The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0956-4233.htm
IJSIM
19,2
252
Received 26 September 2006
Revised 9 March 2007
Accepted 9 March 2007
International Journal of Service
Industry Management
Vol. 19 No. 2, 2008
pp. 252-270
q Emerald Group Publishing Limited
0956-4233
DOI 10.1108/09564230810869766
REGULAR ARTICLE
Fuzzy importance-performance
analysis for determining
critical service attributes
Wei-Jaw Deng
Graduate School of Business Administration,
Chung Hua University, Hsinchu, Taiwan
Abstract
Purpose – The purpose of this paper is to propose a novel approach of fuzzy importance-performance
analysis (FIPA) to replace conventional importance-performance analysis (IPA) for determining critical
service attributes those really need to improve for achieving superior customer satisfaction.
Design/methodology/approach – First, referring numerous studies, conventional IPA has some
erroneous assumptions, the customer satisfaction of attribute performance has the characteristic of
three-factor theory and the novel approach which integrates natural logarithmic transformation and
partial correlation analysis is feasible for acquiring the implicitly derived importance of attributes.
Second, according the fact and nature of fuzziness in human perception, this study applies fuzzy set
theory to revise conventional IPA. Finally, the FIPA is proposed and subsequently implemented in a
Taiwanese hot spring hotel case study.
Findings – The implementation of FIPA shows the determined critical service attributes are almost
completely different from those attributes acquired by conventional IPA. Hence, the application of
conventional IPA may cause practitioners make incorrect decisions of improvement priorities for
service attributes and direct unsuitable quality-based marketing strategies.
Originality/value – The proposed FIPA which integrates fuzzy set theory, three-factor theory,
partial correlation analysis and natural logarithmic transformation avoids the erroneous assumptions
of conventional IPA, considers the nature of fuzziness in human perception and includes the actual
importance of service attributes. Therefore, the proposed FIPA can effectively assist business
managers in determining critical service attributes to improve service quality or customer satisfaction
and to achieve competitive advantage.
Keywords Fuzzy logic, Customer services quality, Performance measures
Paper type Research paper
1. Introduction
Delivering superior customer value and satisfaction are crucial to the competitive edge of a
firm (Kotler and Armstrong, 2000; Weitz and Jap, 1995). Undoubtedly, service quality and
customer satisfaction are principal drivers of financial performance. Matzler et al. (2004a)
contended that customer satisfaction increases customer loyalty, reduces price sensitivity,
increases cross-buying and increases positive word of mouth. Hansemark and Albinsson
(2004) also noted that customer satisfaction directly influences customer retention and firm
market share. Numerous empirical studies have confirmed the positive correlation between
customer satisfaction and profitability (Anderson et al., 1994; Hallowell, 1996; Johnson et al.,
1996; Eklof et al., 1999; Zeithaml, 2000). Therefore, improving customer satisfaction
is a critical issue for business managers in today’s competitive global marketplace.
With this goal in mind, numerous business managers are continually attempting to Critical service
identify critical service attributes that generate customer satisfaction and loyalty in order attributes
to stay abreast of competitors.
Numerous practitioners and researchers have applied importance-performance
analysis (IPA) to identify the critical performance attributes in customer satisfaction
survey data for products and services (Hawes and Rao, 1985; Yavas and Shemwell,
1997; Tikkanen et al., 2000; Chu and Choi, 2000; Huana et al., 2002; Zhang and Chow, 253
2004; O’Neill and Palmer, 2004; Enright and Newton, 2004). Hansen and Bush (1999)
pointed out that IPA is a simple and effective technique that can assist practitioners in
identifying improvement priorities for service attributes and direct quality-based
marketing strategies. Practitioners apply IPA to analyze two dimensions of service
attributes: performance level (satisfaction); and, importance to customers. Analyses of
these dimension attributes are then integrated into a matrix that helps a firm identify
primary drivers of customer satisfaction and, based on these findings, set improvement
priorities (Matzler et al., 2004a). Hence, following a customer satisfaction survey and
IPA, business managers can make rational decisions about how to best deploy scarce
resources to attain the highest degree of customer satisfaction.
Although IPA is an extremely valuable method, previous studies have several
important shortcomings. For example, Matzler et al. (2004a) noted the original IPA has
two implicit assumptions:
(1) attribute performance and attribute importance are independent variables; and
(2) the relationship between attribute performance and overall performance is
linear and symmetrical.

These assumptions are erroneous in the real world, the relationship between attribute-level
performance and overall customer satisfaction (OCS) is asymmetrical (Kano et al., 1984;
Matzler and Sauerwein, 2002; Ting and Chen, 2002; Matzler et al., 2003, 2004a) and the
relationship between attribute performance and attribute importance is causal (Sampson
and Showalter, 1999; Oh, 2001; Ryan and Huyton, 2002; Matzler et al., 2004a).
Berman (2005) noted that customer delight is not same as customer satisfaction.
Customer delight requires that customer receive a positive surprise that exceeds their
expectations. Berman also mentioned that the must-be, satisfier, and delight attribute
categorization system developed by Kano et al. (1984) is a popular approach for better
understanding customer delight. However, other studies of customer satisfaction have
indicated that satisfaction attributes can be understood using three categories: basic
factors, performance factors, and excitement factors (Brandt, 1988; Johnston, 1995;
Matzler et al., 1996; Oliver, 1997; Matzler and Hinterhuber, 1998; Anderson and Mittal,
2000; Matzler and Sauerwein, 2002). The impact of satisfaction attribute performance
on OCS differs according to category. For example, if delight (excitement) attributes are
not met, customers do not feel dissatisfied. However, if delight (excitement) attributes
are met, the result is customer delight. Therefore, practitioners of IPA must consider
three-factor theory to determine critical service attributes that capable of generating
customer satisfaction, delight and loyalty.
Customer service perceptions are characterized by uncertainty and fuzziness.
Traditional assessments of service quality or customer satisfaction that used a Likert scale
(equal-space crisp number) (Yang et al., 2004; Behara et al., 2002) to represent customer
perceptions based on linguistic assessments (for example, “Very satisfied” ¼ 5,
IJSIM “satisfied” ¼ 4, “fair” ¼ 3, “unsatisfied” ¼ 2, “very unsatisfied” ¼ 1) in survey
19,2 questionnaire are impractical. Human perceptions and attitudes are subjective and
vague. Furthermore, variations in individual perceptions and personality mean that the
same words can indicate very different perceptions (Chiou et al., 2005). Consequently, the
use of binary logic and crisp numbers to describe human perceptions or attitudes fails to
address fuzziness (Zadeh, 1965). Zadeh (1965) noted that fuzzy set theory can deal with
254 problems involving uncertainty and fuzziness. Fuzzy number is considered more
appropriate than crisp number to represent the linguistic term scale about the
customer’s perception of delivered-service (Chien and Tsai, 2000; Wu et al., 2004).
Therefore, practitioners of IPA require a psychometrically valid and practical measure of
attribute performance before determining real critical service attributes.
This study proposes a revised IPA approach that comprises fuzzy set theory,
three-factor theory, partial correlation analysis and natural logarithmic transformation.
The proposed fuzzy importance-performance analysis (FIPA) avoids two shortcomings
of traditional IPA and considers the nature of fuzziness in human perception. The
proposed FIPA which includes the actual importance of service attributes, assists
business managers in determining critical service attributes to improve service quality
or customer satisfaction and to achieve competitive advantage.
The remainder of this paper is organized as follows. Section 2 reviews the relevant
literature particularly that about IPA, three-factor theory of customer satisfaction,
fuzzy set theory and assessment of service attribute’s implicitly derived importance.
To elucidate the real importance of attributes, Section 3 introduces a FIPA approach.
Next, Section 4 demonstrates the implementation of the proposed FIPA approach to
determine critical service attributes and enhance customer satisfaction at a Taiwanese
hot spring hotel. Finally, Section 5 draws conclusions.

