This article was downloaded by: [University of Sussex Library]

On: 06 February 2015, At: 20:21

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Total Quality Management & Business

Excellence
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/ctqm20

Perceived quality, customer

satisfaction, and customer loyalty: the
case of lexus in Taiwan
a b c
Chwo-ming Joseph Yu , Lei-yu Wu , Yu-ching Chiao & Hsing-
a
shia Tai
a
Department of Business Administration , National Chengchi
University , Taipei, Taiwan, People's Republic of China
b
Department of International Business , National Taipei College of
Business , Taipei, Taiwan
c
Department of Business Administration , National Changhua
University of Education , Taipei, Taiwan
Published online: 24 Jan 2007.

To cite this article: Chwo-ming Joseph Yu , Lei-yu Wu , Yu-ching Chiao & Hsing-shia Tai (2005)
Perceived quality, customer satisfaction, and customer loyalty: the case of lexus in Taiwan, Total
Quality Management & Business Excellence, 16:6, 707-719, DOI: 10.1080/14783360500077393

To link to this article: http://dx.doi.org/10.1080/14783360500077393

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
 Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
Downloaded by [University of Sussex Library] at 20:21 06 February 2015
 Total Quality Management
Vol. 16, No. 6, 707 –719, August 2005

Perceived Quality, Customer

Satisfaction, and Customer Loyalty: the
Case of Lexus in Taiwan
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

CHWO-MING JOSEPH YU, LEI-YU WU, YU-CHING CHIAO &

HSING-SHIA TAI

Department of Business Administration, National Chengchi University, Taipei, Taiwan, People’s Republic
of China,   Department of International Business, National Taipei College of Business, Taipei, Taiwan,

Department of Business Administration, National Changhua University of Education, Taipei, Taiwan

ABSTRACT The purpose of this study is to examine overall customer satisfaction associated with
Toyota’s Lexus automobiles in Taiwan by applying Fornell et al.’s (1996) customer satisfaction
index (CSI) model. By analysing a set of 320 Taiwanese Lexus owners of more than 7 months, the
study uses the structural equation model with LISREL software to show that the perceived quality
has a direct effect on overall customer satisfaction, and has an indirect effect on customer
complaint-levels and customer loyalty. That is, customer expectations have a direct impact on
customer perceived quality; therefore, such expectations have an indirect effect on overall levels
of customer satisfaction. Not surprisingly, overall customer satisfaction negatively influences
customer complaints and positively influences customer loyalty. Importantly, the study also finds
that customer complaints do not have negative impact on customer loyalty. Additionally, the
study demonstrates that the CSI model should be modified in the case of Lexus’ Taiwanese
customers: because the notion of ‘perceived value’ is not present in the existing model, further
exploration is required. Finally, the study finds that firms should do their utmost to improve
perceived quality, as doing so seems to represent the most effective way of increasing levels of
customer satisfaction.

KEY WORDS : Perceived quality, customer expectation, customer satisfaction, customer

complaints, customer loyalty

Introduction
As the features, quality, and appearance of products and services become increasingly
similar (and, by extension, more and more difficult to differentiate), customers increas-
ingly exhibit fickle and non-loyal purchasing behaviours. In response to ever-more

Correspondnce Address: Yu-Ching Chiao, Department of Business Administration, National Changhua

University of Education, 2, Shi-Da Rd., Changhua 500, Taiwan, People’s Republic of China. Email:
chiaoy@cc.ncue.edu.tw

1478-3363 Print=1478-3371 Online=05=060707–13 # 2005 Taylor & Francis

DOI: 10.1080=14783360500077393
 708 C.-M.J. Yu et al.

prevalent ‘price war’ tactics, and the need to maintain a core group of loyal customers,
companies are becoming increasingly aware of the importance of customer satisfaction.
A satisfied customer might well become a repeat buyer; for example, s/he might purchase
multiple automobiles of a certain brand – and keep coming back to a particular dealer for
every service s/he needs through her or his lifetime. Just as negative word-of-mouth leads
to reduced customer satisfaction (Richins, 1983), it follows that satisfied customers
spreading positive word-of-mouth might create new customers for a business.
For the Lexus division of Hotai Motor Co., the sole distributor of Toyota, Lexus and
Hino in Taiwan, there is no doubt that the road to customers’ patronage is a hard one.
Taiwan’s automobile market is, fundamentally, both extremely complex and incredibly
complicated. Whether winning new customers or defending old customers from competi-
tors, a company’s overall market share will be influenced (Lin, 1996). Although Lexus has
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

