Transport Policy 18 (2011) 318325

Contents lists available at ScienceDirect

Transport Policy
journal homepage: www.elsevier.com/locate/tranpol

Behavioral intentions of public transit passengersThe roles of service

quality, perceived value, satisfaction and involvement
Wen-Tai Lai a, Ching-Fu Chen b,n
a
b

Department of International Trade & Business Administration, Wenzao Ursuline College of Languages, 900 Mintsu 1st Road, Kaohsiung 80793, Taiwan
Department of Transportation & Communication Management Science, National Cheng Kung University, 1, Ta-Hsueh Rd., Tainan 701, Taiwan

a r t i c l e in f o

a b s t r a c t

Available online 28 September 2010

Understanding the behavioral intentions of public transit passengers is important, because customer
loyalty is seen as a prime determinant of long-term nancial performance. This study highlights such
behavioral intentions and explores the relationships between passenger behavioral intentions and the
various factors that affect them. Apart from the factors recognized by past studies, such as service
quality, perceived value, and satisfaction, this study addresses the importance of the involvement of
public transit services in passenger behavioral intentions. By using passenger survey data from the
Kaohsiung Mass Rapid Transit (KMRT), a newly operating public transit system in Taiwan, we apply the
structural equation modeling technique to analyze the conceptualized relationship model. The ndings
reveal that all causal relationships are statistically signicant. Managerial implications are discussed.
& 2010 Elsevier Ltd. All rights reserved.

Keywords:
Involvement
Service quality
Perceived value
Satisfaction
Behavioral intentions
Public transit

1. Introduction
Reducing private transport use and increasing that of public
transport are two critical but challenging tasks for dealing with
the problem of urban transport. Public transport systems need
to become more market oriented and competitive, as they tend to
be viewed as service products. Along with the trend of the
privatization of public transport services over the past decade, the
nancial performance of private service providers has been a key
concern. Understanding the loyalty of public transit passengers is
important, because customer loyalty is seen as a prime determinant of a rms long-term nancial performance and is considered
a major source of competitive advantage (Lam et al., 2004). For
public transit managers, especially for a newly operating public
transit system like the Kaohsiung Mass Rapid Transit System
(KMRT) studied in this paper, understanding passengers behavioral intentions after experiencing the public transit services and
their associated inuential factors is also an essential task. This
information can help public transit managers and marketers
design effective strategies to meet passengers needs, and thus
retain existing passengers as well as attract new ones from other
modes.
Favorable behavioral intentions encompass three dimensions:
word-of-mouth, purchase intentions and price insensitivity, and
together form the customer loyalty. Satisfaction is generally seen

Corresponding author. Tel.: + 886 6 2757575x53230; fax: +886 6 2753882.

E-mail addresses: [email protected] (W.-T. Lai),
[email protected] (C.-F. Chen).
0967-070X/$ - see front matter & 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tranpol.2010.09.003

as the main driver of consumer loyalty and behavior (Olsen,

2007). In studies based on the Theory of Planned Behavior (TPB),
customer satisfaction has been widely identied as the most
important determinant of favorable behavioral intentions. Customer satisfaction is closed linked with perceived service quality
and value (Chen, 2008; Jen and Hu, 2003; Petrick, 2004). Travelers
who perceive good quality of public transit service are thus more
likely to have a higher level of perceived value and satisfaction,
and so continue to use this service.
Apart from satisfaction, involvement, a widespread concept in
both the marketing domain and behavioral research, describes the
strong attitudes that predict or explain behavior (Olsen, 2007).
Involvement is dened as the level of interest or importance that
an object has for an individual, or the centrality of an object to an
individuals ego structure (Zaichkowsky, 1994). The level of
involvement inuences the level of decision importance in a
purchasing process, such as consumers cognitive and behavioral
responses, and customers are likely to display attitudinal loyalty
for high-involvement purchases. Surprisingly, involvement has
only rarely been applied in the transport service literature.
To address this gap in the current literature, this research aims
to develop a relationship model that incorporates the main
determinants of public transit passengers behavioral intentions
and explores their effects on behavioral intentions. Specically,
the roles of service attitudes (i.e. service quality, perceived value
and satisfaction) and involvement that are involved in the
formation of passenger behavioral intentions in the public transit
service context are addressed in this study. The remainder of this
paper is organized as follows. Section 2 presents the theoretical
background and hypotheses of the proposed model. Section 3

W.-T. Lai, C.-F. Chen / Transport Policy 18 (2011) 318325

describes the methodology, followed by empirical results

reported in Section 4. Finally, a discussion of the results and a
conclusion are provided in Section 5.