2. Literature review
2.1 Importance-performance analysis
IPA has been applied as an effective means of evaluating a firm’s competitive position in
the market, identifying improvement opportunities, and guiding strategic planning efforts
(Martilla and James, 1977; Hawes and Rao, 1985; Myers, 1999). IPA, first introduced
by Martilla and James (1977), identifies which product or service attributes a firm should
focus on to enhance customer satisfaction (Matzler et al., 2004a). Typically, data from
customer satisfaction surveys or service quality surveys (using SERVPERF model
(Cronin and Taylor, 1992)) with pre-consuming measurement of customer attribute
importance are utilized to construct a two-dimensioned matrix. In this matrix,
attribute importance is depicted along the x-axis and attribute performance (satisfaction
or service quality) is depicted along the y-axis. Attribute importance is measured using
some form of self-stated importance (e.g. rating scales, constant sum scales, etc.) or
implicitly derived importance (e.g. multiple regression weights, structural equation
modeling weights or partial correlation weights). The means of performance and
importance, commonly utilized in practice, divide the matrix into four quadrants (Figure 1).
Based on this analysis, particular improvement opportunities are determined.
For example, researchers commonly suggest that major weaknesses (Quadrant IV) should
be top priority and targeted for immediate improvement efforts (Martilla and James, 1977).
Conversely, attributes deemed major strengths (Quadrant I) should be maintained,
leveraged, and heavily promoted (Lambert and Sharma, 1990).
 H Critical service
attributes
"Possible Overkill" "Keep Up the Good Work"
Quadrant II Quadrant I
Performance

255

"Low Priority" "Concentrate Here"

Quadrant III Quadrant IV

Figure 1.
L Importance-performance
analysis
L Importance H

Some studies have modified and extended IPA. However, the basic framework has largely
remained the same (Sampson and Showalter, 1999). For example, O’Leary and Adams
(1982) presented a method for generating importance measures as a composite ranking of
median importance scores and Pearson correlation coefficients. Dolinsky and Caputo
(1991) only surveyed consumers to obtain attribute performance ratings for deriving
importance indicators. Performance scores for attributes were then regressed on scores for
OCS and the standard regression coefficients were used as measures of attribute
importance. A minor variation on this approach is found in basic conjoint analysis that
uses dummy variable regression to derive coefficients for specific attributes levels, and
determines importance as a range of coefficients for each attribute (Liljander and
Strandvik, 1993; Malhotra, 1996). Matzler et al. (2003) propose a revised IPA in which
attribute importance is derived by partial correlation analysis between attribute
performance and OCS. Therefore, two erroneous assumptions of traditional IPA (which
are described in Section 1) had been discussed and criticized in his literature.

2.2 Three-factor theory of customer satisfaction

Kano et al. (1984) developed a model that distinguishes between different quality
attribute types. Kano’s model divides product or service quality attributes into five
distinct categories (attractive, one-dimensional, must-be, indifference, and reverse),
each of which influences customer satisfaction differently. Other studies of customer
satisfaction, however, suggest that service attributes can be understood using three
categories: basic factors, performance factors, and excitement factors (Brandt, 1988;
Johnston, 1995; Matzler et al., 1996; Oliver, 1997; Matzler and Hinterhuber, 1998;
Anderson and Mittal, 2000; Matzler and Sauerwein, 2002). The basic factors are similar
to must-be quality elements. The performance factors are similar to one-dimension
quality elements. The excitement factors are similar to attractive quality elements.
Matzler et al. (2004a) elucidate these three factors. Basic factors (dissatisfiers) are
minimum requirements that produce consumer dissatisfaction when not fulfilled, but
do not result in customer satisfaction when fulfilled or exceeded; that is, negative
IJSIM performance for these attributes has a greater impact on overall satisfaction than
19,2 positive performance. Excitement factors (satisfiers) are attributes that increase
customer satisfaction when delivered, but cause no dissatisfaction when not delivered.
That is, positive performance for these attributes has a stronger influence on overall
consumer satisfaction than negative performance. Performance factors produce
satisfaction when performance is high and dissatisfaction when performance is low.
256 The relationship between service attribute performance and OCS is nonlinear and
asymmetrical for basic and excitement attributes. For performance attributes, the
relationship between service attribute performance and overall satisfaction is linear
and symmetrical (Ting and Chen, 2002; Matzler et al., 2004a).
Consequently, service attributes have two key characteristics in three-factor theory:
(1) Importance of a basic or excitement attribute is based on its performance.
Basic attributes are crucial when performance is low and are unimportant when
performance is high. Excitement factors are critical when performance is high
and are irrelevant when performance is low (Sampson and Showalter, 1999;
Ting and Chen, 2002; Matzler et al., 2004a).
(2) The relationship between attribute performance and OCS is asymmetrical.

Consequently, the applicability of the traditional IPA model that utilizes explicit
customer self-stated importance requires modification.

2.3 Fuzzy set theory

Fuzzy set theory was introduced by Zadeh (1965) to deal with problems involving
uncertainty and fuzziness. The basic definition and concept of fuzzy set theory can be
found on Zadeh (1965), Chen (1996), Tsaur et al. (1997), Chien and Tsai (2000), Hsieh et al.
(2004) and Wu et al. (2004). Numerous studies have applied fuzzy set theory to research
problems involving uncertainty. For example, Chien and Tsai (2000) used fuzzy number
to assess perceived service quality and clarify the strengths and weaknesses of
Taiwanese retail stores. Furthermore, Wu et al. (2004) proposed a fuzzy set theory-based
decision model for determining market position and developing improvement strategy
for hospital service quality. In decision-making research field, fuzzy multiple criteria
decision making (fuzzy MCDM) was introduced to replace traditional MCDM. Numerous
studies regarding fuzzy MCDM have been published (Chiou et al., 2005; Chu and Lai,
2005; Hsieh et al., 2004). In above researches, the qualitative data or linguistic terms used
to represent imprecise assessments of decision criteria or performance attributes are all
expressed using fuzzy number. Consequently, researchers or practitioners should
consider the application of fuzzy set theory into IPA.