decided that its strategic thrust will be customer satisfaction, it has no clear benchmarks to
serve as guidelines. Lexus’ main competitors, Mercedes Benz and BMW, have not estab-
lished strong systems in Taiwan for measuring customer satisfaction. In the US, Lexus
dealers such as Sewell Lexus have established such systems with great effectiveness. In
the case of Taiwan, however, the conditions in which the company operates – and there-
fore the goals it must set for itself – differ greatly. Customer satisfaction systems of US or
European dealers may be referred to or studied, but it would be a mistake to treat them as
ideal systems for Taiwan’s Lexus dealers to emulate. Western systems cannot be simply
mimicked in Taiwan.
Technically, ‘Lexus’, as a brand, is not sold in Japan. That is, the products sold under the
Lexus brand name around the world are sold within Japan under the brand name Toyota.
Moreover, Toyota dealers in Japan are segmented according to the kinds of models a
dealer can sell at a particular location. Toyota has multiple groups of dealers, known,
among other names, as Toyota, Toyopet, Toyota Corolla, Nets Toyota, and Toyota
Vista. Each dealer is limited as to which models they can sell.
Although Lexus sell well in Japan, they are not regarded as luxury automobiles in the
way that they are in the rest of the world. Rather, they are treated as higher-end, more
expensive Toyotas. For example, Lexus’ flagship LS 400 model might be driven by an
upwardly-mobile twentysomething – that is, someone who’s not necessarily rich, but
who enjoys fast cars. The reason, perhaps, for the lack of differentiation between
‘Lexus’ and ‘Toyota’ in the mind of Japanese consumers is that ‘Lexus’ has not been
effectively segmented from other Toyota models. In any case, no matter how well the cus-
tomer satisfaction system might work in Japan, Taiwan’s Lexus sellers cannot follow the
Japanese system.
Because Taiwan is a distinct and unique market for Lexus, this study seeks to explore
Lexus’ Taiwanese system of customer satisfaction. Namely, we seek to apply the Custo-
mer Satisfaction Index (CSI) model (Fornell et al., 1996) to current users of Lexus in
Taiwan, and to identify causal relationships leading to overall customer satisfaction in
Taiwan, and possibly in other developing countries as well. The findings should have
implications for firms that hope to improve overall levels of customer satisfaction.

Literature and Hypotheses

In giving reasons for the mounting importance of customer satisfaction, most firms face a
number of problems faced by a variety of industries today: increasing international
 Perceived Quality, Customer Satisfaction, and Customer Loyalty 709

competition, faster economic growth rates, and mature markets. As a result of such cir-
cumstances, firms often find themselves engaged in cut-throat rivalries as they struggle
to win over the same limited pool of customers. Doing so requires the commitment of a
company’s resources; additionally, such competition often puts painful downward
pressure on a firm’s pricing power. Because of the limitations of competing on price
alone, and the costs of pursuing new customers, competing on the basis of quality
(which could become a competitive advantage in its own right), and working to retain
current customers, become more logical and effective strategic methods (Rust & Oliver,
1994). Although customer satisfaction only emerged as a legitimate area of inquiry in
the early 1970s in the US (Churchill & Surprenant, 1982), and companies only truly
began to realize the strategic benefits of quality in the 1990s (Bolton & Drew, 1991),
the achievement of customer satisfaction has already become a cornerstone of smart
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

business practice across a diverse multitude of companies and industries (Szymanski &
Henard, 2001). Thus, the following sections review the literature of customer satisfaction,
including its antecedents (e.g. customer expectation, perceived quality, and perceived
value) and its consequences (e.g. customer loyalty and customer complaints).

Customer Satisfaction Model

Fornell et al. (1996) argued that the nature of the current economic system differs greatly
from that of the past. Therefore, the measurement of economic performance must be
changed so that it takes into account not only the quantity, but also the quality of economic
output. In accordance with this need, the CSI model was introduced in 1996 by Fornell
et al. to evaluate, in the US, the satisfaction-levels of firms, industries, and economic
sectors. The idea was to take more accurate measures of the performance of a modern
economy.
In his earlier study, Fornell (1992) used 15 measurement indicators to evaluate six latent
constructs (e.g. perceived quality, customer expectation, perceived value, overall custo-
mer satisfaction, customer complaints, and customer loyalty); those same measurements
were used to establish questionnaires for telephone surveys. Later, Fornell et al. (1996)
built a CSI model in which customer expectations, perceived quality, and perceived
value comprise the antecedents of overall customer satisfaction. In turn, overall customer
satisfaction affects customer complaints and loyalty (see Figure 1). In short, Fornell’s CSI

