-19.
Kasper, J.D. (1988), “On problem perception, dissatisfaction and brand loyalty”, Journal of
Economic Psychology, Vol. 9, pp. 387-97.
Klemperer, P. (1995), “Competition when consumers have switching costs: an overview with
applications to industrial organization macroeconomics and international trade”, Review of
Economics Studies, Vol. 62, October, pp. 515-39.
LaBarbera, P.A. and Mazursky, D. (1983), “A longitudinal assesment of consumer
satisfaction/dissatisfaction”, Journal of Marketing Research, Vol. 20, November,
pp. 393-404.
Llorens Montes, F.J. (1996), “Una evaluación de la relación entre la calidad de servicio,
satisfacción y actitudes del cliente”, Esic-Market, Julio-Septiembre, pp. 121-32.
Mano, H. and Oliver, R.L. (1993), “Assessing the dimensionality and structure of the consumption
experience: evaluation, feeling, and satisfaction”, Journal of Consumer Research, Vol. 20,
December, pp. 451-66.
Nguyen, N. and LeBlanc, G. (1998), “The mediating role of corporate image on customers’
retention decisions: an investigation in financial services”, International Journal of Bank
Marketing, Vol. 16 No. 2, pp. 52-65.
Oliva, T.A., Oliver, R.L. and MacMillan, I.C. (1992), “A catastrophe model for developing services A model of
satisfaction strategies”, Journal of Marketing, Vol. 56, July, pp. 83-95.
customer loyalty
Oliver, R.L. (1997), Satisfaction. A Behavioral Perspective on the Consumer, McGraw-Hill, New
York, NY.
Oliver, R.L. (1999), “Whence consumer loyalty?”, Journal of Marketing, Vol. 63, October, pp. 33-44.
Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1988), “SERVQUAL: a multiple-item scale for
measuring consumer perceptions of service quality”, Journal of Retailing, Vol. 64 No. 1, 273
pp. 12-40.
Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1994), “Reassessment of expectations as a
comparison standard in measuring service quality: implications for further research”,
Journal of Marketing, Vol. 58 January, pp. 111-24.
Petty, R.E. and Cacioppo, J.T. (1997), Communication and Persuasion: Central versus Peripheral
Routes to Attitude Change, Springer Verlag, New York, NY.
Reichheld, F. (1993), “Loyalty-based management”, Harvard Business Review, March-April,
pp. 64-73.
Reichheld, F. and Sasser, W.E. (1990), “Zero defections: quality comes to service”, Harvard
Business Review, September-October, pp. 105-11.
Reidenbach, R.E. and Sandifer-Smallwood, B. (1990), “Exploring perceptions of hospital
operations by a modified SERVQUAL approach”, Journal of Health Care Marketing,
Vol. 10, December, pp. 47-55.
Rodrı́guez Sánchez, J.A. and Rodrı́guez Parada, S. (1993), “Calidad de servicio: exigencia actual
para entidades financieras competitivas”, Esic-Market, October-November, pp. 93-107.
Ruyter, K., Wetzels, M. and Bloemer, J. (1996), “On the relationship between perceived service
quality and behavioral intentions: a cross-sectional perspective”, in Berács, J., Bauer, A.
and Simons, J. (Eds), Proceedings 25th EMAC Conference, Budapest.
Selnes, F. (1993), “An examination of the effect of product performance on brand reputation,
satisfaction and loyalty”, Journal of Marketing, Vol. 27 No. 9, pp. 19-35.
Sheth, J. and Parvatiyar, A. (1995), “Relationship marketing in consumer markets: antecedents
and consequences”, Journal of the Academy of Marketing Science, Vol. 23 No. 4, pp. 255-71.
Solomon, M.R. (1992), Consumer Behavior, Allyn & Bacon, Boston, MA.
Spreng, R.A. and Mackoy, R.D. (1996), “An empirical examination of a model of perceived
services quality and satisfaction”, Journal of Retailing, Vol. 72 No. 2, pp. 201-14.
Stewart, K. (1998), “An exploration of customer exit in retail banking”, International Journal of
Bank Marketing, Vol. 16 No. 1, pp. 6-14.
Teas, R.K. (1993), “Expectations, performance evaluation and consumers’ perceptions of quality”,
Journal of Marketing Research, Vol. 25, May, pp. 204-12.
Traylor, M.B. (1981), “Product involvement and brand commitment”, Journal of Advertising
Research, December, pp. 51-6.
Wernerfelt, B. (1991), “Brand loyalty and market equilibrium”, Marketing Science, Vol. 10 No. 3,
pp. 229-45.
Woodside, A.G., Frey, L.L. and Daly, R.T. (1989), “Linking service quality, customer satisfaction,
and behavioral intention”, Journal of Health Care Marketing, Vol. 19, December, pp. 5-17.
Yavas, U., Bilgin, Z. and Shemwell, D.J. (1997), “Service quality in the banking sector in an
emerging economy: a consumer survey”, The International Journal of Bank Marketing,
Vol. 15 No. 6, pp. 217-23.
 EJM