2.4 Assessment of service attribute’s implicitly derived importance

Since changes to attribute performance influences the relative importance of attributes,
the self-stated importance explicitly stated by customers for each attribute is not
practically feasible. Implicitly derived importance based on the attribute performance’s
correlation with OCS already includes the characteristic of attribute category in
three-factor theory and is superior to self-stated attribute importance. Thus, Matzler
and Sauerwein (2002) and Matzler et al. (2004a) implicitly derived dimension attribute
importance using a multiple regression analysis with OCS as a dependent variable and
dimension attributes performance as independent variables.
 Recently, Matzler et al. (2004b) noted between the single attribute variables a rather Critical service
strong multicollinearity is to be expected. Therefore, he determines the potential attributes
influence of multicollinearity on regression coefficient estimation. Consequently, he
declares multiple regression analysis is an inappropriate tool for deriving reliable
impact measures when multicollinearity exists within independent variables. As
suggested by Hair et al. (1995), partial correlation analysis is more suitable than
regression analysis for quantifying the influence of independent variables on dependent 257
variables when multicollinearity exists within independent variables. Therefore,
Matzler et al. (2004b) used dichotomized partial correlation analysis with dummy
variables to identify the three factors category of each single attribute.
A regression model utilizing natural logarithmic transformation of independent
variables can capture more diminishing return or sensitivity for independent variables
(Anderson and Sullivan, 1993). Thus, Brandt (1988), Mittal et al. (1998), Anderson and
Mittal (2000) and Ting and Chen (2002) utilize multiple regression analysis with natural
logarithmic dummy variables to determine the asymmetric influence of attribute
performance on OCS.

3. Methodology of the fuzzy importance-performance analysis

3.1 Assignment of triangular fuzzy number to indicate the perceptions of respondents
Generally, surveys examining customer perceptions of satisfaction or service quality have
used questionnaires in which respondents indicate their feelings with reference to selected
linguistic terms. But human judgments of events may vary significantly according to the
subjective perceptions or personality of individuals, even when the same linguistic term is
used (Chiou et al., 2005). Thus, when using fuzzy number to represent specific linguistic
terms, researchers must consider differences among survey respondents.
This study used a triangular fuzzy number to represent the linguistic term of
respondent’s perception of customer satisfaction or service quality. Moreover, the
linguistic terms from among which respondents chose to indicate their perception towards
service were “very satisfied,” “satisfied,” “fair,” “unsatisfied” and “very unsatisfied.” In the
first part in the survey questionnaire of this study, respondents were asked to complete the
question about the range of each linguistic term based on their own subjective decision.
For example, one respondent gave triplets (0, 0, 25), (0, 25, 50), (25, 50, 75), (50, 75, 100) and
(75, 100, 100) meaning “very unsatisfied,” “unsatisfied,” “fair,” “satisfied,” and “very
satisfied,”, respectively, (Figure 2). Another respondent gave triplets (0, 0, 30), (0, 30, 50),
(30, 50, 70), (50, 70, 100) and (70, 100, 100) meaning “very unsatisfied,” “unsatisfied,” “fair,”
“satisfied,” and “very satisfied,” respectively. Finally, this study aggregated respondent
opinions regarding specific linguistic terms by calculating the average triangular fuzzy
VUS US F S VS
f A– (x)

1.0

U Figure 2.
The ith respondent’s
0.0 linguistic term
0 25 50 75 100
IJSIM number for all respondents. Consequently, the final average triangular fuzzy number of
each linguistic term is decided and used for the subsequent assignment of a triangular
19,2 fuzzy number indicating respondent perceptions (Tsaur et al., 1997). The integration
formula is as follows:
!
X n
i X n
ði Þ
Xn
ði Þ
Xn
ði Þ
A~ ak1 ; ak2 ; ak3
k
258 ~ k avg ¼
A i¼1
¼
i¼1 i¼1 i¼1
; i ¼ 1; 2; . . . ; n; k ¼ 1; 2; 3; 4; 5 ð1Þ
n n
where A ~ i is the triangular fuzzy number of kth linguistic term under ith respondent;
k
aðik1Þ ; aðik2Þ and aðik3Þ represent the lower, the moderate and the upper values of the support
of A ~ i , respectively; n denotes the total number of respondent; k denotes the number of
k
linguistic term and there are five linguistic terms in this study, including “very
unsatisfied,” “unsatisfied,” “fair,” “satisfied,” and “very satisfied.”

3.2 Fuzzy number arithmetic and defuzzification for respondent perceptions

After all respondent perceptions are assigned triangular fuzzy number, the necessary
arithmetic and defuzzification can be performed. The necessary triangular fuzzy
number arithmetic and defuzzification are as follows:
.
Average jth attribute performance:
!
Xn
i Xn
ði Þ
X
n
ði Þ
X
n
ði Þ
~
A a ; a ; a
j j1 j2 j3
~ j avg ¼ i¼1 i¼1 i¼1 i¼1
A ¼ ; i ¼ 1;2; . . .; n; j ¼ 1; 2; . .. ;m ð2Þ
n n
~ i is the triangular fuzzy number of jth attribute performance under ith
where A j
respondent; aðij1 Þ , aðij2 Þ and aðij3 Þ represent the lower, the moderate and the upper
values of the support of A ~ i , respectively; n denotes the total number of
j
respondents; m is the total number of attributes.
.
Average OCS:
!
X n X n
ði Þ
Xn
ði Þ
Xn
ði Þ
O~ i
O1 ; O2 ; O3
~ avg ¼ i¼1 i¼1 i¼1 i¼1
O ¼ ; i ¼ 1; 2; . . . ; n ð3Þ
n n
~ i is the triangular fuzzy number of OCS under ith respondent’s perception;
where O
Oði1 Þ , Oði2 Þ and Oði3 Þ represent the lower, the moderate and the upper values of the
support of O ~ i , respectively; n denotes the total number of respondents.
.
Defuzzification of triangular fuzzy number:
As Kaufmann and Gupta (1991), Chen (1996) and Chien and Tsai (2000) noted,
the defuzzification formula for triangular fuzzy number is:
ða1 þ 2a2 þ a3 Þ
V A~ ¼ ð4Þ
4
~ triangular fuzzy number (a1, a2, a3).
where V A~ is the crisp number of A
Crisp numbers derived from the defuzzification of attribute performance and OCS can Critical service
be used to obtain the implicitly derived importance of service attributes and are plotted
in the FIPA matrix.
attributes