Figure 1. Fornell et al.’s (1996) Customer Satisfaction Index (CSI) model

 710 C.-M.J. Yu et al.

system identified and described the causal relationships among these above constructs.
Because the CSI model has such significant implications for managers but has not been
tested in Taiwan, this study both introduces and empirically tests the CSI model on
Lexus users in Taiwan.
On the surface, customer satisfaction seems a straightforward concept. However, its
definition is one that is not fixed, but rather it evolves over time. After reviewing
various related studies, Oliver (1980) concluded that satisfaction is a function of some
initial standard working in combination with some perceived discrepancy from the
initial reference point. To make the idea of an ‘initial standard’ more specific, Bolton &
Drew (1991) defined customer satisfaction or dissatisfaction as a ‘function of the discon-
firmation arising from discrepancies between prior expectations and actual performance.’
Anderson et al. (1994) offered two definitions of customer satisfaction, in accordance with
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

the two broad classes of customer experiences identified by the literature (i.e. transaction-
specific experiences and cumulative experiences). In the case of a transaction-specific
experience, customer satisfaction is defined as the post-choice evaluative judgement of
a specific purchase occasion. In the case of a so-called ‘cumulative’ experience, customer
satisfaction is determined as a result of a customer’s evaluation of his or her total purchase
and consumption experience over time. In their CSI model, Fornell et al. (1996), described
customer satisfaction as a function of customer expectations, perceived quality, and per-
ceived value.

The Antecedents of Customer Satisfaction: Customer Expectations, Perceived Quality,

and Perceived Value
According to Parasuraman et al.’s (1991) findings, four factors impact customers’ expec-
tations. These are word-of-mouth (or reputation), personal needs, past experience, and
external communications (including both the direct and indirect messages communicated
from a company to its customers – i.e. advertising, company brochures, the attitude of the
company’s employees, and the price of the company’s products or services). Other
researchers (e.g. Oliver, 1980) have postulated that customer expectations are influenced
by (1) the product itself, (2) the context in which the product or service is sold, including
the content of communications from salespeople and social referents, and (3) individual
characteristics, including persuasibility and perceptual distortions. Tse & Wilton (1988)
offered three more definitions of customer expectations in their study: (1) equitable per-
formance – a normative standard for performance based on implicit relationships
between an individual’s cost/investment and anticipated rewards. In this case, expec-
tations are likely to be affected by the price paid, the effort invested, and previous
product experiences. (2) Ideal product performance – the optimal product performance;
a consumer’s ideal product-performance scenario. Such expectations may be based on pre-
vious product experiences, messages gleaned from advertisements, or word-of-mouth
communications. (3) Expected product performance – that is, a product’s most likely
performance. These sorts of expectations are affected by the average product performance,
as well as by advertisements. In the CSI model, the determinants of expectations are
customers’ pre-purchase experience of a firm’s product/service, combined with a sort
of forecast of a company’s ability to deliver quality (Fornell et al., 1996). Pre-purchase
experiences include past consumption experiences, as well as non-experiential infor-
mation, such as advertising and word-of-mouth.
 Perceived Quality, Customer Satisfaction, and Customer Loyalty 711

Clearly, different studies have proposed different definitions of customer expectations.

However, the great majority of all such literature has found that the one factor that
consistently and directly affects the perceived quality – and therefore customer satisfac-
tion – is that of the customer’s expectation (Bearden & Teel, 1983; Tse & Wilton, 1988;
Fornell et al., 1996; Halstead et al., 1994). Thus, we adopt the definition of the CSI model,
holding that customer expectations are an antecedent of customer satisfaction. Thus,

H1: Customer expectations will positively affect perceived quality.

H2: Customer expectations will positively affect overall customer satisfaction.

Taking a different point of view, Oliver (1993) made distinct the meanings of ‘satisfac-
tion’ and ‘quality’ by raising four points of difference between the two concepts. First, sat-
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

isfaction may result from any number of causes, and is not necessarily the direct result of a
high-quality product or service experience; quality, on the other hand, is a designation
achieved through the fulfilment of specific attributes. Second, satisfaction is largely the
result of non-quality-related factors, such as needs and equity, whereas expectations of
quality are formed from ideals. Third, dis/satisfaction is felt only after an experience
with a product, while perceived quality does not require an actual experience. Fourth, feel-
ings of satisfaction are influenced by many factors, such as equity, attribution, and
emotion; quality has only a few, specific, antecedents.
Although some researchers have attempted to draw distinctions between customer sat-
isfaction and quality of service, it is fair to say that most of the researchers are not in agree-
ment when it comes to defining the two concepts. Along with this lack of standardization,
differences exist in the perception of the relationship between customer satisfaction and
quality of service. As mentioned earlier, the CSI model does not specifically identify
quality of service; rather, it gives customer satisfaction as a function of customers’ expec-
tations and perceptions of quality and value. However, Bolton & Drew (1991) viewed
customer satisfaction as an antecedent of the quality of service.
It should be noted, however, that while clear differences exist over the definition of
customer satisfaction, as well as over the issue of the relationship between customer sat-
isfaction and quality of service, the fundamental notion of customer satisfaction is in fact
quite similar among the various definitions. Oliver (1993) argued that customers must
have contact experiences of some quality in order to decide whether or not they are satis-
fied. In this light, service quality is akin to the concept of appraisal, and satisfaction
becomes an affective concept. That is, appraisal functions as the antecedent, or cause,
of affective reactions (Oliver, 1999). Likewise, most studies have considered service
quality as the antecedent of customer satisfaction (e.g. Anderson et al., 1994; Chiou
et al., 2002; Cronin et al., 2000; Fornell, 1992; Fornell et al. 1996). In this study, we
adopt most of the arguments and findings of the various prior studies (e.g. Anderson
et al., 1994; Cronin et al., 2000), and take quality of service to be a determinant of custo-
mer satisfaction.