2. Theoretical background and hypotheses

Behavioral intentions, as an afrmed likelihood to engage in a
certain behavior, are important indicators of customers future
behaviors. According to the Theory of Planned Behavior (TPB),
behavioral intentions trigger future behaviors (Ajzen and Fishbein,
1980). Favorable behavioral intentions lead to customer loyalty,
which is dened as a deeply held commitment to repurchase or
repatronise a preferred product or service in the future (Oliver,
1997). Behavioral intentions can be viewed as signals that show
whether a customer will continue to utilize a companys services or
switch to a different provider (Zeithaml et al., 1996). Therefore,
by better understanding the passenger behavior marketers and
managers of transit systems will be better equipped to develop
more appropriate marketing strategies and tailor their products
and services to attract new passengers, as well as retain existing
ones.
Two behaviors, i.e. repeatedly patronizing the service provider
and recommending a service provider to others, are viewed as key
manifestations of customer loyalty (Fornell, 1992). Most researchers tend to employ the behavioral-intention measure to represent
customer loyalty, since action loyalty is difcult to observe and is
often equally difcult to measure (Yang and Peterson, 2004). The
current study thus employs the behavioral-intention measure of
customer loyalty, which includes repurchase and recommendation intentions.
Factors such as service quality, consumer satisfaction, and
value are frequently viewed as key building blocks of customer
loyalty (Chen, 2008). A higher quality of service and levels of
satisfaction can result in positive behavioral intentions and
ultimately loyalty. Some studies suggest that customer satisfaction
should be used in conjunction with perceived value, and perceived
value may be a better predictor of repurchase intentions
than satisfaction (Cronin et al., 2000; Petrick, 2004; Petrick and
Backman, 2002; Woodruff, 1997). Hence, service quality, perceived
value, and satisfaction are adopted in this study to examine their
inuences on behavioral intentions. In addition, this study also
hypothesizes that involvement, as mentioned above, is an antecedent of customer loyalty.
2.1. Service quality
Service quality is a measure of how well the service level
delivered matches customer expectations, while a rm delivering
quality service means conforming to customer expectations on a
consistent basis (Joewono and Kubota, 2007; Transportation
Research Board, 1999, 2004). Therefore, the extent to which
service performance matches customer expectations inuences to
the degree which customers feel satised. The SERVQUAL scale
proposed by Parasuraman et al. (1985, 1988) based on the
disconrmation of expectations model (Oliver, 1980) is widely
used to measure service quality. This scale contains 22 items
under ve dimensions including tangibility, reliability, responsiveness, assurance, and empathy. However, these ve dimensions and 22 associated items have been challenged as being
inappropriate for some service businesses (Cronin and Taylor,
1992). In practice, appropriate modications are thus generally
needed in order to reect the specic characteristics of the service
context being studied.
Hu and Jen (2006) proposed a scale of bus service quality with
20 items under four dimensions interactions with passengers,

319

tangible service equipment, convenience of service, and operating

management support and undertook a survey of bus services in
Taipei, Taiwan. Interaction with passengers refers to the respect
and care passengers feel when interacting with service providers,
and how they respond to passengers problems. Tangible service
equipment relates to the level of comfort of the facilities and
equipment operated by the service providers. Convenience of
service concerns accessibility, the information provided, and the
convenience of the service network. Finally, operating management support pertains to elements such as bus schedules, service
periods, and the number of staff (Hu and Jen, 2006).
Joewono and Kubota (2007) measured the service quality of
Indonesian paratransit systems using nine factors with 54
attributes. The nine factors are as follows: availability, accessibility, reliability, information, customer service, comfort, safety,
fare, and environmental impact. They explored user-perceived
service quality and overall satisfaction with the paratransit
service in order to make forecasts with regard to competition
from motor vehicles in Bandung, Indonesia. The results showed
that service quality has positive effects on both overall satisfaction and customer loyalty, and overall satisfaction has a positive
impact on customer loyalty. In their study on measuring the
service quality of Greek public transit systems, Tyrinopoulos and
Antonious (2008) used a total of 23 attributes classied into four
categories, including general characteristics of the public transport system, terminals & stops, vehicles, and transport points,
based on the Handbook for Measuring Customer Satisfaction and
Service Quality (Transportation Research Board, 1999). In the
current study, the service attributes are mainly based on those in
Tyrinopoulos and Antonious (2008), and amended according to
the specic service characteristics of the Kaohsiung Mass Rapid
Transit System (KMRT).
2.2. Perceived value
Perceived value pertains to a consumers overall assessment
of the utility of a product (or service) based on perceptions of
what is received and what is given (Zeithaml, 1988). More
specically, perceived value comes from a trade-off between
perceived benets and perceived costs (Lovelock, 2000). Previous
studies have suggested that perceived value may be a better
predictor of repurchase intentions than either satisfaction or
quality (Cronin et al., 2000; Petrick, 2004; Petrick and Backman,
2002; Woodruff, 1997). Perceived value has been identied as an
antecedent to satisfaction and behavioral intentions (Chen, 2008,
Cronin et al., 2000; Petrick, 2004; Petrick and Backman, 2002). In
addition, many studies have concluded that service quality
positively affects perceived value (Cronin et al., 2000; Zeithaml,
1988).
In the public transport context, Jen and Hu (2003) established
and tested the perceived value model, which they applied to
identify factors affecting passengers repurchase intentions toward public transit services. Their results revealed that passengers behavioral intentions are signicantly affected by perceived
value, which is determined by perceived benets and perceived
costs. Moreover, service quality is found to have a positive effect
on perceived value.
2.3. Satisfaction
Satisfaction is an overall affective response to a perceived
discrepancy between prior expectations and perceived performance after consumption (Oliver, 1980, 1999). There are two
general conceptualizations of satisfaction in the literature:
transaction-specic satisfaction (individual level) and cumulative

320

W.-T. Lai, C.-F. Chen / Transport Policy 18 (2011) 318325

satisfaction (customers total consumption experience) (Johnson

et al., 1995). In practice, service quality and satisfaction are often
used interchangeably, but differences between service quality
and customer satisfaction are also claried in the literature.
For example, Oliver (1997) suggests that service quality judgments
are more specic while customer satisfaction judgments are
more holistic. Service quality is also related to cognitive
judgments while customer satisfaction is associated with affective
judgments.
This study refers service quality to transaction-specic
satisfaction, which concerns the (dis)conrmation in the expectance gap of individual attributes, while cumulative satisfaction is
conceptualized as overall satisfaction. In other words, overall
satisfaction pertains to a holistic evaluation after a service
delivery experience, and acts as a consequence of satisfaction
with individual attributes (i.e. service quality) (Chen, 2008;
Fornell, 1992). Past studies have shown evidence that customer
satisfaction signicantly affects customer loyalty and behavior
intentions in various industries (Anderson and Sullivan, 1993;
Cronin and Taylor, 1992; Fornell, 1992; Oliver, 1980; Petrick and
Backman, 2002), including public transit services (Joewono and
Kubota, 2007; Nathanail, 2008).