3.3 Acquiring the implicitly derived importance of attributes

As previous studies described in Section 2.4, this study presents a novel method for
measuring the implicitly derived importance of attributes that combines partial 259
correlation analysis and natural logarithmic transformation. The proposed method
comprises three steps:
(1) Transform all attribute performance (APij) into a natural logarithmic form:
APij ! lnðAPij Þ i ¼ 1; 2; . . . ; n; j ¼ 1; 2; . . .m ð5Þ
where APij is the crisp number of attribute performance on jth attribute under
ith respondent; n denotes the total number of respondents; m is the total number
of attributes.
(2) Natural logarithmic attribute performance (ln(APij)) and OCS (the crisp number
of OCS) are included in a multivariate normal correlation model as variables
(total m þ 1 variables). Each variable has total n data. In the practical
application via feasible statistical software, participators just enter all ln(APij)
and OCSi in data sheet. Columns denote variables and rows denote respondents.
About the concept of multivariate normal correlation model, participators can
refer the Chapter 15 (Normal correlation models) of Applied Linear Statistical
Models (Neter et al., 1985) or other feasible statistics book.
(3) Perform partial correlation analysis of each ln(APj) with OCS via feasible
statistical software (e.g. SPSS; SAS; STATISTICA). The coefficient of partial
correlation between OCS and ln(APj) is the implicitly derived importance of jth
attribute.
For example, the formula of coefficient of partial correlation between OCS and
ln(AP1) given fixed ln(AP2), ln(AP3), . . . ln(APm) is as follows (Neter et al., 1985):
rOCS lnðAP1 Þ · lnðAP2 ÞlnðAP3 Þ... lnðAPm Þ
sOCS lnðAP1 Þ · lnðAP2 ÞlnðAP3 Þ... lnðAPm Þ ð6Þ
¼
sOCS · lnðAP2 ÞlnðAP3 Þ... lnðAPm Þ · slnðAP1 Þ · lnðAP2 ÞlnðAP3 Þ... lnðAPm Þ
where sOCS lnðAP1 Þ · lnðAP2 ÞlnðAP3 Þ... lnðAPm Þ is the standard deviation of the
conditional joint distribution of OCS and ln(AP1) when ln(AP2), ln(AP3), . . .
ln(APm) are fixed; sOCS · lnðAP2 ÞlnðAP3 Þ... lnðAPm Þ is the standard deviation of the
conditional distribution of OCS when ln(AP2), ln(AP3),. . . ln(APm) are fixed;
slnðAP1 Þ · lnðAP2 ÞlnðAP3 Þ... lnðAPm Þ is the standard deviation of the conditional
distribution of ln(AP1) when ln(AP2), ln(AP3), . . . ln(APm) are fixed.
Because proposed method does not apply common multiple regression analysis to acquire
the implicitly derived importance of service attribute, then it can eliminate the potential
problem of a linear and symmetrical relationship between attribute performance and OCS.
Unlike the self-stated importance which do not consider the influence of service attributes’
present performance, the implicitly derived importance of service attributes derived via
proposed method can consider the relationship between OCS and attribute’s present
IJSIM performance and can include the attribute category characteristic of three-factor theory.
19,2 Furthermore, the potential problem of multicollinearity among independent variables
when employing multiple regression analysis to measure implicitly derived attribute
importance can also be eliminated via partial correlation analysis (Hair et al., 1995).
Notably, the natural logarithmic transformation of attributes’ performance used in this
method also captures more sensitivity for correlation model variables (Anderson and
260 Sullivan, 1993).

3.4 FIPA approach

The implicitly derived importance is calculated using the method proposed in above
subsections and then inputted into the FIPA matrix. The information acquired using
this FIPA is extremely useful for managers for using in relation to customer
satisfaction or service quality improvements. The FIPA comprises seven steps:
(1) Gather customer perception for attribute performance and OCS for focal
services. A questionnaire survey based on linguistic variables is commonly
used for this step. Notably, in the first part of questionnaire, the question about
the range of each linguistic term must be designed.
(2) Assign a final average triangular fuzzy number to customer perceptions of
attribute performance and OCS as mentioned in Section 3.1.
(3) Transform the fuzzy numbers of customer perceptions for attribute
performance and OCS into crisp numbers by performing some necessary
arithmetic and conducting defuzzification as mentioned in Section 3.2.
(4) Obtain the implicitly derived importance of each attribute through natural
logarithmic transformation and partial correlation analysis as mentioned in
Section 3.3.
(5) Use the mean of all implicitly derived degrees of importance for attributes and
the mean of all performance for attributes to divide the FIPA matrix into four
quadrants.
(6) Plot all attributes on the FIPA matrix.
(7) According to the management scheme of each quadrant, FIPA practitioners
determine a reasonable action plan for each attribute in each quadrant.
Particularly, service attributes in Quadrant IV (management scheme action is
“concentrate here”) is the critical service attributes that can be improved for
higher customer satisfaction and competitive advantage. Moreover, the
improvement priority for critical service attributes is the sequence of
implicitly derived importance for attributes. Restated, improvement priority
is based on the degree of attribute importance.

The proposed FIPA approach allows practitioners to consider the nature of fuzziness in
human perception or attitude and the three-factor theory of customer satisfaction. Thus,
an appropriate, effective and reasonable action plan for each attribute can be obtained.

4. Case study
In this section, an example case is presented to demonstrate the implementation of the
proposed FIPA approach to determine critical service attributes that can enhance
customer satisfaction at a Taiwanese hot spring hotel.
4.1 Gather the data of customer perception about focal delivered service Critical service
This section shows the Step 1 of FIPA approach. The questionnaire in this case study attributes
comprised four parts. The first part contained 1 statement about the range of each
linguistic term. The second part contained 20 statements reflecting the dimensionality
of service attribute performance in focal hot spring hotel. To provide a comparison
between IPA and FIPA, the answer column of self-stated raw importance is also design
in second part’s statements. But when the actual application of FIPA is performed, 261
statements of self-stated raw importance are unnecessary. The third part contained
1 statement reflecting the dimensionality of OCS. The fourth part included respondent
demographic information. The scale of answer is five linguistic terms (very satisfied,
satisfied, fair, unsatisfied and very unsatisfied) in parts 2 and 3. The statements of
questionnaire are closed-response questions and are developed based on a review of
hotel customer satisfaction literatures and practical hot spring hotel circumstance
(Miyoung and Haemoon, 1998; Tsang and Qu, 2000; Antony et al., 2004; Lau et al., 2005;
Mohsin and Ryan, 2005).
Customers with who had consumed services from the focal hot spring hotel were
asked to help the survey. Firstly, they fill up the triplet of each linguistic term by their
own subjective decision. Secondly, they rate their degree of satisfaction and self-stated
importance for each attribute. Thirdly, they rate their degree of OCS for focal hotel’s
service. Lately, they provide their demographic information. A total 324 valid
questionnaires were collected for analysis.
To verify the reliability and construct validity of the formal questionnaire before
survey data can use for subsequent analysis, factor analysis was performed using the
defuzzification crisp number of attribute performance to verify the construct validity
and the Cronbach’s a value for each dimension was calculated to verify the reliability.
The factor analysis was based on the principal component analysis with varimax
rotation, eigenvalue exceeding 1 and factor loadings exceeding 0.4. The test value of the
Kaiser-Meyer-Olkin was 0.910, and the p value of the Bartlett’s sphericity test was
almost zero. Furthermore, the cumulative explained variance is 60.297 percent.
Consequently, the construct validity of the questionnaire was quite good (Kaiser, 1974).
The 20 customer satisfaction statements regarding the service of hot spring hotels in
Taiwan were classified into three dimensions, namely “empathy and assurance,”
“responsibility and reliability,” and “tangibility.” Cronbach’s a values for each
dimension of hot spring hotel service satisfaction ranged from 0.8239 to 0.8915 (Table I).
This range demonstrates that the scales of the formal questionnaire have good reliability
(with Cronbach’s a values for each dimension exceeding 0.7) (Nunnally, 1978).