H3: Perceived quality will positively affect overall customer satisfaction.

In contrast to the treatment of customer satisfaction and expectation, a number of

researchers have not, as yet, offered definitions of customers’ perceptions of quality
(Churchill & Surprenant, 1982; Oliver, 1980; Tse & Wilton, 1988). Indeed, some seem
 712 C.-M.J. Yu et al.

to positively avoid defining the term. Not surprisingly, too, those definitions that have been
put forth vary greatly. To wit: Fornell et al. (1996) defined perceived quality as ‘formed
from customization and reliability.’ In this case, customization refers to how well the com-
pany’s product is tailored or customized to satisfy the various needs of a specific customer,
and reliability refers to how reliable, standardized, and free from deficiencies a company’s
product is. Other research has defined perceived quality as ‘. . .the customer’s assessment
of the overall excellence or superiority of the service’ (Zeithaml, 1988). As mentioned
above, some researchers (Churchill & Surprenant, 1982; Oliver, 1980; Tse & Wilton,
1988) have discussed the idea of the perception of quality in their literature, but have
avoided defining the term.
‘Perceived value’, then, is a unique construct; both part of the CSI model and one of the
antecedents of overall customer satisfaction. Fornell and his colleagues defined the con-
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

struct as one that represents the perceived level of product quality relative to the price
of the product (Fornell et al., 1996). Further, they argued that by including perceived
quality as a variable, price information can be incorporated into the model, thereby
increasing the comparability of results across firms, industries, and sectors.
Zeithaml (1988) also recognizes the functionality of the notion of perceived value.
Unlike Fornell, however, she does not treat it as one of the constructs present in the
SERVQUAL model, but rather as a component of perceived quality. Her definition is
‘. . .the customer’s overall assessment of the utility of a product based on perceptions of
what is received and what is given’ (Zeithaml, 1988).
In summary, then, it can be said that on the one hand, Fornell et al. (1996) and Zeithaml
(1988) recognized perceived value as an important factor in the marketing of products and
services. Further, Fornell et al. even incorporate the construct to measure the overall cus-
tomer satisfaction model. On the other hand, both Fornell et al. and Zeithaml considered
customer perceptions of quality to be comprised of perceived quality and perceived value.
(In light of this consideration, it might be reasonable, in the future, to explore whether per-
ceived value really exists as an independent construct influencing customer satisfaction.)
As recognized by the existent literature, however (e.g. Cronin et al., 2000; Zeithaml,
1988), it is no easy task to draw a clear delineation between the two constructs (i.e. per-
ceived value and perceived quality). For this reason, the current study does not propose to
test any hypothesis in terms of perceived value. Instead, we attempt to discuss the role of
perceived value through the use of the structural equation model.

The Consequences of Customer Satisfaction: Customer Loyalty and

Customer Complaints
Few studies have investigated the outcomes of customer satisfaction (Cronin et al., 2000;
Szymanski et al., 2001). However, understanding the outcomes of customer satisfaction,
including (Cronin et al., 2000), consumer loyalty (Bei & Chiao, 2001) and the intention to
continue to do business with a particular provider (Burnham et al., 2003), is of central
importance to marketers. Among such possible outcomes, this study examines both
customer complaints and customer loyalty.
Fewer customer complaints will lead to greater customer loyalty (Tax et al., 1998).
Customers’ complaints to vendors serve as a mechanism for relieving cognitive disso-
nance, venting anger and frustration, and initiating or seeking redress when a consumer
experience is unpleasant or dissatisfying (Nyer, 2000). Although dissatisfied consumers
 Perceived Quality, Customer Satisfaction, and Customer Loyalty 713

have the simple option of discontinuing their patronage of a particular seller, a deeply felt
dissatisfaction can also prompt complaining to the seller. The expectation, then, is that the
more frequent customers’ deep dissatisfaction, the more frequent customers’ complaints to
sellers. Thus, we propose the following hypotheses.

H4: Overall customer satisfaction will negatively affect customer complaints.

H5: Overall customer satisfaction will positively affect customer loyalty.
H6: Customer complaints will negatively affect customer loyalty.