2.4. Involvement
The construct of involvement is crucial when considering an
individuals purchasing process and is addressed by attitudinal and
behavioral theories when tackling the issues of brand loyalty and
purchase intentions (Bennett et al., 2005). Involvement is intended to
cover an individuals subjective sense of the concern, care, importance, personal relevance, and signicance attached to an attitude, a
persons motivational state of mind with regard to an object or
activity, or the mobilization of behavioral resources for the achievement of relevant goals (Olsen, 2007). Involvement refers to a persons
feelings about the relevance or importance of an object based on their
inherent needs, values, and interests (Zaichkowsky, 1985).
Since an individual can be involved with advertisements,
products, or purchasing decisions, the concept of involvement
with these different objects leads to various responses. Involvement with products has been hypothesized to lead to a greater
perception of attribute differences, a perception of greater
product importance, and a greater commitment to brand choice.
Involvement with purchases leads to searches for additional
information and more time spent searching for the right selection.
For example, involvement with advertising can lead an individual
to give more counterarguments to the messages it contains
(Zaichkowsky, 1985). Zaichkowsky (1985) argued that a measure
of involvement which can be used across various research studies
might include three categories: personal (i.e. the inherent
interests, values, or needs that motivate an individuals feelings
toward the object), physical (i.e. the characteristics of the object
that cause differentiation and increase interest), and situational
(i.e. factors that temporarily increase or decrease the relevance of
or interest in the object). Evidence that these three factors
inuence a consumers level of involvement or response to
products, advertising, and purchasing decisions has been widely
found in the literature. In the public transport context, Lai and
Lu (2007) relied on Zaichkowsky (1985) to develop a public
transport involvement scale covering personal (e.g. connection
with public transport), product (e.g. fare and performance) and
situational factors (e.g. time pressure and weather conditions) to
measure Taiwanese travelers involvement in public transport
services. Public transit involvement is dened as the level of
interest or importance of public transit to a passenger in the
current study.

Involvement is found to affect the decision process of a

consumers purchasing behavior, and consumers are more likely
to display attitudinal loyalty for high-involvement purchases
(Bennett et al., 2005; Chen and Tsai, 2008; Tsiotsou, 2006).
Previous studies suggest a positive relationship between product
involvement and service/brand loyalty (Beatty et al., 1988;
Pritchard et al., 1999). Consumer involvement is also found to
have direct effects on perceived product (service) quality
(Tsiotsou, 2006), the level of satisfaction (Richins and Bloch,
1991), and therefore perceived value. In other words, the level of
involvement that a consumer has with respect to the object of
interest works as an important determinant of consumer evaluations and behaviors (Chen and Tsai, 2008).
It is noteworthy that involvement is also suggested to be a
moderator variable in many studies of attitudebehavioral
relationships and within other aspects of marketing research
(Olsen, 2007). Some studies also investigate the moderation effect
of involvement on the satisfactionloyalty relationships (Bloemer
and de Ruyter, 1998). In fact, involvement can be changed due to
changes in customer attitudes towards services and perceptions
of service evaluations, such as perceived value, service quality,
and satisfaction. In the above mentioned studies, it is the factor of
modied involvement after customers service experiences that is
used as a moderator.
Based on the review of the literature, the current study
proposes the conceptual model shown in Fig. 1. The hypotheses
are stated as follows:












H1: Service quality has a positive effect on perceived value.

H2: Service quality has a positive effect on overall satisfaction.
H3: Service quality has a positive effect on involvement.
H4: Service quality has a positive effect on behavioral intentions.
H5: Perceived value has a positive effect on overall satisfaction.
H6: Perceived value has a positive effect on involvement.
H7: Perceived value has a positive effect on behavioral
intentions.
H8: Overall satisfaction has a positive effect on involvement.
H9: Overall satisfaction has a positive effect on behavioral
intentions.
H10: Involvement has a positive effect on behavioral intentions.

Each of the model constructs is dened as follows:

Behavioral intention: the public transit passengers judgment
about the likelihood of continuing to use public transit or
willingness to recommend it to others.
Satisfaction: the extent of overall pleasure or contentment felt
by the public transit passenger, resulting from the ability of the

H4
H2

Service
quality

H1

Satisfaction

H9

Behavioral
intention

H3
H8

H5
Perceived
value

Involvement
H6
H7
Fig. 1. The conceptual model.

H10

W.-T. Lai, C.-F. Chen / Transport Policy 18 (2011) 318325

3.1. Kaohsiung Mass Rapid Transit System (KMRT)

Before the introduction of the KMRT, public transport services
were poorly provided in Kaohsiung city, and private cars and
motorcycles were the main transport modes for the commuters.
The KMRT was constructed and is operated by the Kaohsiung Rapid
Transit Corporation (KRTC) under a build-operate-transfer (BOT)
contract through public and private cooperation, in an approach
that is intended to benet both public and private sectors. The
system, offering a network of two lines (see Fig. 2), was developed
to increase the quality and coverage ratio of public transportation
in Kaohsiung and to minimize noise and air pollution in the urban
area. The total length of the current network, i.e. Orange and Red
Lines, is 42.7 km with 38 stations, of which 28 are underground,
two at ground level, and eight elevated. The Red line (northsouth
bound) and Orange line (westeast bound) started operating in
March and September 2008, respectively. The frequency of train
services is every 6 min during peak periods and every 10 min
during off-peak ones. Fig. 3 depicts the ridership of the KMRT for
the Red and Orange lines. The total ridership was around 141,000
passengers per day at the end of January 2009, which was still far
behind the expected ridership of 450,000 passengers per day that

140,000
120,000

Red Line
Orange Line
Total

100,000
80,000
60,000
40,000
20,000
0

04
/2
00
05 8
/2
00
06 8
/2
00
07 8
/2
00
08 8
/2
00
09 8
/2
00
10 8
/2
00
11 8
/2
00
12 8
/2
00
01 8
/2
00
9

3. Methods

160,000

Ridership (passenger)

trip experience to fulll the passengers relevant desires, expectations, and needs.
Perceived value: the public transit passengers overall appraisal
of the value of the service provided, based on their assessment of
what is received (benets) and what is given (costs or sacrice).
Service quality: the public transit passengers assessment of the
standard of the service delivered.
Involvement: the level of interest in or importance of public
transit to a passenger.

321

Average (day)
Fig. 3. KMRT ridership.

art need to break-even (Mass Rapid Transit Bureau, 2009).

Therefore, marketing and promotion strategies to increase ridership are of concerns for the KMRT.
3.2. Measures
A self-administered questionnaire survey of KMRT passengers
in Kaohsiung City was conducted to collect empirical data for this
study. The questionnaire was designed based upon a review of the
related literature. To have adequate content validity, the survey
instrument was revised and nalized according to feedback from
two transport management professors and a pre-test sample of 25
postgraduate students studying a transport management program
at a university in southern Taiwan. The questionnaire examined
service quality (19 items; example item: I feel the service
frequency of the KMRT is good.), perceived value (three items,
example item: Taking the KMRT is worth it compared with the
money I spent), overall satisfaction (a single item: Overall, I am
satised with the service the KMRT provided), public transit
involvement (nine items; example item: I always pay attention
to the information about public transit), behavioral intentions
(two items: I am willing to recommend the KMRT to others and
I am willing to keep on taking the KMRT in the future), and the
respondents demographic and travel behavior information. Apart
from respondents demographic and travel behavior, which were
measured by categorical scales, the items of all the other
constructs are measured using a ve-point Likert scale ranging
from strongly disagree (1) to strongly agree (5).
3.3. Data collection

Fig. 2. KMRT network (Source: http://mtbu.kcg.gov.tw).

An on-site survey was conducted at KMRT stations in

Kaohsiung City on both weekdays and weekends in October
2008. The convenience sampling technique was employed due to
the fact that the population of public transit passengers was not
available. Passengers were asked about their willingness to take
part in the survey, and if they answered yes then they were asked
to complete a pencil and paper questionnaire under the guidance
of the data collectors. Eight hundred questionnaires were
distributed and 763 usable ones were obtained, and after deleting
those with incomplete responses there was an effective response
rate of 95.4%. Among the usable samples, 60.8% respondents were
female and 39.2% are male. Nearly half of the respondents (47.1%)
had a monthly income of less than NT$20,000 (roughly US$600),
followed by 32.8% with NT$ 20,00140,000 (US$6011200).

322

W.-T. Lai, C.-F. Chen / Transport Policy 18 (2011) 318325

Forty-ve percent were aged between 25 and 34, and 30.9% were
between 15 and 24. Students, housewives, and civil servants
accounted for 53.5% of the sample.

tested using structural equation modeling (SEM) in the second

stage by using LISREL 8.0 (Joreskog

and Sorbom,
1996). All
analyses were based on the correlation matrix of indicators of
interest.

3.4. Data analysis

4. Empirical results
The data analysis was conducted in two stages. First,
exploratory factor analyses using principal component with
VARIMAX rotation technique were performed to examine construct dimensionalities of both public transport involvement and
service quality. In this study, a factor was retained only if it had an
eigenvalue, i.e. the amount of variance accounted for a factor,
greater than 1.0. The items under each factor were retained
only if they had factor loadings greater than 0.5. To assess the
reliability of measures, Cronbachs Alpha was calculated to
examine the reliability of variables retained in each factor,
and coefcients greater than 0.5 were considered acceptable,
indicating a good construct reliability (Hair et al., 2006). On that
basis, the relationships of involvement, service quality, perceived
value, satisfaction, and behavioral intentions were empirically

4.1. Dimensionality of service quality and involvement

This study employs a multi-attribute approach to measure
public transport involvement and service quality. The results from
the principal component factor analysis, a uni-dimensional
solution with nine involvement items explaining 58.8% of total
variance, are shown in Table 1. The involvement factor has an
eigenvalue of 5.26 and reliability value (a) of 0.91. A similar
procedure is applied to service quality. Two factors with
eigenvalues greater than one explain 55.6% of the total variance
for the service quality scale (see Table 2). Note that three items
are removed from the scale as their factor loadings are less than
0.5. According to the varimax-rotated factor pattern, the rst

Table 1
Factor analysis of involvement.
Factor/item

Factor loading

Involvement (INV)
INV5: Taking public transit is important to my daily life
INV6: I feel that taking public transit is consistent with my lifestyle
INV3: No matter what the trip purpose is, I always prefer taking public transit
INV2: Whether the weather is good or bad, I always prefer taking public transit
INV7: I like others to know the fact that I take public transit
INV1: Whether there is time pressure or not, I always prefer taking public transit
INV4: I like the feeling of taking public transit
INV8: I always pay attention to the information about public transit
INV9: I believe taking public transit will change how people judge me

0.83
0.83
0.82
0.80
0.79
0.76
0.74
0.73
0.58

Eigenvalues

Variance
explained (%)

Cronbach a

5.29

58.84

0.91

KaiserMeyerOlkin measure of sampling adequacy 0.91.

Bartletts test of sphericity 4131.32 (d.f. 36, p 0.000).
Total variance explained 58.84%.

Table 2
Factor analysis of service quality.
Factor/item
Core service (CS)
SQ4: general information provision
SQ3: network coverage
SQ2: service provision hours
SQ5: prices of tickets
SQ1: service frequency
SQ8: complaint dealing
SQ6: ticket selling network
SQ9: train information provision
SQ7: personnel behavior
Psychical environment (PE)
SQ13: facility cleanliness
SQ15: vehicle cleanliness
SQ18: vehicle safety
SQ14: safety at terminals and stops
SQ19: vehicle stability
SQ12: conditions at terminals and stops
SQ17: onboard information provision

Factor loading

Eigenvalue

Variance explained (%)

Cronbach a

8.17

28.12

0.89

1.29

27.48

0.88

0.73
0.73
0.69
0.67
0.64
0.64
0.59
0.58
0.52
0.75
0.75
0.74
0.73
0.71
0.65
0.65

KaiserMeyerOlkin measure of sampling adequacy 0.952.

Bartletts test of sphericity 6588.309 (d.f. 136, p 0.000).
Total variance explained 55.60%.
Note: SQ10, SQ11, and SQ16 are removed due to their factor loadings are less than 0.5.

W.-T. Lai, C.-F. Chen / Transport Policy 18 (2011) 318325

factor concerns core service (nine items, a 0.89) while the

second relates to physical environment (seven items, a 0.88).
The factors, including involvement and service quality, are
measured through a single index by averaging the item scores
for each factor for subsequent analyses.

4.2. Measurement model

The measurement model t is assessed by a conrmatory
factor analysis (CFA), with the results shown in Table 1. Six
common measures are used to estimate the model t, namely the
ratio of the chi-square value to degrees of freedom (w2/d.f.),
goodness-of-t index (GFI), root mean square error of approximation (RMSEA), normed t index (NFI), non-normed t index
(NNFI), and the comparative t index (CFI). A GFI, a NFI, a NNFI,
and a CFI above 0.90 indicate a close t of the model to the data.
Moreover, an RMSEA of 0.08 or less also indicates a good t (Hair
et al., 2006). As shown in Table 3, the chi-square statistic
(w2 88.20, d.f. 18) is signicant, however, the ratio of the chisquare value to degrees of freedom (w2/d.f.4.9) is less than 5.
Other t indices namely, GFI (0.97), CFI (0.99), NFI (0.99), NNFI
(0.98), and RMSEA (0.07) exceed their respective common
acceptance levels, as mentioned above. Therefore, the measurement model has a good t with the data collected.
Item reliability, construct reliability, and average variance
extracted are all examined with regard to the convergent validity
of CFA (Hair et al., 2006). Item reliability indicates the amount
of variance in an item due to the underlying construct. t-Values for
all the standardized factor loadings of items are found to be
signicant (po0.05), and all loadings are larger than 0.5, assuring
item reliability (see Table 3). Construct reliability estimates
range from 0.84 to 0.87, greater than 0.6, which is the threshold
value of acceptable reliability. The average variance extracted (AVE)
measure reects the overall amount of variance in the indicators
accounted for by the latent construct. The average variances
extracted of all constructs range from 0.69 to 0.74, above the cutoff value of 0.5, indicating all constructs are satisfactory. The
convergent validity of the measurement model appears to be
adequate.
In addition, discriminant validity is assessed by comparing the
average variance extracted (AVE) of each individual construct
with shared variances between this individual construct and all
the other constructs. A higher AVE of the individual construct
than shared variances suggests discriminant validity. As shown in
Table 4, comparing the square root of the AVE with the
correlations among the constructs indicates that each construct
is more closely related to its own measures than to those of other
constructs, implying adequate discriminant validity (Fornell and
Larcker, 1981).

323

4.3. Structural model

Having established a reliable and valid measurement model, a
structural model is used to test the predictive relationships
between constructs of the proposed conceptual model. The
simultaneous maximum-likelihood-estimation procedures are
used to examine the hypothesized relationships among service
quality, perceived value, overall satisfaction, involvement, and
behavioral intentions. The goodness-of-t indices of the nal
estimated structural model after the model modication process
include w2/d.f. (83.52/184.65), GFI (0.97), AGFI (0.94), CFI (0.99),
NFI (0.99), and RMSEA (0.07), indicating that the structural model
has a reasonable explanation of the observed covariance among
the constructs of interest. Fig. 4 depicts the results of the
estimated structural model.

4.3.1. Hypotheses testing

Regarding the hypothesis tests, as shown in Table 3 and Fig. 4,
all structural path estimates are statistically signicant (po0.01).
Service quality has a signicantly positive effect on perceived
Table 4
Correlation matrix of constructs.
Construct

Mean

S.D.

IN

SQ

PV

SAT

BI

IN
SQ
PV
SAT
BI

3.77
3.92
3.81
4.14
3.97

0.66
0.52
0.70
0.68
0.73

0.62
0.69
0.56
0.64

0.86
0.66
0.63
0.60

0.83
0.58
0.60

0.63

0.85

Note: The values on the diagonal are square roots of average variance extracted.

0.17 (2.95)

Service
quality

0.75 (18.64)

0.49 (9.17)
Satisfaction

0.24 (4.55)
0.25 (4.80)

Perceived
value

0.31 (7.98)

Behavioral
intention

0.09 (2.54)
Involvement

0.27 (6.38)

0.50 (9.62)
0.18 (3.24)
Fig. 4. The estimated model.

Table 3
CFA for the measurement model.
Constructs

Involvement (IN)
Service quality (SQ)
Perceived value (PV)

Satisfaction (SAT)
Behavioral intention (BI)

Items

INV
CS
PE
PV1
PV2
PV3
OSAT
BI1
BI2

Item reliability
Factor
loading

Standard
errors

Error variance

t-value

1.00
0.90
0.82
0.80
0.80
0.89
1.00
0.88
0.82

0.02
0.03
0.03
0.03
0.03
0.03
0.02
0.03
0.03

0.00
0.19
0.33
0.35
0.35
0.21
0.00
0.22
0.33

39.04
29.14
26.48
23.80
25.34
28.50
39.04
28.82
25.47

Construct
reliability

Average variance
extracted


0.85


0.74

0.87

0.69


0.84


0.72

324

W.-T. Lai, C.-F. Chen / Transport Policy 18 (2011) 318325

Table 5
Direct, indirect, and total effects of behavior intention.
Path

Effect

Estimates

t-value

Involvement-Behavioral intentions

Direct effect
Indirect effect
Total effect
Direct effect
Indirect effect
Total effect
Direct effect
Indirect effect
Total effect
Direct effect
Indirect effect
Total effect

0.27
0.42
0.69
0.17
0.25
0.42
0.18
0.06
0.24
0.31

0.31

6.30**
10.86**
18.97**
2.95**
7.17**
9.09**
3.24**
3.17**
4.06**
7.98**

7.98**

Service quality-Behavioral intentions

Perceived value-Behavioral intentions

Satisfaction-Behavioral intentions

nn

Denotes p o 0.01.

value (g1 0.75, t-value18.64), overall satisfaction (g2 0.49,

t-value 9.17), involvement (g3 0.24, t-value4.55), and behavioral intentions (g4 0.17, t-value 2.95), and thus H1, H2, H3, and
H4 are supported. The results show that service quality plays a
signicant role in passenger evaluations of public transit services,
involvement, and behavioral intentions.
The effects of perceived value on overall satisfaction (g5 0.25,
t-value 4.81), involvement (g6 0.50, t-value9.62), and behavioral intentions (g7 0.18, t-value3.24) are signicantly positive, indicating that H5, H6, and H7 are supported. Overall
satisfaction has a signicantly positive effect on both involvement
(g8 0.09, t-value2.54) and behavioral intentions (g9 0.31,
t-value7.98), suggesting support for H8 and H9. Finally, involvement is also found to have a signicantly positive effect on behavioral
intentions (g10 0.27, t-value6.38), and thus H10 is supported.
4.3.2. Indirect, direct, and total effects on behavioural intentions
Table 5 reports the effects (i.e. direct, indirect, and total) of
various determinants on passengers behavioral intentions.
According to the estimated model, apart from satisfaction with
mere direct effects, all antecedents have both direct and indirect
effects on behavioral intentions. The total effect is estimated as
the sum of direct and indirect effects. In terms of total effects,
involvement (0.69) appears to have the largest effect on
behavioral intentions followed by service quality (0.42), satisfaction (0.31), and perceived value (0.24) in order. With respect to
direct effects, involvement (0.27) also has a considerably large
effect on behavioral intentions while satisfaction (0.31) has the
largest direct effect. Moreover, the indirect effect of involvement
(0.42) on behavioral intentions mediated by service quality,
perceived value, and satisfaction is also the largest among various
antecedents. These results suggest that the role of involvement
inuencing passengers behavioral intentions, either toward reusing public transit services or recommending the public transit
services to others, should not be negligible. Furthermore, how to
increase passengers level of involvement deserves careful consideration in the marketing and promotion strategies of public
transit companies.

5. Discussion and conclusion

This study investigated public transit passengers behaviors by
constructing a comprehensive model considering public transit
involvement, service quality, perceived value, satisfaction, and
behavioral intentions. In particular, different from previous
studies, the role of involvement is addressed in the process of
passenger experience while taking public transit services. The

relationship model was tested using data obtained from a

passenger questionnaire survey of the KMRT system in Taiwan.
The results of the relationships between service quality,
perceived value, satisfaction, and behavioral intentions suggest
important implications for public transit companies. Passenger
behavioral intentions or loyalty signicantly rely on passenger
satisfaction. To enhance passenger satisfaction, two antecedents,
i.e. service quality and perceived value, must be addressed.
Service quality is measured as passenger evaluations of the
service attributes of public transit services, such as the general
characteristics of the public transport system, terminals and
stops, vehicles, and transport points (Tyrinopoulos and Antonious,
2008). Quality improvement and management, such as identifying those specic attributes about which passengers are most
concerned and endeavoring to offer quality services, are prerequisites to obtaining passenger satisfaction. According to
our results, service attributes such as vehicle safety, facility
cleanliness, and complaint handing have signicant inuences on
passenger behavioral intentions.1 These ndings can provide
useful information for the KMRT in its efforts to prioritize the
important service attributes and ensure its service quality meets
or exceeds passenger expectations. However, offering better
service quality is always costly, and such costs are frequently
shifted to passengers through higher ticket prices, and these
might offset any benets thus obtained.
In other words, passenger satisfaction is not guaranteed if the
quality of the services that public transit delivers do not result in
increased passenger perceptions of value in relation to the fares
paid. Therefore, providing passenger-value-oriented quality services is crucial for public transit companies if they are to satisfy
their passengers and thus increase re-patronage/word-of-mouth
behavior, and consequently customer loyalty. Currie and Wallis
(2008) provided a synthesis of the evidence with respect to
effective ways to grow urban bus markets in Europe, North
America, and Australia. They focused on service improvement
measures, including network structure and service level, bus
priority measures, vehicles and stop infrastructure, fare and
ticketing systems, passenger information and marketing, personal
safety and security, and synergy effects of measures. In addition,
the effects of psychological factors such as service quality,
satisfaction and perceived value on passenger behaviors have
also been addressed in the eld of transportation research (Hu
and Jen, 2006; Jen and Hu, 2003; Joewono and Kubota, 2007;
Tyrinopoulos and Antonious, 2008).
While offering quality transit services is essential, increasing
passengers involvement with, by the use of advertisements, for
example, can be employed as a pull-strategy to develop the transit
market. Another contribution of this study is to explore the role
and effect of passengers involvement in public transit on
passenger behavioral intentions. Our results show that involvement has a positive and signicant effect on passenger behavioral
intentions. In the public transit context, passengers with higher
public transit involvement are those who have higher needs,
values, and interest with regard to public transit. Therefore, it is
rational that higher public transit involvement is positively
related to higher behavioral intentions and other antecedents.
Given the importance of involvement with regard to passenger
behavioral intentions, activities aim at increasing passenger
involvement could be used to design effective marketing
strategies in the public transit context. Marketers of public transit
services should seek various ways, such as eye-catching advertisements and celebrity endorsements, to increase passengers

1
Due to limited space, the details of the regression results are not reported in
this paper. However, they are available upon request from the authors.

W.-T. Lai, C.-F. Chen / Transport Policy 18 (2011) 318325

involvement with such services. Increasing ridership is extremely

important for a public transit system like the KMRT, which is a
recent BOT transit system start-up with current ridership far
below its expected break-even point. Therefore, in addition to
retaining existing passengers, more importance should be
attached to attracting new passengers. A more marketingoriented approach rather than the traditional supply-oriented
approach is thus believed to be more appropriate for the KMRT.
Several limitations of this study provide research direction for
future studies. First, given the importance of involvement, the
development of a comprehensive and valid involvement scale
which can apply to general public transport services is recommended. Second, how to design and implement effective marketing strategies incorporating involvement to attract more
passengers also need further investigation. Third, involvement is
only specied as one of the determinants of passenger behavioral
intentions in this study. However, as frequently discussed in the
behavioral and marketing literature, involvement might act as a
mediator or moderator in the attitudebehavioral relationships
(Kokkinaki and Lunt, 1997). In this study, involvement is affected
by customer evaluations of services, such as perceived value,
service quality, and satisfaction. When people are satised with
the public transport, their involvement toward it will increase.
Increased involvement might play a role as a moderator in the
relationship between service evaluations and behavioral intentions. Although the potential moderating role of involvement is
not explored in this study, it deserves further investigation in the
public transit context.

Acknowledgements
The authors would like to thanks the two anonymous
reviewers for their helpful suggestions and comments and Joyce
Chen for her great assistance in this paper.
References
Ajzen, I., Fishbein, M., 1980. Understanding Attitudes and Predicting Social
Behavior. Prentice-Hall, Englewood Cliffs, NJ.
Anderson, W., Sullivan, M., 1993. The antecedents and consequences of customer
satisfaction for rms. Market Sci. 12, 125143.
Beatty, S.E., Kahle, L.R., Homer, P., 1988. The involvementcommitment model:
theory and implications. J. Bus. Res. 16, 149167.
Bennett, R., Hartel, C.E.J., MaColl-Kennedy, J.R., 2005. Experience as a moderator of
involvement and satisfaction on brand loyalty in a business-to-business
setting. J. Bus. Res. 34, 97107.
Bloemer, J.M., de Ruyter, K., 1998. On the relationship between store image, store
satisfaction and store loyalty. Eur. J. Market 32, 499513.
Chen, C.-F., 2008. Investigating structural relationships between service quality,
perceived value, satisfaction, and behavioral intentions for air passengers:
evidence from Taiwan. Transport. Res. Part A 42 (4), 709717.
Chen, C.-F., Tsai, M.-H., 2008. Perceived value, satisfaction, and loyalty of TV travel
product shopping: involvement as a moderator. Tourism Manage. 29, 11661171.
Cronin, J.J., Taylor, S.A., 1992. Measuring service quality: a reexamination and
extension. J. Market 56, 5568.
Cronin, J.J., Brady, M.K., Hult, G.T.M., 2000. Assessing the effects of quality, value
and customer satisfaction on consumer behavioral intentions in service
environments. J. Retail. 76 (2), 193218.
Currie, G., Wallis, I., 2008. Effective ways to grow urban bus marketsa synthesis
of evidence. J. Transp. Geo. 16, 419429.

325

Fornell, C., 1992. A national customer satisfaction barometer: the Swedish

experience. J. Market 5, 121.
Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with
unobservable variables and measurement error. J. Market Res. 18 (1), 3950.
Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 2006. Multivariate Data Analysis
6th edition Prentice Hall, New Jersey.
Hu, K.C., Jen, W., 2006. Passengers perceived service quality of city buses in Taipei:
scale development and measurement. Transp. Rev. 26 (5), 645662.
Jen, W., Hu, K.-C., 2003. Application of perceived value model to identify factors
affecting passengers repurchase intentions on city bus: a case of the Taipei
metropolitan area. Transportation 30, 307327.
Joewono, T.B., Kubota, H., 2007. User satisfaction with paratransit in competition
with motorization in Indonesia: anticipation of future implications. Transportation 34, 337354.
Johnson, M.D., Anderson, E.W., Fornell, C., 1995. Rational and adaptive performance expectation in a customer framework. J. Cust. Res. 21, 595707.

Joreskog,
K., Sorbom,
D., 1996. LISREL 8: Users Reference Guide. Scientic Software
International, Chicago, IL.
Kokkinaki, F., Lunt, P., 1997. The relationship between involvement, attitude
accessibility and attitudebehavioral consistency. Br. J. Soc. Psychol. 36, 497609.
Lai, W.T., Lu, J.L., 2007. Modeling the working mode choice, ownership and usage
of car and motorcycle in Taiwan. Journal of the Eastern Asia Society for
Transportation Studies, 7, 869895.
Lam, S.Y., Shankar, V., Erramilli, M.K., Murthy, B., 2004. Customer value,
satisfaction, loyalty, and switching costs: an illustration from a business-tobusiness service context. J. Acad. Market. Sci. 32 (3), 293311.
Lovelock, C.H., 2000. Service Marketing 4th edition Prentice Hall International,
New Jersey.
Mass Rapid Transit Bureau, 2009. Kaohsiung Mass Rapid Transit, /http://mtbu.
kcg.gov.twS. Accessed February 25 2009.
Nathanail, E., 2008. Measuring the quality of service for messengers on the
Hellenic railways. Transp. Res. Part A 42, 4866.
Olsen, S.O., 2007. Repurchase loyalty: the role of involvement and satisfaction.
Psychol. Market 24 (4), 315341.
Oliver, R., 1980. A cognitive model of the antecedents and consequences of
satisfaction decisions. J. Market. Res. 17, 460469.
Oliver, R., 1997. Satisfaction: A Behavioral Perspective on the Consumer. McGrawHill, New York.
Oliver, R., 1999. Whence consumer loyalty? J. Market 63 (3344) 1999.
Parasuraman, A., Zeithaml, V.A., Berry, L., 1985. A conceptual model of service
quality and its implications for future research. J. Market 49 (4), 4150.
Parasuraman, A., Zeithaml, V.A., Berry, L., 1988. A multiple item scale for
measuring consumer perceptions of service quality. J. Retail. 64 (1), 1240
SERVQUAL.
Petrick, J.F., 2004. The roles of quality, value and satisfaction in predicting cruise
passengers behavioral intentions. J. Travel Res. 42 (4), 397407.
Petrick, J.F., Backman, S.J., 2002. An examination of the construct of perceived
value for the prediction of golf travelers intentions to revisit. J. Travel Res. 41
(1), 3845.
Pritchard, M.P., Havitz, M.E., Howard, D.R., 1999. Analyzing the commitment
loyalty link in service contexts. J. Acad. Market. Sci. 27, 333348.
Richins, M.L., Bloch, P.H., 1991. Post-purchase product satisfaction: incorporating
the effects of involvement and time. J. Bus. Res. 23, 145158.
Transportation Research Board, 1999. A handbook for measuring customer
satisfaction and service quality. TRCP Report 47.
Transportation Research Board, 2004. Transit capacity and quality of service
manual. TRCP Report 100 (2nd edition).
Tsiotsou, R., 2006. The role of perceived product quality and overall satisfaction on
purchase intentions. Int. J. Consum. Stud. 30 (2), 207217.
Tyrinopoulos, Y., Antonious, C., 2008. Public transit user satisfaction: variability
and policy implications. Trans. Pol. 15, 260272.
Woodruff, R.B., 1997. Customer value: the next source for competitive edge.
J. Acad. Market. Sci. 25 (2), 139153.
Yang, Z., Peterson, R.T., 2004. Customer perceived value, satisfaction, and loyalty:
the role of switching costs. Psychol. Market 21 (10), 799822.
Zaichkowsky, J.L., 1985. Measuring the involvement construct. J. Consumer Res. 12
(3), 341352.
Zaichkowsky, J.L., 1994. The personal involvement inventory: reduction, reversion,
and application to advertising. J. Advertising 23 (4), 5970.
Zeithaml, V.A., 1988. Consumer perceptions of price, quality and value: a meansend model and synthesis of evidence. J. Market 52 (July), 222.
Zeithaml, V.A., Berry, L.L., Parasuraman, A., 1996. The behavioral consequences of
service quality. J. Market 60 (2), 3146.