4.2 Triangular fuzzy number assignment, arithmetic and defuzzification

This section shows the Steps 2 and 3 of FIPA approach. Using survey data of the
triplet of each linguistic term, the final average triangular fuzzy number of each
linguistic term is calculated by equation (1) in Section 3.1 and listed in Table II. Use
fifth linguistic term “Very Satisfied” as an example, the A ~ 5 avg is P324 A
~ i =324 and
i¼1 5
equal to (78.0159, 95.9524, 100) with A ~ 1 ¼ ð80; 100; 100Þ; A~ 2 ¼ ð70; 90; 100Þ; . . .
5 5
~ 324 ¼ ð75; 100; 100Þ. Subsequently, perceptions of service attribute performance and
A 5
OCS expressed using linguistic terms are transferred into triangular fuzzy numbers
using the final average triangular fuzzy number for each linguistic term. Finally, fuzzy
IJSIM
Variance
19,2 Extracted Factor explained Cronbach’s
dimension Statement loading Eigenvalue (percent) a

Responsibility 20 Personal warm care given by staff 0.698

and empathy 18 Have customers’ best interest at
262 heart 0.674
17 Easy to get staff’s attention & help 0.659
10 Readiness to respond to customer’s
requests 0.657
15 Knowledgeable to answer
customers’ request 0.647 5.254 26.268 0.8984
14 Courtesy and friendliness of staff 0.646
16 Individual attention for customer 0.620
12 Willingness to help customers 0.620
19 Understand the specific needs of
customers 0.570
13 Provision of safe environment and
equipment 0.532
Reliability 07 Provision of services as promised 0.737
and assurance 08 Dependability in handling
customers’ service problem 0.705
06 Reasonable price 0.679 3.577 17.886 0.8515
11 Prompt reply to customers 0.568
09 Perform service right at the first
time 0.509
Tangibility 01 The physical facilities are visually
appealing 0.839
02 Multiple hot spring facilities 0.760
Table I. 05 Availability of adequate fire &
Results of factor analysis first aids facilities and instructions 0.626 3.229 16.143 0.8407
by using defuzzification 03 Cleanness of hot spring facilities 0.610
crisp number of 04 Convenient hotel location 0.402
attribute’s performance Cumulative variance explained 60.297 percent

Linguistic term-very unsatisfied L M H

Average triplet 0 2.381 21.8254
Linguistic term-unsatisfied L M H
Average triplet 5.5556 24.9206 46.746
Linguistic term-fair L M H
Average triplet 27.6984 49.2857 69.7619
Linguistic term-satisfied L M H
Average triplet 52.619 73.4127 92.2222
Table II. Linguistic term-very satisfied L M H
Average triangular fuzzy Average triplet 78.0159 95.9524 100
number of each
linguistic term Notes: L – the low bound; M – the median; H – the high bound
numbers representing customer perceptions for service attribute performance and OCS Critical service
were transformed into crisp numbers via equations (2)-(4) mentioned in Section 3.2. attributes
Table III shows the average triangular fuzzy number and defuzzification crisp number
for the service attribute performance of focal hotel in columns 3 and 4. To provide a
comparison, average self-stated raw importance and average performance (in Likert
five-point scale) of each attribute of focal hotel are listed in columns 5 and 6 in Table III.
The bottom two rows of Table III list the average triangular fuzzy number and the 263
average defuzzification crisp number of OCS.

4.3 Attribute’s implicitly derived importance and IPA

This section shows the Steps 4-7 of FIPA approach. By using the defuzzification crisp
number of attribute performance and OCS, The implicitly derived importance of
service attributes is acquired by approach described in Section 3.3. After doing natural
logarithmic transformation and partial correlation analysis via SPSS software, the
results of partial correlation coefficient are represented the implicitly derived
importance of service attributes and are listed in column 3 of Table IV. For example,
the 17th customer attribute “Easy to get staff’s attention & help” has a partial
correlation coefficient of 0.244. Figure 3 shows the SPSS output for partial correlation
coefficients for the 17th service attribute. Moreover, column 4 of Table IV lists the
ranking of the implicitly derived importance of service attributes.
Subsequently, traditional IPA matrix was constructed using a Likert scale based
self-stated raw importance and attitude performance to represent the x- and y-axes,
respectively. Mean self-stated raw importance and mean performance of attitude were
used to separate the axes. After plotting 20 attributes’ self-stated raw importance and
attribute performance into IPA matrix. Figure 4 shows the traditional IPA matrix for
the example case. To build FIPA matrix, this study uses total average implicitly
derived importance of attribute and average OCS to separate the x- and y-axes,
respectively, and plot 20 attributes’ implicitly derived importance and performance
into FIPA matrix. Figure 5 shows the FIPA matrix for the example case.
Based on traditional IPA results, business managers should concentrate on
customer attributes 7, 15, and 18. However, analytical results for the FIPA indicate that
business managers should concentrate on customer attributes 6, 17, 18, and 19 which
are the critical service attributes of focal hotel. The improvement priority for these
critical service attributes is 6, 17, 19, and 18. That is, service attribute 6 is the most
important and service attribute 18 is the least important in Quadrant IV (Figure 5).
Therefore, the determination result of critical service attributes is mostly different.
Furthermore, business managers may utilize different management strategies for
the same service attribute. For example, service attribute 17, “Easy to get staff’s
attention & help,” was located in Quadrant III in the traditional IPA matrix and is a
minor weaknesses and do not require additional effort. The management scheme
action is “low priority.” However, based on the FIPA matrix, service attribute 17 was
located in Quadrant IV and is a competitive disadvantage for customer satisfaction and
the management scheme action is “concentrate here.” Service attribute 9, “Perform
service right at the first time,” as another example, was located in Quadrant II in the
traditional IPA matrix and is the minor strength attribute for improving customer
satisfaction. The management scheme action is “possible overkill.” Managers can
decide to redirect business resources to other attributes that require resources.
 19,2