Research Design
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

In order to avoid the potential problems associated with non-users (or students) as respon-
dents (Syzmanski & Henard, 2001), we sampled only actual users of the product. With
support from the Lexus Division of Hotai Motor Co., it was possible to work only with
actual Lexus owners. The sample group comprised individual Lexus owners (versus
corporate owners) who had themselves purchased his/her Lexus and used it for more
than 7 months. The reason for using such parameters was that we postulated that new auto-
mobiles would be less likely to have operational problems. In addition, respondents who
had owned their car for longer periods of time would be more likely to be familiar with the
automobile than new owners. To avoid the very real possibility of a non-user (i.e. secretary
or assistant) filling out the questionnaire instead of the actual user, we did not include cor-
porate owners in our survey. Finally, all qualified respondents were asked to fill out the
questionnaire in order to make the survey more complete.
We sent 879 questionnaires to individual Lexus owners in Taiwan. 320 questionnaires
(36.41%) had been collected one month after the send-date. One of these questionnaires
came back with only the customer’s personal data, rendering it useless. Not surprisingly,
some of the questionnaires contained unanswered questions. However, due to the
limitation of the sample size, those unanswered questions were treated as missing data.
Therefore, statistically analysed data were 319 (36.29%). The final analysed results,
then, represent 36.29% of the total (personal use) Lexus users in Taiwan who have
been using their automobiles for more than 7 months.
The questionnaire contained six parts. Those parts were: customer expectations, per-
ceived quality, perceived value, overall customer satisfaction, customer complaints, and
customer loyalty; there was an additional section for customer’s personal data (see
Table 1 for a description). All the rated questions were measured on a seven-point
scale, with the exception of two questions, which asked for responses in terms of
percentages. The weightings of the seven-point scale were not standardized, and differed
according to each question. In most cases, though, 1 represented the most negative poss-
ible response, while 7 represented the most positive possible response. As reflected in
Table 1, SEM (structural equation model) with LISREL software, a statistical technique
with the ability to deal with multiple indicators and multiple constructs, was used in the
analysis of the data.
For the customer complaint section, a number of questions and much of the content was
expanded (in comparison to the original Fornell et al., 1996, questionnaire) to take into
consideration the fact that there are multiple methods and channels through which a
Lexus owner might voice his/her complaints. In the personal data section, we did not
 714 C.-M.J. Yu et al.

Table 1. Constructs and measurements

Latent constructs Measurement indicators

Perceived quality(h1) (Y1) Overall evaluation of Lexus’ quality experience

(Y2) Evaluation of Lexus’ customization experience
(Y3) Evaluation of Lexus’ reliability experience
Perceived value(h2) (Y4) Lexus’ rating of quality given price
(Y5) Lexus’ rating of price given quality
Overallcustomer (Y6) Overall satisfaction from all experiences with Lexus
satisfaction(h3) (Y7) Expectancy disconfirmation from Lexus
(Y8) Lexus’ performance compared to the customer’s
ideal product or services in the category
Customer complaints (h4) (Y9) Has the customer complained about Lexus or its
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

service and how many times?

(Y10) If your friend wants to buy a car, would you
recommend Lexus?
Customer loyalty(h5) (Y11) In the future, if you want to buy another automobile
or change to another automobile, would you buy
Lexus again?
(Y12) If you are willing to repurchase Lexus, how
important is the price factor?
or
If you are not willing to repurchase Lexus, how related the
price factor to your decision?
(Y13) Price tolerance (increase) given repurchase
or
Price tolerance (decrease) to induce repurchase
Customer expectation (j) (X1) Pre-purchase overall expectation of Lexus’ quality
(X2) Pre-purchase expectation to Lexus regarding customization
(X3) Pre-purchase expectation to Lexus regarding reliability

elect to ask more sensitive or personal questions, such as those regarding education or
income level. In order to heighten the response rate, and to demonstrate our appreciation,
a gift was sent to all those who sent back the questionnaire.

Results and Analyses

Pre-analysis
In the first stage of statistical analysis using LISREL (Jöreskog & Sörbom, 1996), one of
the indicators (Y13) of customer loyalty was found to be less than ideal in the measure-
ment of the construct in question. As the results came out negatively, the indicator was
taken out of the analysis. Thus, the indicator of customer complaints was the only
means (Y9) for measuring customer complaints. In an attempt to run LISREL, the
theta-epsilon (TE) of Y9 could not be identified, thus influencing the entire model.
More specifically, standard errors, t-values, modification indices, and standardized
residuals could not be computed. Thus, the theta-epsilon value of Y9 was fixed
(meaning its value was set to 1).
 Perceived Quality, Customer Satisfaction, and Customer Loyalty 715

In a further attempt to run LISREL with the exact model introduced by Fornell et al. as
in Figure 1, the outcome in terms of our sample profile was far from ideal. The CSI model
showed a simple and absolute disconnection between perceived value and overall custo-
mer satisfaction, indicating that none of the three constructs – perceived quality, customer
expectation, and perceived value – bore any relationship to overall customer satisfaction.
Further, the relationship between some constructs appeared to be insignificant and/or
negative when they should have emerged as positive. In terms of the overall goodness-
of-fit of the model, the results were as follows: x2(df¼82) ¼ 200.00 (p ¼ 0.00),
GFI ¼ 0.92, RMSEA ¼ 0.067, NNFI ¼ 0.90, AGFI ¼ 0.89, NFI ¼ 0.87, where the chi-
square statistic, AGFI, and NFI values are below their recommended values (0.9).
As we had doubted the existence of ‘perceived value’ from the beginning of the
research, and based upon the LISREL results, perceived value was removed from any
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

further analysis. In the CSI model, at least, the existence of perceived value as one of
the antecedents of overall customer satisfaction was proven. Considering that only
Fornell et al. incorporate perceived value as one of the antecedents of customer satisfac-
tion, the reason for the absence of perceived value among Taiwanese Lexus owners may
be that when reflecting on perceived value, they consider it simply as a part of perceived
quality, rather than as a separate construct existing apart from perceived quality.