274
38,1/2

Table AI.
Correlation matrix
1.0000
0.5524 1.0000
Appendix

0.4251 0.5070 1.0000

0.5979 0.5763 0.5555 1.0000
0.1477 0.1493 0.2113 0.1742 1.0000
0.1917 0.1997 0.2764 0.2193 0.7633 1.0000
0.1700 0.2224 0.1353 0.2027 0.4563 0.5023 1.0000
0.7088 0.4801 0.3408 0.5413 0.0877 0.1412 0.1265 1.0000
0.5387 0.4088 0.2819 0.4009 0.0262 0.0943 0.0816 0.5276 1.0000
0.3552 0.2557 0.2502 0.2983 0.0814 0.1414 0.1063 0.2592 0.1868 1.0000
0.1998 0.2376 0.1579 0.1659 0.0352 0.0762 0.1090 0.1826 0.1495 0.4436 1.0000
0.3459 0.3113 0.2765 0.3419 0.1208 0.1372 0.1461 0.2946 0.2691 0.4045 0.3468 1.0000
0.3446 0.2318 0.1602 0.2465 0.0282 0.1113 0.1411 0.3024 0.2891 0.2593 0.2436 0.3755 1.0000
0.3028 0.3334 0.1990 0.3662 0.1003 0.0962 0.1336 0.2754 0.2572 0.1438 0.2502 0.3023 0.2248 1.0000
0.5187 0.4324 0.2935 0.4498 0.0513 0.1172 0.1246 0.4578 0.3918 0.3066 0.2540 0.4054 0.3722 0.4081 1.0000
0.5586 0.5022 0.3950 0.4845 0.1009 0.1681 0.1750 0.5060 0.4287 0.3175 0.2295 0.4554 0.3763 0.3781 0.7620
0.4321 0.3961 0.3282 0.4066 0.0608 0.0935 0.1384 0.3981 0.3584 0.2909 0.2580 0.4483 0.3593 0.3494 0.5821
0.4263 0.3738 0.2802 0.3801 0.0870 0.0966 0.1261 0.3418 0.3292 0.2726 0.2243 0.3579 0.3983 0.5116 0.5430
0.4769 0.4408 0.3280 0.4611 0.0086 0.0535 0.0706 0.3919 0.3301 0.2682 0.2294 0.4453 0.3524 0.3491 0.5768
0.4675 0.4640 0.3980 0.4769 0.0499 0.1066 0.1516 0.4159 0.3464 0.2906 0.2554 0.4431 0.3934 0.3298 0.6372
0.4206 0.4074 0.2686 0.4211 0.0306 0.0614 0.1226 0.3709 0.3239 0.2606 0.2304 0.4218 0.3354 0.4609 0.5564
0.3791 0.3540 0.2328 0.3954 0.0687 0.0686 0.1135 0.3747 0.3143 0.1924 0.1976 0.3864 0.3331 0.3186 0.4861
0.5284 0.4628 0.4150 0.4698 0.0708 0.1143 0.1062 0.4676 0.3319 0.3692 0.2573 0.5117 0.3926 0.3424 0.5551
0.4758 0.4436 0.3312 0.4605 0.0995 0.1281 0.1901 0.4176 0.3544 0.2652 0.2105 0.4881 0.3386 0.3278 0.5403
0.5234 0.4457 0.3480 0.4540 0.1085 0.1096 0.1641 0.4536 0.3622 0.3214 0.2082 0.4687 0.4295 0.3882 0.5207
0.4461 0.3714 0.2906 0.4084 0.0475 0.0767 0.1172 0.3545 0.2796 0.3012 0.2154 0.3815 0.3809 0.2823 0.5145
0.3601 0.2991 0.2610 0.3782 0.0719 0.1295 0.0566 0.2942 0.2239 0.2951 0.1848 0.2509 0.2517 0.2537 0.3519
0.5194 0.3969 0.2935 0.4932 0.0916 0.1556 0.1297 0.4179 0.3083 0.2085 0.2119 0.3426 0.3508 0.3250 0.4526
0.4826 0.4333 0.3462 0.4323 0.1205 0.1235 0.0901 0.4253 0.3450 0.1769 0.2090 0.3256 0.2933 0.2708 0.5076

(continued)
1.0000
0.6396 1.0000
0.5672 0.6316 1.0000
0.6336 0.5209 0.5222 1.0000
0.7145 0.5951 0.5552 0.7291 1.0000
0.6295 0.5691 0.5003 0.7171 0.7091 1.0000
0.5249 0.5427 0.4657 0.5664 0.5634 0.5598 1.0000
0.6486 0.5390 0.5007 0.6381 0.6921 0.6194 0.5257 1.0000
0.6349 0.4891 0.4577 0.6174 0.6960 0.6145 0.4730 0.7181 1.0000
0.6110 0.5349 0.5295 0.5369 0.5979 0.5386 0.5048 0.6552 0.6294 1.0000
0.5431 0.4409 0.4392 0.5205 0.5540 0.4823 0.4963 0.5591 0.4926 0.5560 1.0000
0.3677 0.3338 0.3295 0.3970 0.3568 0.3765 0.3577 0.3775 0.3387 0.4084 0.3778 1.0000
0.4842 0.4096 0.4072 0.4519 0.4367 0.4349 0.4809 0.4562 0.3978 0.4837 0.4886 0.4684 1.0000
0.5022 0.4291 0.4130 0.5034 0.5246 0.4996 0.5199 0.4974 0.4508 0.4776 0.4421 0.3736 0.5730 1.0000
customer loyalty

Table AI.
275
A model of