264
IJSIM

Table III.

of attributes
raw importance
Average performance
and average self-stated
Focal hotel’s Focal hotel’s
Focal hotel’s performance performance performance
(in triangular fuzzy (in defuzzification Self-stated (in five-point
No. Service attribute number) crisp number) importance Likert scale)

1 The physical facilities are visually appealing (45.40, 65.86, 83.90) 65.25 4.08 3.69
2 Multiple hot spring facilities (43.33, 63.84, 82.21) 63.30 4.14 3.61
3 Cleanness of hot spring facilities (47.64, 67.92, 84.62) 67.03 4.27 3.78
4 Convenient hotel location (44.40, 65.01, 82.56) 64.25 4.02 3.66
5 Availability of adequate fire & first aids facilities and instructions (46.62, 66.82, 83.57) 65.96 4.09 3.74
6 Reasonable price (39.59, 59.76, 77.07) 59.04 3.94 3.45
7 Provision of services as promised (43.31, 63.22, 80.83) 62.65 4.18 3.58
8 Dependability in handling customers’ service problem (45.33, 65.12, 81.50) 64.27 4.28 3.67
9 Perform service right at the first time (45.43, 65.73, 82.98) 64.97 4.09 3.69
10 Readiness to respond to customer’s requests (43.42, 63.51, 80.65) 62.77 4.08 3.60
11 Prompt reply to customers (38.59, 58.54, 76.37) 58.01 4.08 3.39
12 Willingness to help customers (45.44, 65.56, 82.57) 64.78 4.18 3.68
13 Provision of safe environment and equipment (43.33, 63.58, 81.23) 62.93 4.07 3.60
14 Courtesy and friendliness of staff (48.89, 69.10, 84.83) 67.98 4.30 3.84
15 Knowledgeable to answer customers’ request (42.57, 62.96, 80.50) 62.25 4.21 3.58
16 Individual attention for customer (38.52, 58.72, 76.59) 58.14 3.94 3.40
17 Easy to get staff’s attention & help (41.18, 61.35, 79.24) 60.78 3.96 3.51
18 Have customers’ best interest at heart (41.64, 61.49, 78.92) 60.88 4.18 3.52
19 Understand the specific needs of customers (42.36, 62.77, 80.22) 62.03 4.11 3.57
20 Personal warm care given by staff (45.18, 65.18, 82.19) 64.43 4.27 3.67
Total average (43.61, 63.80, 81.13) 63.09 4.12 3.61
Average OCS (in triangular fuzzy number) (44.99, 65.53, 83.89)
Average OCS (in defuzzification crisp number) 64.99
 Critical service
Implicitly
derived attributes
No. Service attribute importance Ranking

1 The physical facilities are visually appealing 0.006 18

2 Multiple hot spring facilities 0.220 4
3 Cleanness of hot spring facilities 0.006 19 265
4 Convenient hotel location 0.010 17
5 Availability of adequate fire & first aids facilities and instructions 0.001 20
6 Reasonable price 0.259 1
7 Provision of services as promised 0.012 16
8 Dependability in handling customers’ service problem 0.129 8
9 Perform service right at the first time 0.179 6
10 Readiness to respond to customer’s requests 0.052 12
11 Prompt reply to customers 0.071 11
12 Willingness to help customers 0.073 10
13 Provision of safe environment and equipment 0.031 14
14 Courtesy and friendliness of staff 0.232 3
15 Knowledgeable to answer customers’ request 0.020 15
16 Individual attention for customer 0.100 9
17 Easy to get staff’s attention & help 0.244 2
18 Have customers’ best interest at heart 0.133 7 Table IV.
19 Understand the specific needs of customers 0.191 5 Implicitly derived
20 Personal warm care given by staff 0.048 13 importance of service
Total average 0.101 attributes

- - - PARTIAL CORRELATION COEFFICIENTS - - -

Controlling for LOGP1 LOGP2 LOGP3 LOGP4 LOGP5 LOGP6

LOGP7 LOGP8 LOGP9 LOGP10 LOGP11 LOGP12
LOGP13 LOGP14 LOGP15 LOGP16 LOGP18 LOGP19
LOGP20

LOGP17 OCSCRISP

LOGP17 1.0000 0.2441

( 0) ( 304)
P= . P= .010

OCSCRISP 0.2441 1.0000

( 304) ( 0) Figure 3.
P= .010 P= . Partial correlation
coefficient of 17th service
(Coefficient / (D.F.) / 2-tailed Significance) attribute
" . " is printed if a coefficient cannot be computed

However, in the FIPA matrix, service attribute 9 was located in Quadrant I and is an
opportunity for achieving or maintaining competitive advantage and is a major
strength. The management scheme action is “keep up the good work.” That is, the
consequences of final management action would be inappropriate. Thus, managers
must note that traditional IPA did not consider the nature of fuzziness in human
perception and three-factor theory of customer satisfaction. The referential information
IJSIM H
19,2 "possible overkill" "keep up the 14
good work"
3
5

Actual performance
266 4
1 9 12
20 8

3.61 2
10
13 7 15
"low priority" 19

17 18

16 "concentrate here"
L 1
Figure 4. L 4.12 H
Traditional IPA matrix for
attributes Self-Stated Importance
Note: The number in grid is the statement number of questionnaire (see Table III)

H "possible overkill" "keep up the

Fuzzy Set Theory Based Performance

good work" 14
3
5
1 9
20 12
4 8
0.631 2
7 13 10
(%)
19
15
18 17

6
16
11
"low priority" "concentrate here"
L
0.101 L H
Figure 5. Implicitly Derived Importance
FIPA matrix for attributes Notes: 1.The number in grid is the statement number of questionnaire (see Table III);
2.The value of performance is presented in percentage

acquired by traditional IPA can cause managers take incorrect actions when
attempting to improve service quality or customer satisfaction.