Model Estimation
The final result of the modified CSI model is expressed as Figure 2, where x2(df¼60) ¼
148.90 (p ¼ 0.00), GFI ¼ 0.93, RMSEA ¼ 0.068, NNFI ¼ 0.91, AGFI ¼ 0.90,
NFI ¼ 0.89. This research employed a Maximum Likelihood method for the estimation
of its parameters. The values of LAMBDA-Y determine whether or not each measurement
indicator can reflect the latent constructs. If the result of a measurement indicator is sig-
nificant, it indicates that the observed indicator can measure latent construct. From the
outputs of the measurement model (i.e. confirmatory factor analysis, CFA), we observed
that the t-values of the quality customization, and reliability of an experience were all sig-
nificant. Therefore, it may reasonably be argued that those three indicators in particular
can reflect the latent construct of perceived quality. As tests measuring overall levels of

Figure 2. Modified CSI model

 716 C.-M.J. Yu et al.

customer satisfaction, the results indicate significance. Therefore, three indicators of

overall customer satisfaction may be considered reflective of the latent construct.
Because there was only one indicator available for the measure of customer complaints,
theta-epsilon was set as 0 and Lambda Y was freed. The t-value indicates that the indicator
is significant; our data show that it did indeed reflect customer complaints. Finally, the
results of the three tested indicators of customer loyalty also appeared to be significant,
thereby confirming that the three indicators of customer loyalty could reflect the latent
construct.
Worth mentioning, too, is the fact that, in the case of customer loyalty, the coefficient
value for the importance of price factor is negative. This result reveals that whether or not
current Lexus owners intended to purchase another Lexus at some future time, price would
not be an important factor in their decision.
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

The values of LAMBDA-X determine whether each of the various measurement indi-
cators does or does not reflect the latent constructs. From the LISREL results, we found
that the t-values of pre-purchase expectations of quality, customization, and reliability
were significant. These results support the contention that those three indicators are effec-
tive in measuring the latent construct (i.e. customer expectations).
Tables 2 and 3 give the estimate of the model and are used to verify the hypotheses.
Referring to Table 2, we see that the t-value of customer expectations to perceived
quality is 3.78, which indicates that the relationship is significant (p , 0.05). Therefore,
hypothesis 1 was supported; that is, customer expectations positively influence
perceptions of quality.
From Table 2, we see that the t-value of customer expectations with respect to overall
customer satisfaction was –0.63, and, thus, that the relationship is insignificant
(p . 0.05). Therefore, hypothesis 2 was not supported. For hypothesis 2, the result indi-
cates that customer expectations do not significantly influence overall customer satisfac-
tion for Lexus owners. Fornell et al. (1996) hypothesize that customer expectation
positively influences overall customer satisfaction (CSI). However, according to
Churchill & Surprenant (1982), ‘. . .an individual’s expectations are: (1) confirmed
when a product performs as expected, (2) negatively disconfirmed when the product per-
forms more poorly than expected, and (3) positively disconfirmed when the product per-
forms better than expected.’ Therefore, the CSI model assumes that customers’
expectations will be positively disconfirmed. On the other hand, Lin’s (1996) research
results show that customer expectations negatively influence customer satisfaction. In
the case of this research, it can be assumed that Lexus owners’ expectations are simply
confirmed by the actual performance of the automobile, and that customers’ expectations
have neither a positive or negative influence on their satisfaction.

Table 2. GAMMA values

Customer expectation

Perceived quality 0.61 (0.16) 3.78

Overall customer satisfaction 20.06 (0.10) 20.63
Customer complaints –
Customer loyalty –

Standard error in parentheses and the third value is t-value.


Significant at a , 0.05.
 Perceived Quality, Customer Satisfaction, and Customer Loyalty 717

Table 3. BETA values

Perceived Overall customer Customer Customer

quality satisfaction complaints loyalty

Perceived quality – – – –
Overall customer 1.10 (0.08) – – –
satisfaction 13.94
Customer complaints – 20.51 (0.07) 27.24 – –
Customer loyalty – 0.93 (0.06) 14.81 0.05 (0.04) 1.12 –

Standard Error in parentheses and the third value is t-value.


Significant at a , 0.05.
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

According to Table 3, the t-value of perceived quality with respect to overall customer
satisfaction was 13.94, and the relationship is significant (p , 0.05). Therefore, hypothesis
3 was supported – that is, perceived quality positively influences overall customer
satisfaction.
For hypothesis 4, the t-value of overall customer satisfaction with respect to customer
complaints was – 7.24 and the relationship is significant (p , 0.05). Therefore, hypothesis
4, which states that overall customer satisfaction negatively influences customer com-
plaints, finds support in our data.
From Table 3, the t-value of overall customer satisfaction with respect to customer
loyalty was 14.81, and the relationship is significant (p , 0.05). Therefore, hypothesis 5
is supported. From this support, we may deduce that overall customer satisfaction posi-
tively influences customer loyalty.
For the final hypothesis, number 6, the t-value of customer complaints to customer
loyalty was 1.12 and the relationship was insignificant (p . 0.05). Therefore, our hypoth-
esis 6, that customer complaints negatively influence customer loyalty, was not supported.
However, t-value was positive, meaning that Lexus is successful in turning unsatisfied
customers into satisfied customers.
The rejection of hypothesis 6 merits further discussion. In the CSI model it is assumed that
customer complaints are negatively associated with customer loyalty; however, in these
empirical results, it was found that customer complaints did not in fact negatively affect
customer loyalty. We argued that when the relationship is not negative, it simply means
that the company is successful in turning customers with complaints into loyal customers.
This indicates that customer complaints are handled with tremendous skill and care at
Lexus. However, the positive relationship between customer complaints and customer
loyalty did not reach statistically significant levels. Further study in examination of this
relationship would help further understanding of this issue. Generally speaking, in Fornell
et al.’s study, all of the hypotheses were found to be true, except hypotheses 2 and 6.

Conclusions
Customer satisfaction leads to profitability (Jacobs et al., 2001; Srivastava et al., 1998).
The purpose of this research is to explore Taiwan Lexus’s system of customer satisfaction.
Applying a survey approach, and based on the estimated results of the modified CSI
model, the following conclusions with respect to the implications of customer satisfaction
can be inferred.
 718 C.-M.J. Yu et al.

First, perceived quality is the only construct that positively and directly influences
overall customer satisfaction. Thus, as the level of perceived quality increases, the level
of overall customer satisfaction also increases.
Second, customer expectations also have a positive effect on overall customer satisfac-
tion. However, they can only indirectly influence overall customer satisfaction by means
of the perception of quality. It was found that customer expectations do not have a signifi-
cant direct effect on overall levels of customer satisfaction. As a consequence, increases in
the level of customer expectations also increase the perceived level of quality, and
increases in the level of perceived quality increase, in turn, the level of overall customer
satisfaction.
Third, different levels of overall customer satisfaction have significantly negative direct
effects on customer complaints or significantly positive direct effects on customer loyalty.
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

Therefore, as overall levels of customer satisfaction escalate, customer complaint-levels

reduce, and levels of customer loyalty improve. As the rate of customer loyalty improves,
it becomes more likely that a customer will again purchase a Lexus, recommend Lexus
automobiles to friends or relatives, and become less sensitive to the price of a Lexus.
Finally, and perhaps most interestingly, we found that the influence of customer com-
plaints on customer loyalty was not negative. This non-negative result indicates that Lexus
is successful in turning complaining customers into loyal customers. In another words,
Lexus does a masterful job of handling customer complaints. It might be said that after
a customer has voiced a complaint with Lexus, the possibility of that customer becoming
satisfied with their overall Lexus experience is better than that of a customer who never
voices a complaint. It is encouraging to know that effectively handling customer com-
plaints enables a firm to turn seeming negatives into positives.

Managerial implications and further research

Our major finding – that perceived quality is the only construct that directly affects overall
levels of customer satisfaction – should have some implications for firms intending to
increase the level of customer satisfaction associated with the sale of durable goods.
Because perceived quality is measured by overall quality, customization, and reliability,
firms can emphasize these indicators in providing services or products to better serve
their customers. For example, customization may include adapting quality, style,
variety, price, prestige, drivability, warranty, or convenience of repair to appeal to
customers’ needs. By examining the findings of this research, ample room for improve-
ment may become evident to various sorts of firms.
As for further directions along this line of inquiry, we propose three possible points of
view for future research based on our findings. First, in sharp contrast to the findings of
US-based research (e.g. Fornell et al., 1996), customer complaints in Taiwan did not
negatively influence customer loyalty in Taiwan. What might be the cause of such a strik-
ing difference? Might it be due to cultural differences, the maturity of customers, or firms’
skill at handling complaints, as we have suggested? Certainly, such inconsistent findings
merit further study. Second, with the goal of more general applicability in mind, the CSI
model might be examined in some other contexts – i.e. in different industries or countries.
Doing so would allow for investigation into the validity of our findings. Finally, the ques-
tion of whether consumers treat perceived value as part and parcel of perceived quality is
also one that merits a great deal more exploration.
 Perceived Quality, Customer Satisfaction, and Customer Loyalty 719

References
Anderson, E.W., Fornell, C. & Lehmann, D.R. (1994) Customer satisfaction, market share, and profitability:
findings from Sweden, Journal of Marketing, 58(3), pp. 53 –66.
Bearden, W.O. & Teel, J.E. (1983) Selected determinants of customer satisfaction and complaint reports, Journal
of Marketing Research, 20(1), pp. 21–28.
Bei, L.T. & Chiao, Y.C. (2001) An integrated model for the effects of perceived product, perceived service
quality, and perceived price fairness on consumer satisfaction and loyalty, Journal of Consumer Satisfaction,
Dissatisfaction and Complaining Behavior, 14, pp. 125–140.
Bolton, R.N. & Drew, J.H. (1991) A multistage model of customers’ assessments of service quality and value,
Journal of Consumer Research, 17(March), pp. 375–384.
Burnham, T.A., Frels, J.K. & Mahajan, V. (2003) Consumer switching costs: a typology, antecedents, and
consequences, Journal of the Academy of Marketing Science, 31(2), pp. 109– 126.
Churchill, Jr., & Surprenant, C. (1982) An investigation into the determinants of Customer Satisfaction, Journal
Downloaded by [University of Sussex Library] at 20:21 06 February 2015

of Marketing Research, 19(November), pp. 491–504.

Cronin, J.J., Jr., Brady, M.K. & Hult, G.T.M. (2000) Assessing the effects of quality, value, and customer
satisfaction on consumer behavioral intentions in service environments, Journal of Retailing, 76(2),
pp. 193 –218.
Fornell, C. (1992) A national customer satisfaction barometer: the Swedish experience, Journal of Marketing,
56(1), pp. 6 –21.
Fornell, C., Johnson, M.D., Anderson, E.W, Cha, J. & Bryant, B.E. (1996) American customer satisfaction index:
nature, purpose, and findings, Journal of Marketing, 60(4), pp. 7 –18.
Halstead, D., Hartman, D. & Schmidt, S.L. (1994) Multisource effects on the satisfaction formation process,
Journal of the Academy of Marketing Science, 22(2), pp. 114 –129.
Jacobs, F.A., Johnston, W. & Kotchetova, N. (2001) Customer profitability, Industrial Marketing Management,
30(4), pp. 353 –363.
Jöreskog, K.G. & Sörbom, D. (1996) LISREL 8: User’s Reference Guide (Chicago, IL: Scientific Software
International).
Lin, Y.Z. (1996) Research of Customer Satisfaction Decision Model and Effects: Substantiation of Personally
Used Small Automobile in Taiwan, PhD Dissertation, Graduate Institute of Commerce, National Taiwan
University.
Nyer, P.U. (2000) An investigation into whether complaining can cause increased consumer satisfaction, Journal
of Consumer Marketing, 17(1), pp. 9– 19.
Oliver, R.L. (1980) A cognitive model of the antecedents and consequence of satisfaction decisions, Journal of
Marketing Research, 17(4), pp. 460– 469.
Oliver, R.L. (1993) Conceptual model of service quality and service satisfaction: compatible goals, different con-
cepts. In: T.A. Swartz, D.E. Bowen & S.W. Brown (Eds) Advances in Service Marketing and Management:
Research and Practice (Greenwich: JAI).
Oliver, R.L. (1999) Whence consumer loyalty? Journal of Marketing, 63(1), pp. 33–44.
Parasuraman, A., Zeithaml, V.A. & Berry, L.L. (1991) Refinement and Reassessment of the SERVQUAL scale,
Journal of Retailing, 67(4), pp. 420–450.
Richins, M.L. (1983) Negative word-of-mouth by dissatisfied customers: a pilot study, Journal of Marketing,
47(1), pp. 68 –78.
Rust, R.T. & Oliver, R.L. (1994) Service Quality: New Directions in Theory and Practice (Thousand Oaks,
California: Sage Publications).
Srivastava, R.K., Shervani, T.A. & Fahey, L. (1998) Market-based assets and shareholder value: A framework for
analysis, Journal of Marketing, 62(1), pp. 2–18.
Szymanski, D.M. & Henard, D.H. (2001) Customer satisfaction: a meta-analysis of the empirical evidence,
Journal of the Academy of Marketing Science, 29(1), pp. 16 –35.
Tax, S., Brown, S. & Chandrashekaran, M. (1998) Customer evaluations of service complaint experiences: impli-
cations for relationship marketing, Journal of Marketing, 60(1), pp. 60–76.
Tse, D.K. & Wilton, P.C. (1988) Models of consumer satisfaction formation: an extension, Journal of Marketing
Research, 25(2), pp. 204 –212.
Zeithaml, V.A. (1988) Consumer perceptions of price, quality and value: A means-end model and synthesis of
evidence, Journal of Marketing, 52(July), pp. 2– 22.