5. Conclusions
Traditional IPA was developed as a tool to facilitate prioritization of improvements
and resource allocation. The three-factor theory of customer satisfaction indicates the
existence of a nonlinear relationship between attribute performance (satisfaction) and Critical service
importance; however, this theory creates questions regarding the applicability of IPA attributes
and resulting managerial recommendations. Managers must be aware that changes to
attribute performance (satisfaction) are associated with changes to attribute
importance (Matzler et al., 2003). Typically, managers must work with limited
resources in competitive business environments. Restated, potential service
improvements must be prioritized, with resources allocated to facilitate changes to 267
achieve competitive advantage (Beach and Burns, 1995).
This study presented a FIPA integrating fuzzy set theory, three-factor theory,
partial correlation analysis and natural logarithmic transformation. The importance of
service attributes is implicitly derived via natural logarithmic transformation and
partial correlation analysis. The partial correlation coefficient represents the actual
importance of attribute that had considered the attribute category in three-factor
theory. The application of fuzzy set theory enables practitioners to consider the nature
of fuzziness in human perceptions or attitudes. Furthermore, from the perspective of
workload in questionnaire survey (20 customer satisfaction statements plus 1 OCS
statement in FIPA approach; 20 customer satisfaction statements plus
20 pre-consuming self-stated attributes’ importance statements in IPA approach for
completing analysis), the FIPA approach avoids the task need for measuring the
pre-consuming importance of attributes (almost 50 percent questionnaire survey
workload saving). This unnecessary process is time-consuming for both analysts and
respondents. Consequently, business managers can effectively determine critical
service attributes of focal service and obtain an appropriate action plan for each
critical service attribute via proposed FIPA approach to improve service quality or
customer satisfaction and to achieve competitive advantage.

References
Anderson, E.W. and Mittal, V. (2000), “Strengthening the satisfaction-profit chain”, Journal of
Service Research, Vol. 3 No. 2, pp. 107-20.
Anderson, E.W. and Sullivan, M.W. (1993), “The antecedents and consequences of customer
satisfaction for firms”, Marketing Science, Vol. 12, 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 No. 3, pp. 53-66.
Antony, J., Antony, F.J. and Ghosh, S. (2004), “Evaluating service quality in a UK hotel chain:
a case study”, International Journal of Contemporary Hospitality Management, Vol. 16
No. 6, pp. 380-4.
Beach, L.R. and Burns, L.R. (1995), “The service quality improvement strategy: identifying priorities
for change”, International Journal of Service Industry Management, Vol. 6 No. 5, pp. 5-15.
Behara, R.S., Fisher, W.W. and Lemmink, J. (2002), “Modelling and evaluating service quality
measurement using neural networks”, International Journal of Operations & Production
Management, Vol. 22 Nos 9/10, pp. 1162-85.
Berman, B. (2005), “How to delight your customers”, California Management Review, Vol. 48
No. 1, pp. 129-51.
Brandt, D.R. (1988), “How service marketers can identify value-enhancing service elements”,
The Journal of Service Marketing, Vol. 2 No. 3, pp. 35-41.
Chen, S.M. (1996), “Evaluating weapon systems using fuzzy arithmetic operations”, Fuzzy Sets
and Systems, Vol. 77, pp. 265-76.
IJSIM Chien, C.J. and Tsai, H.H. (2000), “Using fuzzy numbers to evaluate perceived service quality”,
Fuzzy Sets and Systems, Vol. 116, pp. 289-300.
19,2
Chiou, H.K., Tzeng, G.H. and Cheng, D.C. (2005), “Evaluating sustainable fishing development
strategies using fuzzy MCDM approach”, Omega, Vol. 33, pp. 223-34.
Chu, R.K.S. and Choi, T. (2000), “An importance-performance analysis of hotel selection factors in
the Hong Kong hotel industry: a comparison of business and leisure travelers”, Tourism
268 Management, Vol. 21 No. 4, pp. 363-77.
Chu, T.C. and Lai, M.T. (2005), “Selecting distribution centre location using an improved fuzzy
MCDM approach”, The International Journal of Advanced Manufacturing Technology,
Vol. 26 No. 3, pp. 293-9.
Cronin, J. and Taylor, S.A. (1992), “Measuring service quality: a reexamination and extension”,
Journal of Marketing, Vol. 56 No. 3, pp. 55-68.
Dolinsky, A.L. and Caputo, R.K. (1991), “Adding a competitive dimension on
importance-performance analysis: an application to traditional health care systems”,
Health Marketing Quarterly, Vol. 8 Nos 3/4, pp. 61-79.
Eklof, J.A., Hackl, P. and Westlund, A. (1999), “On measuring interactions between customer
satisfaction and financial results”, Total Quality Management, Vol. 10 Nos 4/5, pp. 514-22.
Enright, M.J. and Newton, J. (2004), “Tourism destination competitiveness: a quantitative
approach”, Tourism Management, Vol. 25 No. 6, pp. 777-88.
Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1995), Multivariate Data Analysis,
4th ed., Prentice-Hall, Upper Saddle River, NJ.
Hallowell, R. (1996), “The relationship of customer satisfaction, customer loyalty, and
profitability: an empirical study”, International Journal of Service Industry Management,
Vol. 7 No. 4, pp. 27-42.
Hansemark, O.C. and Albinsson, M. (2004), “Customer satisfaction and retention: the experiences
of individual employees”, Managing Service Quality, Vol. 14 No. 1, pp. 40-57.
Hansen, E. and Bush, R.J. (1999), “Understanding customer quality requirements: model and
application”, Industrial Marketing Management, Vol. 28 No. 2, pp. 119-30.
Hawes, J.M. and Rao, C.P. (1985), “Using importance-performance analysis to develop health care
marketing strategies”, Journal of Health Care Marketing, Vol. 5, pp. 19-25.
Hsieh, T.Y., Lu, S.T. and Tzeng, G.H. (2004), “Fuzzy MCDM approach for planning and design
tenders selection in public office buildings”, International Journal of Project Management,
Vol. 22, pp. 573-84.
Huana, T.C., Beamanb, J. and Shelbyc, L.B. (2002), “Using action-grids in tourism management”,
Tourism Management, Vol. 23 No. 3, pp. 255-64.
Johnston, R. (1995), “The determinants of service quality: satisfiers and dissatisfiers”,
International Journal of Service Industry Management, Vol. 6 No. 5, pp. 53-71.
Johnson, M.D., Nader, G. and Fornell, C. (1996), “Expectations, perceived performance, and
customer satisfaction for a complex service: the case of bank loans”, Journal of Economic
Psychology, Vol. 17 No. 2, pp. 163-82.
Kaiser, H.F. (1974), “An index of factorial simplicity”, Psychometrika, Vol. 39, pp. 31-6.
Kano, N., Seraku, N., Takahashi, F. and Tsuji, S. (1984), “Attractive quality and must-be quality”,
Hinshitsu (Quality, The Journal of Japanese Society for Quality Control), Vol. 14, pp. 39-48.
Kaufmann, A. and Gupta, M.M. (1991), Introduction to Fuzzy Arithmetic Theory and Application,
Van Nostrand Reinhold, New York, NY.
Kotler, P. and Armstrong, G. (2000), Marketing: An Introduction, Prentice-Hall, New York, NY.
Lambert, D.M. and Sharma, A. (1990), “A customer-based competitive analysis for logistics Critical service
decisions”, International Journal of Physical Distribution & Logistics Management, Vol. 20
No. 1, pp. 17-24. attributes
Lau, P.M., Akbar, A.K. and Fie, Y.G. (2005), “Service quality: a study of the luxury hotels in
Malaysia”, Journal of American Academy of Business, Vol. 7 No. 2, pp. 46-55.
Liljander, V. and Strandvik, T. (1993), “Estimating zones of tolerance in perceived service quality
and perceived service value”, International Journal of Service Industry Management, Vol. 4 269
No. 2, pp. 6-28.
Malhotra, N.K. (1996), Marketing Research: An Applied Orientation, Prentice-Hall, Englewood
Cliffs, NJ.
Martilla, J.A. and James, J.C. (1977), “Importance-performance analysis”, Journal of Marketing,
Vol. 41 No. 1, pp. 77-9.
Matzler, K. and Hinterhuber, H.H. (1998), “How to make product development projects more
successful by integrating Kano’s model of customer satisfaction into quality function
deployment”, Technovation, Vol. 18 No. 1, pp. 25-38.
Matzler, K. and Sauerwein, E. (2002), “The factor structure of customer satisfaction: an empirical
test of the importance grid and the penalty-reward-contrast analysis”, International
Journal of Service Industry Management, Vol. 13 No. 4, pp. 314-32.
Matzler, K., Fuchs, M. and Schubert, A.K. (2004a), “Employee satisfaction: does Kano’s model
apply?”, Total Quality Management & Business Excellence, Vol. 15 Nos 9/10, pp. 1179-98.
Matzler, K., Sauerwein, E. and Heischmidt, K.A. (2003), “Importance-performance analysis
revisited: the role of the factor structure of customer satisfaction”, The Service Industries
Journal, Vol. 23 No. 2, pp. 112-29.
Matzler, K., Hinterhube, H.H., Bailom, F. and Sauerwein, E. (1996), “How to delight your
customers”, Journal of Product & Brand Management, Vol. 5 No. 2, pp. 6-18.
Matzler, K., Bailom, F., Hinterhuber, H.H., Renzl, B. and Pichler, J. (2004b), “The asymmetric
relationship between attribute-level performance and overall customer satisfaction:
a reconsideration of the importance-performance analysis”, Industrial Marketing
Management, Vol. 33, pp. 271-7.
Mittal, V., Ross, W.T.R. and Baldasare, P.M. (1998), “The asymmetric impact of negative and
positive attribute-level performance on overall satisfaction and repurchase intentions”,
Journal of Marketing, Vol. 62 No. 1, pp. 33-47.
Miyoung, J. and Haemoon, O. (1998), “Quality function deployment: an extended framework for
service quality and customer satisfaction in the hospitality industry”, International
Journal of Hospitality Management, Vol. 17 No. 4, pp. 375-90.
Mohsin, A. and Ryan, C. (2005), “Service quality assessment of 4-star hotels in Darwin, Northern
Territory, Australia”, Journal of Hospitality & Tourism Management, Vol. 12 No. 1,
pp. 25-36.
Myers, J.H. (1999), Measuring Customer Satisfaction: Hot Buttons and Other Measurement
Issues, American Marketing Association, Chicago, IL.
Neter, J., Wasserman, W. and Kutner, M.H. (1985), Applied Linear Statistical Models, Irwin,
Homewood, IL.
Nunnally, J.C. (1978), Psychometric Theory, McGraw-Hill, New York, NY.
Oh, H. (2001), “Revisiting importance-performance analysis”, Tourism Management, Vol. 22
No. 6, pp. 617-27.
IJSIM O’Leary, J.T. and Adams, M.B. (1982), Community Views Concerning Urban Forest Recreation
Resources, Facilities and Services, US Forest Service, North Central Forest Experiment
19,2 Station, Chicago, IL.
Oliver, R.L. (1997), Customer Satisfaction: A Behavioral Perspective on the Consumer,
McGraw-Hill, New York, NY.
O’Neill, M.A. and Palmer, A. (2004), “Importance-performance analysis: a useful tool for directing
270 continuous quality improvement in higher education”, Quality Assurance in Education,
Vol. 12 No. 1, pp. 39-52.
Ryan, C. and Huyton, J. (2002), “Tourists and aboriginal people”, Annals of Tourism Research,
Vol. 29 No. 3, pp. 631-47.
Sampson, S.E. and Showalter, M.J. (1999), “The performance-importance response function:
observations and implications”, The Service Industries Journal, Vol. 19 No. 3, pp. 1-25.
Tikkanen, H., Alajoutsijarvi, K. and Tahtinen, J. (2000), “The concept of satisfaction in industrial
markets: a contextual perspective and a case study from the software industry”, Industrial
Marketing Management, Vol. 29 No. 4, pp. 373-86.
Ting, S.C. and Chen, C.N. (2002), “The asymmetrical and non-linear effects of store quality
attributes on customer satisfaction”, Total Quality Management, Vol. 13 No. 4, pp. 547-69.
Tsang, N. and Qu, H. (2000), “Service quality in China’s hotel industry: a perspective from
tourists and hotel managers”, International Journal of Contemporary Hospitality
Management, Vol. 12 No. 5, pp. 316-26.
Tsaur, S.H., Tzeng, G.H. and Wang, K.C. (1997), “Evaluating tourist risks from fuzzy
perspectives”, Annals of Tourism Research, Vol. 24 No. 4, pp. 796-812.
Weitz, B.A. and Jap, S.D. (1995), “Relationship marketing and distribution channels”, Journal of
the Academy of Marketing Science, Vol. 23, pp. 305-20.
Wu, W.Y., Hsiao, S.W. and Kuo, H.P. (2004), “Fuzzy set theory based decision model for
determining market position and developing strategy for hospital service quality”, Total
Quality Management, Vol. 15 No. 4, pp. 439-56.
Yang, Z., Jun, M. and Peterson, R.T. (2004), “Measuring customer perceived online service
quality: scale development and managerial implications”, International Journal of
Operations & Production Management, Vol. 24 Nos 11/12, pp. 1149-74.
Yavas, U. and Shemwell, D.J. (1997), “Analyzing a bank’s competitive position and appropriate
strategy”, Journal of Retail Banking Services, Vol. 19 No. 4, pp. 43-51.
Zadeh, L.A. (1965), “Fuzzy sets”, Information and Control, Vol. 8, pp. 338-53.
Zeithaml, V.A. (2000), “Service quality, profitability, and the economic worth of customers: what
we know and what we need to learn”, Journal of the Academy of Marketing Science, Vol. 28
No. 1, pp. 67-85.
Zhang, H.Q. and Chow, I. (2004), “Application of importance-performance model in tour guides’
performance: evidence from mainland Chinese outbound visitors in Hong Kong”, Tourism
Management, Vol. 25 No. 1, pp. 81-91.

Corresponding author
Wei-Jaw Deng can be contacted at: simond@chu.edu.tw

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints