International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012 45

Applying the Theory of Planned

Behavior to Predict Low-Carbon
Tourism Behavior:
A Modified Model from Taiwan
Nae-Wen Kuo, Department of Geography, National Taiwan Normal University, Taipei City,
Taiwan
You-Yu Dai, Graduate Institute of Recreation, Tourism, and Hospitality Management,
National Chiayi University, Chiayi City, Taiwan

ABSTRACT
To reduce carbon emissions resulted from tourism industry, low-carbon tourism is advocated and has become
an important tourism policy in many countries. Previously, studies concerned about low-carbon tourism
were focused on how to plan and design low-carbon tourism activities and itinerary products. However, little
research was paid attention to the low-carbon tourism behavior of tourists and the factors that will influence
their low-carbon tourism behavior were still unclear. Factors affecting tourists’ low-carbon tourism behavior
are important and need to be explored. The main purpose of this research was to find the important factors
that will affect tourists’ behavior, and a modified Theory of Planned Behavior (TPB) mode was used. In this
study, an effective sample of 387 Taiwanese who visited the first “ECO Taiwan Expo” in Taiwan was col-
lected. The results showed that the respondents were independently involved in low-carbon tourism, rather
than influenced by the views of significant others or groups. In addition, past travel experiences could improve
perceived behavioral control and behavioral intention toward “low-carbon tourism” behavior. Finally, a
moderating effect of perceived behavioral control between behavioral intention and preferred behavior was
found in this study.

Keywords: Carbon Reduction, Low-Carbon Tourism, Taiwan, Theory of Planned Behavior (TPB), Tourism

1. INTRODUCTION tions and subsequent behavior of individuals

To promote carbon reduction in the 21st cen-
tury, low-carbon tourism is a rising tourism and
recreation paradigm. The Theory of Planned
Behavior (TPB) explains behavioral inten-
DOI: 10.4018/jthi.2012100103
as a result of three factors: personal attitudes,
subjective norms, and perceived behavioral
control. TPB is often used in tourism and hos-
pitality research to predict and examine tour-
ists’ intention for choosing destination (Lam
& Hsu, 2006), decision-making (Quintal, Lee,
& Soutar, 2010), tour planning (Sparks, 2007),

Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
46 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012
and so on. However, a research gap exists in
the model about explaining the low-carbon
tourism orientation.
This research shall focus on the topic of
“low-carbon tourism” behavior in Taiwan.
“Low-carbon tourism” behavior in the context of
this study refers to “Tourists may reduce carbon
dioxide emission by their choice of destinations,
favouring environmentally friendly means of
transport, and choosing environmentally cer-
tified hotels, as well as eating in restaurants
providing local and/or organic food. Tourists
can also demand transportation in new, fuel-
efficient aircraft, or to stay in environmentally
friendly” (Simpson, Gössling, Scott, Hall, &
Gladin, 2008). Globally, environmental con-
cerns receive relatively little coverage in tourism
industry due to consumer purchasing decisions
and corporate environmental performance
is largely unknown (Buckly, 2002). Locally,
public reception toward environmental issues
has also been lukewarm in Taiwan. In response,
the government has introduced new initiatives
to increase the environmental awareness of the
public in an effort to transform Taiwan into a
“low-carbon island.” The regency of this topic
in the public mindset makes it a good point of
focus for this study.
The purposes of this study are, first of all,
to establish a reasonable model via TPB to
predict which factors may impact “low-carbon
tourism” toward behavior. Next, we want to
make up gaps in theories by integrating possible
constructs. The specific objectives of this study
were to address the impact of attitude, subjec-
tive norms, and perceived behavioral control,
past travel experiences on behavioral intention,
to exam the impact of behavioral intention on
preferred behavior, to investigate the moderat-
ing role of perceived behavioral control in the
relationship between behavioral intention and
preferred behavior, and to identify perceived
barriers that tourists possibly perceive.
Consequently, this study choices the TPB
framework to determine the persuasive efficacy
on behavioral intention and preferred behavior
toward low-carbon tourism. The relationship
between past travel experiences, the three
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
predictive factors of behavioral intention, and
preferred behavior will be investigated. In ad-
dition, perceived behavioral control may affect
as moderator between behavioral intention and
preferred behavior. Thus, we proposed a modi-
fied model in this research to exam the moderator
effect of perceived behavioral control.
2. LITERATURE REVIEW
2.1. The Theory of
Planned Behavior
2.1.1. The Core Constructs
In social psychology, there is still conclusive in
theories or models of how attitude could predict
or influence the behavior. However, fortunately,
the salt scholars in these studies (e.g., Armitage
& Conned, 2000; Garcia & Mann, 2003) were
thought that the TPB can provide researchers
useful ideas. In this way, they can predict the
possible influential behavior on environment
when people engaged in tourism, and then
provide prevention and advocacy opinions. The
TPB was proposed by Ajzen (1985) to predict
factors that may influence human beings’ be-
havior. There are five mainly constructs build
up TPB, which are attitude, subjective norm,
perceived behavioral control, behavioral inten-
tion, and behavior.
According to the TPB, the most proximal
determinant of an individual’s behavior is his
or her intention to engage in the behavior (Fig-
ure 1). In other words, behavioral intention is
predicted by three main antecedents: attitude,
subjective norm and perceived behavioral
control (Ajzen, 1985, 1991; Ajzen & Fishbein,
2005; Abraham & Sheeran, 2003; Hagger &
Chatzisarants, 2005). The three predictors
weighted for their importance in relation to the
behavior and population of interest. Attitudes
refer to the overall positive or negative evalua-
tion of performing the behavior. Attitude toward
a behavior is the degree to which performance
of the behavior is positively or negatively
valued (Reber, Allen, & Reber, 2009). Ajzen
(1988, 1991) and Ajzen and Fishbein (2005) all
 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012 47
Figure 1. The TPB model (adapted from Ajzen & Fishbein, 2005)
mentioned that, according to the expectancy-
value model (EVT) (Fishbein & Ajzen, 1975),
attitude toward a behavior is determined by
the total set of accessible behavioral beliefs
linking the behavior to various outcomes and
other attributes. Specifically, the strength of
each belief is weighted by the evaluation of
the outcome or attribute.
Subjective norm is based on an individual’s
perception of whether a key person in their life
would like them to perform the behavior (Reber,
Allen, & Reber, 2009). According to Ajzen
(1998, 1991) and Ajzen and Fishbein (2005),
subjective norm is the perceived social pressure
to engage or not to engage in a behavior. Draw-
ing an analogy to the expectancy-value model
of attitude, it is assumed that subjective norm
is determined by the total set of accessible
normative beliefs concerning the expectations
of important referents. Specifically, the strength
of each normative belief is weighted by motiva-
tion to comply with the referent in question.
Perceived behavioral control reflects the
extent to which individuals perceive the behav-
ior to be under their volitional control (Reber,
Allen, & Reber, 2009). Ajzen (1998, 1991)
and Ajzen and Fishbein (2005) mentioned that,
Perceived behavioral control refers to people’s
perceptions of their ability to perform a given
behavior. Drawing an analogy to the expectan-
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
cy-value model of attitude, it is assumed that
perceived behavioral control is determined by
the total set of accessible control beliefs, i.e.,
beliefs about the presence of factors that may
facilitate or impede performance of the behav-
ior. Specifically, the strength of each control
belief is weighted by the perceived power of
the control factor.
Intention is an indication of a person’s
readiness to perform a given behavior, and it
is considered to be the immediate antecedent
of behavior (Ajzen, 1988, 1991; Ajzen & Fish-
bein, 2005). Actual behavioral control refers
to the extent to which a person has the skills,
resources, and other prerequisites needed to
perform a given behavior (Ajzen & Fishbein,
2005). Successful performance of the behavior
depends not only on a favorable intention but
also on a sufficient level of behavioral control.
To the extent that perceived behavioral control is
accurate, it can serve as a proxy of actual control
and can be used for the prediction of behavior.
The last mainly construct is behavior. Behavior
is the manifest, observable response in a given
situation with respect to a given target (Ajzen,
1988, 1991; Ajzen & Fishbein, 2005). Single
behavioral observations can be aggregated
across contexts and times to produce a more
broadly representative measure of behavior.
48 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012
2.2. Moderating Effects of
Perceived Behavioral Control
Intention is a key determinant of action
(Amireault, Godin, Vohl, & Pérusse, 2008).
However, there used to be a gap between inten-
tion and behavioral performance that remains
to be explained. Therefore, many studies aimed
to identify moderators of the intention-behavior
and perceived behavioral control-behavior re-
lationships for personal activity. According to
literature review, perceived behavioral control
has received mixed evidence for its moderating
effect of the intention-behavior relationship
(Armitage & Conner, 2001; Yang-Wallentin,
Schmidt, Davidov, & Bamberg, 2003). In each
case, higher perceived behavioral control was
associated with better intention-behavior con-
sistency. Moderation effect was reported for
past behavior for both the intention-behavior
(Sheeran & Abraham, 2003) and perceived be-
havioral control-behavior (Norman, Conner, &
Bell, 2000) relationships. As Ajzen (1988, 1991)
and Ajzen and Fishbein (2005) mentioned, in
the TPB, behavior is a function of compatible
intentions and perceptions of behavioral control.
Conceptually, perceived behavioral control is
expected to moderate the effect of intention
on behavior, such that a favorable intention
produces the behavior only when perceived
behavioral control is strong.
Consequently, TPB could show that (1)
significant behavioral intention will be directly
affected by personal beliefs and the results of
the evaluation of behavior (attitude), views of
significant others or groups and the degree of
follow (subjective norms), and personal percep-
tion for the ability to overcome obstacles and
other effects (perceived behavioral control).
(2) The three significant directly impact on
behavioral intentions will follow-up actions. (3)
Perceived behavioral control may have signifi-
cantly moderating affect between subsequent
behavior and behavioral intention. Ajzen and
Fishbein (2005) than modified TPB to explain
that (4) perceived behavioral control may have
an antecedent existence, the effect named of
actual behavioral control.
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
2.3. The Role of Past
Travel Experiences
On the other hand, personal past experiences
may impact his/her behavioral intention directly
(Sönmez & Graefe, 1998). Lam and Hsu (2006)
thought it is reasonable to assume that the past
behavior could predict behavioral intention
of TPB. Experiences would associate with
perceived behavioral control (Cestac, Paran,
& Delhomme, 2011). Thus, in our study, we
modified TPB to add personal past experi-
ences to replace actual behavioral control as
an antecedent both of behavioral intention and
perceived behavioral control.
Understanding the beliefs associated
with particular low-carbon practices, and how
these beliefs may differ from people, is criti-
cal. It can provide important information for
policy makers, extension agents, and other
stakeholders seeking to promote the use of
low-carbon tourism resource management
practices. Nevertheless, previous research us-
ing this framework to understand the adoption
of low-carbon tourism in tourism operations is
relatively limited and has not included all belief-
based measures (i.e., behavioral, normative and
control beliefs). Further research is needed in
order to test the efficacy of this model across
a range of low-carbon tourism management
contexts and practices.
An important aspect of the TPB is that
it goes beyond merely identifying the direct
determinants of intentions and behavior, by
theorizing about the factors that underpin these
determinants. According to the TPB, our atti-
tudes are formed via behavioral beliefs that are
weighted by outcome evaluations. Subjective
norms are thought to be a function of norma-
tive beliefs that are weighted by the motivation
to comply. Additionally, the perceptions of
behavioral control are control beliefs that are
weighted by perceived power. Thus, attitudes,
subjective norms and perceived behavioral con-
trol are determined by behavioral, normative,
and control beliefs (Ajzen, 1985, 1991; Ajzen
& Fishbein, 2005).
 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012 49
The TPB provides a pool model for un-
derstanding and predicting how information
and motivation influence behavior (Manstead
& Parker, 1995). The TPB also highlights the
critical role that beliefs play, as antecedents to
intentions and behavior. It further highlights the
need for additional research, in order to identify
the salient beliefs associated with low-carbon
practices. Not only does the TPB provide a pool
model, it also offers a rigorous methodology
to identify the beliefs that underpin attitudes,
norms, and control perceptions. Moreover,
it allows for an investigation that is separate
from the difference between individuals who
adopt or do not adopt a particular low-carbon
practice (or individuals who are intenders or
non-intenders).
2.4. Past TPB Studies
Over the past few decades, TPB has been de-
veloped to a mature state. It was used in many
empirical behavior sciences studies. Such as
environmental behavior, for example, Fielding,
Terry, Masser, Bordia, and Hogg (2005) studied
soil and water conservation by TPB; Andrews,
Silk, and Eneli (2010) used TPB to understand
health promotion behavior, occupational behav-
ior, for instance, Fraser, Johnson, Hebert, Ajzen,
Copeland, Brown, and Chan (2010) invested
employers’ hiring intentions; and so on.
In addition, TPB was also used by many
tourism researches. Lam and Hsu (2006) used
TPB on behavioral intention of choosing a
travel destination. Sparks (2007) predicted
wine tourist intentions based on TPB. Quintal,
Lee, and Soutar (2010) using TPB to exam the
differential impacts risk and uncertainty have
on travel decision-making. Lee, Song, Bendle,
Kim, and Han (2012) introduced a new concept
of non-pharmaceutical intervention (NPI) for
influenza, and tested the impact of NPI on the
behavioral intention of potential international
tourists. They also extended the model of goal-
directed behavior (MGB) by incorporating the
new concepts of NPI, and the perception of 2009
H1N1. The model found that desire, perceived
behavioral control, frequency of past behavior,
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
and non-pharmaceutical interventions predicted
tourists” intention but perceptions of 2009 H1N1
had nil effect on desire and intention.
3. RESEARCH METHODOLOGY
3.1. Study Design and Hypotheses
Our research is to understand how people, who
involve low-carbon technology and issues, will
be influences by antecedent factors to do “low-
carbon tourism”. The first “ECO Taiwan Expo”
fairs with low-carbon products, technologies,
services and other elements, through envi-
ronmental education, trend discussion, game
experience, and product announcements, etc.
to achieve diversified purposes, such as new
green consumption patterns by people, promot-
ing environmental protection concepts, and the
creation of green products business. People
who went to visit the first “ECO Taiwan Expo”
initiatively may concern about those issues. So
we expect them have more ideas about “low-
carbon tourism.”
In addition, according to Ajzen’s TPB
model, first of all, attitude, subjective norm,
and perceived behavioral control are all positive
impact behavioral intentions. Next, behavioral
intention is a mediate variable between its three
antecedents and preferred behavior. Therefore,
the three predictive variables might predict
participants’ behavioral intentions to differing
extents as well. Third, past travel experience
would have positive impacts on perceived
behavioral control and behavioral intention.
In other words, past travel experiences as an
antecedent of perceived behavioral control
and behavioral intentions. Finally, perceived
behavioral control may play a moderating effect
when behavioral intentions directly influence
preferred behavior.
Moreover, if individuals are familiar with
the behavior to be adopted, which is likely to
be the case for individuals with high levels of
perceived behavioral control, behavioral inten-
tion should be more accurate and, consequently,
behavioral intention should adequately reflect
such perceived control and the intention-
50 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012
behavior relationship should be stronger.
Indeed, perceived behavioral control was a
significant determinant of behavior for samples
familiar with the behavior, whereas it was not
a significant determinant of behavior among
samples unfamiliar with the behavior. When
the frequency of perceived behavioral control
was moderate or high, behavioral intention
was a significant determinant of travel activity
whereas it was not a significant determinant
when the frequency was low.
In summary, this study aimed at testing
psychological moderator as well as perceived
behavioral control of the behavioral intention
and behavior relationships for “low-carbon tour-
ism” activity. The study structure is shown as
Figure 2. Given the lack of scholarly attention
in this field, we proposed the study hypotheses
as follows:
H1: Personal attitude toward “low-carbon tour-
ism” behavior is positively associated with
behavioral intentions toward “low-carbon
tourism” behavior.
H2: Subjective norms of participants toward
“low-carbon tourism” behavior are posi-
Figure 2. The modified TPB model of this study
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
tively associated with behavioral intentions
toward “low-carbon tourism” behavior.
H3: Perceived behavioral control toward “low-
carbon tourism” behavior is positively as-
sociated with behavioral intentions toward
“low-carbon tourism” behavior.
H4: P eople’s behavioral intention toward
“low-carbon tourism” positively influences
“low-carbon tourism” behavior.
H5-1: Personal past travel experiences is posi-
tively associated with behavioral intentions
toward “low-carbon tourism” behavior.
H5-2: Personal past travel experiences is
positively associated with perceived
behavioral control toward “low-carbon
tourism” behavior.
H6: Perceived behavioral control toward “low-
carbon tourism” behavior has moderating
effect between “low-carbon tourism”
behavior intention and preferred behavior.
3.2. Pilot Study
In accordance with the recommendations of
Ajzen and Fishbein (1980), an elicitation study
was conducted to identify the salient beliefs as-
sociated with in “low-carbon tourism” behavior
 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012 51
operations. So the purpose of the step is to
confirm what elements may be factors to influ-
ence personal behavioral intention. 6 Taiwanese
“low-carbon tourism” enthusiast (5 males and 1
female) were interviewed. The respondents were
asked a set of open-ended questions about the
advantages and disadvantages of “low-carbon
tourism” behavior in leisure time. Respondents
were also asked, “What individuals or groups
would approve or disapprove of ‘low-carbon
tourism’ behavior in leisure time?” Moreover,
they were asked, “What circumstances or factors
might prevent [them] from ‘low-carbon tour-
ism’ behavior in leisure time?” The four most
frequently cited advantages and disadvantages
were included in the main questionnaire, as a
measure of behavioral beliefs. The four most
frequently cited referents were used to measure
normative beliefs, and the eight most cited bar-
riers assessed control beliefs.
Next, we established the initial question-
naire and did a pilot study. To further refine the
scale and to test for reliability, field testing was
conducted using a convenience sample consist-
ing of undergraduate students from a variety
of disciplines and years of academic study at
two large universities in Taipei City and Chiayi
City, Taiwan. Beyond convenience, this sample
was deemed acceptable for testing the scale for
two reasons. First, like any part of the general
population, these students represented potential
travellers who may or may not reflect a predis-
position towards low-carbon tourism rather than
demonstrated behaviour or self-identification.
Second, as argued by Calder, Phillips, and
Tybout (1982) and empirically demonstrated
by Peterson (2001), using a relatively homo-
geneous sample, such as students, allows for
testing of the scale and for outcomes that are
less subject to confounding factors that could
introduce variance unrelated to the scale items.
Once the data had been collected, an itera-
tive process of reviewing and analyzing the
items within each dimension for conceptual and
empirical ‘fit’ was undertaken. Two procedures
were relied upon to help inform the best items
within each dimension. Principal component
analyses were run for each set of items within
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
a dimension examining the communalities (i.e.,
that they exceeded 0.40) and forcing a one-
factor solution to ensure that each item loaded
highly (i.e., loadings .0.40) on the underlying
dimension (Hair, Anderson, Tatham, & Black,
1998). When items failed to meet these mini-
mum requirements, they were again evaluated
for conceptual fit and, in most cases, discarded.
Subsequently, the item-to-total correlations
were considered and each set of items assessed
for internal consistency (i.e., reliability) by
calculating Cronbach’s coefficient alpha (α)
in order to identify a parsimonious number of
items defining the dimension that maximized its
reliability, with a minimum acceptable criterion
of 0.70 (Nunnally & Bernstein, 1994). The fit
and reliability of the dimensions in defining the
overall TPB questionnaire were also done once
each dimension was finally defined.
3.3. Main Study
3.3.1. Measurements
Measurement dimensions and operated vari-
ables are shown. In accordance with the rec-
ommendations of Ajzen and Fishbein (1980),
belief-based measures about the “low-carbon
tourism” behavior on respondents” properties
were obtained. Throughout the questionnaire,
“low-carbon tourism” behavior was defined
as “Visitors through the journey process, such
as select tourism destination, environmental
protected transportation, accommodation with
a certification of environmental protection, and
eat local food or organic food, etc., to lower the
amount of carbon dioxide emissions” (Simpson,
Gössling, Scott et al., 2008). And “low-carbon
tourism” behavior intention was defined as hav-
ing the intention to such behaviors. To assess
behavioral beliefs, respondents were asked to
indicate the likelihood, on a 5-point scale (from
-2 extremely unlikely to +2 extremely likely),
that each of the costs and benefits would lead
to “low-carbon tourism” behavior. The benefits/
costs were: (1) lower the amount of carbon
dioxide emissions, (2) reduce the amount of
garbage generated, (3) reduce food transport
52 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012
miles, (4) promoting good health, (5) protect the
global environment, and (6) let yourself have
a little more inspiration on the environment,
individual or society (-2 means “extremely bad”
to +2 means “extremely good”).
To assess the normative beliefs, respon-
dents were asked to assess how much they
thought each of the referents would think
that they should do “low-carbon tourism” (-2
extremely unlikely, +2 extremely likely). The
salient referents of participants were (1) parents,
(2) teachers/directors, (3) peers, (4) friends,
(5) environmental protection groups, and (6)
tourism industries. The motivation to comply
with the referents was measured by asking the
respondents how willing was to do what they
thought referents wanted them to do (-2 means
“not at all” to +2 means “very much”).
To assess control beliefs, respondents were
asked how much each of eight barriers would
prevent them from “low-carbon tourism” (-2
means “not at all” to +2 means “very much”).
The eight barriers were: (1) do not have the
concept of sustainable environment, (2) do not
have enough low-carbon tourism information,
(3) run-length price is not reasonable, (4) the
itinerary is not easy, (5) tour transportation is
not convenient, (6) without enough time and
money, (7) without a healthy body, and (8) the
tourism industry is not willing to cooperate.
Perceived power was inventoried as if each
of the barriers was removed, and respondents
asked how willing they were to “low-carbon
tourism” (-2 means “not at all” to +2 means
“very much”).
Behavioral intentions were measured
from the answer to five questions. An example
question is “Do you intend to plan ‘low-carbon
tourism’ in your next journey?” (-2 means “defi-
nitely intend not to” to +2 means “definitely
intend to”). Behaviors were measured from
the answer to two questions: (1) do you prefer
“low-carbon tourism” styles? And (2) do you
prefer use low-carbon transportation in your
trip? The questionnaire also included a ques-
tion assessing past travel experiences. Finally,
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
participators’socio-demographic variables were
also collected.
3.3.2. Sampling and Statistical
Analysis Method
The formal questionnaires were collected from
people who visited the first “ECO Taiwan
Expo.” We used random sampling method.
The survey was hold during Oct 21-23, 2010.
In the three days, there are 462 questionnaires
were granted, 425 were retrieved, and 387 are
valid questionnaires. The effective response rate
was 83.8%. All descriptive statistical analyses
of social demographic variables were made by
SPSS 19.0. As for the reliabilities, validities,
correlation coefficients, factor analyses, and
path analyses of constructs were examined both
by SPSS 19.0 and LISREL 8.80.
4. RESEARCH RESULTS
AND ANALYSIS
4.1. Descriptive Statistics
Table 1 summarizes the participants’ back-
ground of this investigation. Gender (59.4%
female) and age (42.4% 21-30 years) are in-
cluded. The Social Economic Status (SES) of
the respondents is captured via their occupation
(students are the most, n = 172), location (much
from Taipei/Keelung, n = 242), education (bach-
elor or above are the most, n = 338), marriage
status (single and others is popular, n = 254),
and personal average monthly income (Lower
than NT $20,000 is nearly half, n = 178).
As references mentioned, behavior beliefs
(weighted by evaluation of the outcome or at-
titude) has positive correlation with attitude;
normative beliefs (weighed by motivations to
comply) has positive correlation with subjective
norm; and control beliefs (weighed by perceived
power) has positive correlation with perceived
behavior control. Interrelationships among the
study variables, measure items, and references
are presented in Table 2.
 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012 53

Table 1. Descriptive statistics of respondents’ social demographic variables

Variables Items Number (%)

Valid 387 (83.8%)
Invalid 80 (16.2%)
Male 154 (39.8%)
Gender Female 230 (59.4%)
Missing values 3 (0.8%)
Below 20 years old 75 (19.4%)
21-30 years old 164 (42.4%)
Age 31-40 years old 63 (16.3%)
41-50 years old 56 (14.5%)
Behind 51 years old 29 (7.5%)
Administrators; business executives; the company responsible person / dealer 28 (7.3%)
Community service (police and civil servants); teachers or researchers 38 (9.8%)
Professionals 43 (11.1%)
Technical staff (workers) or assistants; service personnel or salespersons; service
Occupation 51 (13.2%)
staff
Student 172 (44.4%)
Housewives; retirees 39 (10.1%)
Others 16 (4.2%)
Kaohsiung / Pingtong 4 (1.0%)
Yunlin / Chiayi / Tainan 16 (4.1%)
Taichung / Chunghwa / Nantou 41 (10.6%)
Location
Taoyuan / Hsinchu / Miaoli 67 (17.3%)
Taipei / Keelung 242 (62.5%)
Others 17 (4.5%)
High school 44 (11.4%)
Bachelor 251 (64.9%)
Education
Master or above 87 (22.5%)
Missing values 5 (1.3%)
Married with children 104 (26.9%)

Marriage Married but no children 21 (5.4%)

status Single and others 254 (65.6%)
Missing values 8 (2.1%)
Less than NT $20,000 178 (46.0%)
NT $20,001-40,000 93 (24.0%)
Average NT $40,001-60,000 62 (16.0%)
monthly
income NT $60,001-80,000 26 (6.7%)
More than NT $80,001 17 (4.4%)
Missing values 11 (2.8%)

Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
54 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012
4.2. Confirmatory Factor Analysis
(CFA)
As a next step, a measurement model including
all study variables was estimated before the
structural equation modeling (SEM) to evaluate
the data quality (Anderson & Gerbing, 1988).
Table 3 shows that the composite reliability
(CR) of each aspect met the standard of 0.60
(Bagozzi & Yi, 1988). Besides attitude, the aver-
age variance extracted (AVE) was higher than
the standard of 0.50. In addition, the RMSEA,
CFI, and NNFI in this study were respectively
0.07, 0.96 and 0.95, all suggesting a satisfactory
goodness of fit index.
On the other hand, with regard discriminant
validity from measuring the correlation between
two different dimensions; a low correlation in-
dicated discriminant validity (Fornell & Larck-
er, 1981). When testing discriminant validity, if
the root mean square of AVE of each aspect were
larger than the number of correlationship coef-
ficient, and accounted for more than 83.3% of
the entire compared number (Hair, Anderson,
Tatham et al., 1998), it would be proven that
each aspect shared correlated but non-identical
factors, as presented in Table 4.
4.2.1. Structural Model and Empirical
Analysis
The hypothesis models in this study all showed
a satisfactory goodness of fit index (χ2 = 780.58
(df = 265), RMSEA = 0.07, SRMR = 0.08, CFI
= 0.95, GFI = 0.90, AGFI = 0.90, NFI = 0.93,
Table 2. Matrix of Pearson correlation coefficient among three predicting factors and anteced-
ent beliefs
Behavior beliefs (weighted by evaluation of the outcome
or attitude) (Cronbach’s α= 0.85)
Normative beliefs (weighed by Motivations to comply)
(Cronbach’s α= 0.88)
Control beliefs (weighed by Perceived power) (Cronbach’s
α= 0.80)
Note: ** p < 0.01.
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
NNFI = 0.95; χ2/df = 2.95, as showed in Table
4). Attitude and perceived behavior control had
positive impacts on behavioral intention (γ21 =
0.32, t = 5.02; β21 = 0.22, t = 4.25). Behavioral
intention also had a positive impact on behav-
ior (β32 = 0.16, t =2.01). Furthermore, positive
impacts of past travel experience were found
for perceived behavior control (γ13 = 0.26, t =
6.51), and behavioral intention (γ23 = 0.23, t =
3.26). Therefore, except our hypothesis H2, the
other hypotheses H1, H3, H4, H5-1, and H5-2
proposed in this study were accepted (Table 5).
In other words, personal attitude and per-
ceived behavioral control toward “low-carbon
tourism” behavior are significant positively
associated with behavioral intentions toward
“low-carbon tourism” behavior. Next, people’s
behavioral intention toward “low-carbon tour-
ism” significantly and positively influences
“low-carbon tourism” behavior. Third, per-
sonal past travel experiences have significant
positively impacts on associated with behav-
ioral intention and perceived behavioral control
toward “low-carbon tourism” behavior.
4.2.2. Empirical Testing of
Hypothesized Moderating Effect
of Perceived Behavioral Control
4.2.2.1. Grouping
To test the moderating role of perceived behav-
ioral control, the respondents were divided into
high and low groups based on the responses to
positive and negative perceived “low-carbon
tourism” barriers. K-means cluster analysis was
Subjective Perceived behavior
Attitude
norm control
0.52**
0.57**
0.44**
 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012 55

Table 3. CFA results of measurement model

SFL SE t-value SMC CR AVE

Attitude (Cronbach’s α= 0.70). 0.82 0.54
Engaged in low-carbon tourism is worth/unworthy in terms of 0.77 0.04 15.96 0.59
your business.
Engaged in low-carbon tourism is unneeded/need in terms of 0.81 0.04 17.24 0.65
your business.
Engaged in low-carbon tourism is very adverse / beneficial in 0.69 0.04 13.83 0.47
terms of your business.
Engaged in low-carbon tourism is inconvenient / convenient in 0.66 0.05 13.21 0.44
terms of your business.
Subjective norm (Cronbach’s α= 0.80). 0.81 0.52
In your daily life, most important person to you (e.g., family 0.79 0.04 16.48 0.63
members, teachers, competent...) considers that you engaged in
low-carbon tourism is should not be/should be.
In your daily life, most important group to you (e.g., school, work 0.72 0.04 14.48 0.52
place, community...) considers that you engaged in low-carbon
tourism is should not be/should be.
In your daily life, most important person to you (e.g., family 0.73 0.04 14.67 0.53
members, teachers, competent...) considers that you engaged in
low-carbon tourism is not supported/ supported.
In your daily life, most important group to you (e.g., school, work 0.63 0.04 12.10 0.39
place, community...) considers that you engaged in low-carbon
tourism is not supported/ supported.
Perceived behavior control (Cronbach’s α= 0.92). 0.95 0.69
If I do not gain the concept of sustainable environment during my 0.70 0.05 14.71 0.48
travel, I would not / would be engaged in low-carbon tourism.
If I do not have enough information of low-carbon tourism during 0.75 0.05 16.29 0.56
my travel, I would not / would be engaged in low-carbon tourism.
If the tour’s price is unreasonable during my travel, I would not 0.89 0.05 21.14 0.79
/ would be engaged in low-carbon tourism.
If the whole tour’s schedule is inconvenience during my travel,
0.91 0.05 21.86 0.82
I would not / would be engaged in low-carbon tourism.
If the tour’s transport is inconvenience during my travel, I would
0.90 0.05 21.49 0.81
not / would be engaged in low-carbon tourism.
If I do not have sufficient time and funding during my travel, I
0.87 0.06 20.27 0.75
would not / would be engaged in low-carbon tourism.
If I am not healthy during my travel, I would not / would be
0.79 0.05 17.64 0.63
engaged in low-carbon tourism.
If there is no tour operator willing to cooperate during my travel,
0.82 0.06 18.71 0.68
I would not / would be engaged in low-carbon tourism.
Behavioral intention (Cronbach’s α= 0.81). 0.82 0.47
From now on, I’m willing to plan low-carbon tourism during
0.71 0.03 14.29 0.51
my next tour.
From now on, I’m willing to tell my friends and family to engage
0.61 0.03 11.60 0.37
in low-carbon tourism.

continued on following page

Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
56 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012

Table 3. Continued

SFL SE t-value SMC CR AVE

During the next one year, I might (about 30% chances) engage
0.62 0.04 11.86 0.38
in low-carbon tourism.
During the next one year, I would (about 50% chances) engage
0.77 0.04 15.95 0.60
in low-carbon tourism.
During the next one year, I will (about 80% chances) engage in
0.72 0.05 14.35 0.51
low-carbon tourism.
Past travel experiences (Cronbach’s α= 0.72). 0.76 0.62
Did you have travel in the past one year? 0.59 0.06 9.78 0.34
Did you have “low-carbon tourism” in the past one year? 0.94 0.08 13.70 0.89
Preferred behavior (Cronbach’s α= 0.70). 0.86 0.75
Do you used to prefer “low-carbon tourism” style? 0.62 0.06 13.66 0.68
Do you used to prefer “low-carbon tourism” transportation? 0.52 0.04 14.22 0.81
Note: SFL: standard factor loading; SE: standard error; SMC: square of multiple correlations; CR: construct reliability;
AVE: average variance extracted; all t-values are significant at 0.05 levels (two-tail t-test p < 0.05).

used in grouping because the user can specify
the number of clusters; this analysis is useful
when the sample size is large (200 or more
cases) (Hair, Anderson, Tatham et al., 1998). In
statistics and data mining, K-means clustering
is a method of cluster analysis which aims to
partition n observations into k clusters in which
each observation belongs to the cluster with the
nearest mean (Yuksel & Yuksel, 2002).
It is similar to the expectation-maximiza-
tion algorithm for mixtures of Gaussians in that
they both attempt to find the centers of natural

Table 4. Matrix of phi correlation coefficient among constructs

Perceived Behavior
Standard Subjective Behavioral Past travel
Dimension Mean Attitude behavior
Deviation norm intention experiences
control
0.73
Attitude 1.13 0.73

Subjective 0.72
1.17 0.62 0.72
norm 1
Perceived
0.83
behavior 1.07 0.55 0.34 0.21
1
control
Behavioral 0.69
1.83 0.61 0.61 0.48 0.39
intention 1
Past travel 0.79
1.24 0.43 0.27 0.16 0.40 0.49
experiences 1
0.87
Behavior 1.84 0.52 0.29 0.59 0.80 0.55 0.51
1
Note: The values in boldface on the diagonal line are the root square of AVE. The values in the lower triangular matrix
are the non-standardized correlation coefficients.

Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012 57

Table 5. The latent variables and the estimation of hypothesis path parameters of latent variables
in the theoretical structural model

Estimated Parameter
Standard
Parameter Path Hypothesis Non- t-value
Standardized Deviation
Standardized
Attitude toward behavior on
γ21 H1 0.32 0.33 0.12 5.02
Behavioral intention
Subjective norm on
γ22 H2 0.12 0.11 0.11 0.30
Behavioral intention
Perceived behavioral control
β21 H3 0.22 0.23 0.06 4.25
on Behavioral intention
Behavioral intention on
β32 H4 0.16 0.16 0.04 2.01
Preferred behavior
Past travel experiences on
γ23 H5-1 0.23 0.23 0.06 3.26
Behavioral intention
Past travel experiences on
γ13 H5-2 0.26 0.26 0.04 3.11
Perceived behavioral control
Note: A level significance was achieved when t-value was greater than 1.65 (p < 0.05), in which case the hypothesis
was accepted.

clusters in the data as well as in the iterative
refinement approach employed by both algo-
rithms. The number of cases using the responses
for each component of perceived “low-carbon
tourism” barriers were split into two groups in
sequence. Specifically, the cases were divided
into high (95 cases) and low perceived behav-
ioral control groups (108 cases).
4.2.2.2. Measurement Invariance
According to Steenkamp and Baumgartner
(1998), if the measurement model is not support-
ed, findings from the structural invariance test
are mistaken. Therefore, prior to the structural
invariance test, measurement invariance was
assessed to determine if a measurement model’s
across groups were invariant. The general
method of assessing measurement invariance is
the chi-square difference test. The measurement
models are invariant when the chi-square does
not show a significant difference (Yoo, 2002). A
non-restricted model using confirmatory factor
analysis was first assessed, and then the equality
of the factor loadings across groups (full metric
invariance of the CFA model) was assessed.
Table 6 shows the results of the measurement

Table 6. Measurement invariance and structural invariance

Models χ2 df RMSEA CFI NFI

Non-restricted model 84.11 25 0.06 0.90 0.90
Measurement
invariance Full metric invariance of CFA model (L(X)
87.57 30 0.06 0.94 0.91
Y = IN*)a
Full metric invariance of structural model
104.63 37 0.07 0.91 0.92
Structural invari- (L(X)Y = IN*)
ance Full path invarianceb (L(X)Y = IN, GA = IN,
166.01 47 0.05 0.93 0.95
BE = IN)
Note: *IN = invariance.

Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
58 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012

Figure 3. The final model to predict low-carbon tourism behavior: Note: * p < 0.05, ** p < 0.01

invariance. Since the chi-square differences
between the non-restricted model and the full
metric invariance model were not significant,
the full metric invariances was supported (Δχ2(5)
= 3.46, p > 0.01). These findings indicated that
factor loadings were equivalent. Accordingly,
the full metric invariance models were used in
subsequent analyses. All non-restricted CFA
models and full metric invariance models
showed an excellent fit to the data (Table 6).
4.2.2.3. Structural Invariance
As a next step, structural invariance was tested.
The baseline models were generated by run-
ning the entire structural models, which were
rooted in the full and partial metric invariance
models. In addition, the constrained models
(full path invariance models) in which all
causal paths were fixed to be invariant across
groups were generated (Table 6). All models
showed satisfactory fits. Tests for chi-square
differences between the baseline models and the
constrained models were conducted to ensure
the equality of path coefficients (Yoo, 2002).
The results showed that because the chi-square
differences between the baseline models and
the constrained models were all significant, full
structural invariances were not supported for
the perceived behavioral control groups (Δχ2(10)
= 61.38, p < 0.01). This finding indicated that
paths across groups differed or at least some
of the paths were not equivalent.
a. Chi-square difference test: Δχ2(5) = 3.46, p
> 0.01 (non-significant), thus full metric
invariance is supported.
b. Chi-square difference test: Δχ2(10) = 61.38,
p < 0.01 (significant), thus full structural
invariance is not supported, and paths
across two groups are different.
4.2.2.4. Invariance Test for the Path from
Behavioral Intention and Preferred Behavior
A more rigorous test was performed to as-
sess the invariance of a specific path because
the overall cross-group invariance test of the
structural model did not allow us to examine
invariance in the specific parameter of inter-
est across groups. The particular parameter
of interest (behavioral intention -> preferred
behavior) in the nested models was constrained
to be equal across groups, and all paths in the

Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012 59
baseline models were able to be freely estimated.
Findings for the invariance test for the path as
expected, significant chi-square differences
across groups (Δχ2 = 7.75, Δdf = 1, p < 0.01)
was found, indicating that both positive and
negative components of perceived behavioral
control barriers has a significant moderating
role in the relationship between behavioral
intention and preferred behavior. The coef-
ficient value of strong perceived behavioral
control group (β32 = 0.23, t = 3.05, p < 0.01)
was higher than the values of weak perceived
behavioral control group (β32 = 0.07, t = 1.14,
p > 0.05). This result supported H6: Perceived
behavioral control toward “low-carbon tour-
ism” behavior has moderating effect between
“low-carbon tourism” behavior intention and
preferred behavior. Figure 3 presents a structure
model of the final hypothesis.
5. CONCLUSION AND
SUGGESTION
The current research proposed and tested a
theoretical framework for understanding the
relationships among attitude, subjective norms,
perceived behavioral control, behavioral inten-
tion, and preferred behavior toward “low-carbon
tourism.” The results of the SEM addressed
the effect of the three predicting factors on
behavioral intention and preferred behavior.
The proposed moderating impact of perceived
behavioral control was verified by testing for
metric invariances.
The study results find that tourists’ at-
titude and perceived behavioral control are
significant positive impact on their behavioral
intentions. In addition, the higher behavioral
intention, the higher preferred behavior toward
“low-carbon tourism.” These are as the same as
Ajzen’s expectances. Subjective norms do not
have significantly effect on behavioral inten-
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
tions. This does not mean it’s not an important
predicting factor. But in the study situation,
mentioned to people’s behavioral intentions
toward involve in low-carbon tourism, their
personal self-awareness is high, thus they would
not be susceptible influenced by the opinions
of other masses.
In addition, past travel experiences play
important role in this study, as the same results
as Sönmez and Graefe (1998) and Lam and
Hsu (2006). Anyone would have perceptions
of carbon reduction in past travel experiences,
this leads them more or less have tendency
toward “low-carbon tourism”. Experiences
indeed associate with perceived behavioral
control, as the same results as Cestac, Paran,
and Delhomme (2011). So these experiences can
increase personal perceived behavioral control,
it’s really a possible antecedent of perceived
behavioral control.
A significant finding in this research is
that “perceived behavioral control” plays an
important moderating effect between behavioral
intentions and preferred behavior. This result
is as the same as findings of Fitch and Ravlin
(2005), which indicated that PBC as a modera-
tor of the intention/behavior relationship, they
suggested that PBC may be more relevant in
situations where the actor has acquired some
familiarity with specific situational contingen-
cies. Specifically, perceived behavioral control
plays like a catalyst. Even though behavioral
intention sure can promote behavior occurred,
perceived behavioral control strengthens the
relationship.
ACKNOWLEDGMENTS
The authors would like to thank the anonymous
reviewers and the editor for their insightful
comments and suggestions. This research was
60 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012
supported by National Science Council, Taiwan
(Program No.: NSC99-3113-S-003-004-NE).
REFERENCES
Ajzen, I. (1985). From intentions to actions: A theory
of planned behavior. In Kuhl, J., & Beckman, J.
(Eds.), Action-control: From cognition to behavior
(pp. 11–39). Heidelberg, Germany: Springer-Verlag.
doi:10.1007/978-3-642-69746-3_2
Ajzen, I. (1988). Attitudes, personality and behavior.
Milton Keynes, UK: Open University Press.
Ajzen, I. (1991). The theory of planned behavior.
Organizational Behavior and Human Decision
Processes, 50, 179–211. doi:10.1016/0749-
5978(91)90020-T
Ajzen, I. (2002). Constructing a TPB questionnaire:
Conceptual and methodological considerations.
Retrieved from http://people.umass.edu/aizen/pdf/
tpb.measurement.pdf
Ajzen, I., & Fishbein, M. (1980). Understanding at-
titudes and predicting social behavior. Upper Saddle
River, NJ: Prentice Hall.
Ajzen, I., & Fishbein, M. (2005). The influence of
attitudes on behavior. In Albarracín, D., Johnson, B.
T., & Zanna, M. P. (Eds.), The handbook of attitudes
(pp. 173–221). Mahwah, NJ: Lawrence Erlbaum.
Amireault, S., Godin, G., Vohl, M.-C., & Pérusse,
L. (2008). Moderators of the intention-behavior and
perceived behavioral control-behavior relationships
for leisure-time physical activity. The International
Journal of Behavioral Nutrition and Physical Activ-
ity, 5(7), 1–11.
Anderson, J. C., & Gerbing, D. W. (1988). Structural
equation modeling in practice: A review and recom-
mended two-step approach. Psychological Bulletin,
103(3), 411–423. doi:10.1037/0033-2909.103.3.411
Andrews, K., Silk, K., & Eneli, I. (2010). Parents
as health promoters: A theory of planned behavior
perspective on the prevention of childhood obesity.
Journal of Health Communication, 15(1), 95–107.
doi:10.1080/10810730903460567
Armitage, C. J., & Conner, M. (2001). Efficacy of
the theory of planned behavior: A meta-analytic
review. The British Journal of Social Psychology,
40, 471–499. doi:10.1348/014466601164939
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of
structural equation models. Academy of Marketing
Science, 16(1), 76–94. doi:10.1007/BF02723327
Buckly, R. (2002). Tourism ecolabels. Annals of
Tourism Research, 29(1), 183–208. doi:10.1016/
S0160-7383(01)00035-4
Calder, B. J., Phillips, L. W., & Tybout, A. M. (1982).
The concept of external validity. The Journal of Con-
sumer Research, 9(3), 240–244. doi:10.1086/208920
Cestac, J., Paran, F., & Delhomme, P. (2011). Young
drivers” sensation seeking, subjective norms, and
perceived behavioral control and their roles in
predicting speeding intention: How risk-taking
motivations evolve with gender and driving experi-
ence. Safety Science, 49(3), 424–432. doi:10.1016/j.
ssci.2010.10.007
Fielding, K. S., Terry, D. J., Masser, B. M., Bordia,
P., & Hogg, M. A. (2005). Explaining landholders”
decisions about riparian zone management: The
role of behavioral, normative, and control beliefs.
Journal of Environmental Management, 77, 12–21.
doi:10.1016/j.jenvman.2005.03.002
Fishbein, M., & Ajzen, I. (1975). Belief, attitude,
intention, and behavior: An introduction to theory
and research. Reading, MA: Addison-Wesley.
Fitch, J. L., & Ravlin, E. C. (2005). Willpower and
perceived behavioral control: Influences on the
intention-behavior relationship and post behavior
attributions. Social Behavior and Personality, 33(2),
105–124. doi:10.2224/sbp.2005.33.2.105
Fornell, C., & Larcker, D. F. (1981). Evaluating
structural equation models with unobservable vari-
ables and measurement error. Journal of Marketing
Research, 18(1), 39–50. doi:10.2307/3151312
Fraser, R. T., Johnson, K., Hebert, J., Ajzen, I.,
Copeland, J., Brown, P., & Chan, F. (2010). Un-
derstanding employers’ hiring intentions in relation
to qualified workers with disabilities: Preliminary
findings. Journal of Occupational Rehabilitation,
20, 420–426. doi:10.1007/s10926-009-9220-1
Garcia, K., & Mann, T. (2003). From ‘I wish’ to ‘I
will’: Social-cognitive predictors of behavioral inten-
tions. Journal of Health Psychology, 8(3), 347–360.
Hagger, M. S., & Chatzisarants, N. L. D. (2005).
First- and higher-order models of attitudes, norma-
tive influence, and perceived behavioral control
in the theory of planned behavior. The British
Journal of Social Psychology, 44, 513–535.
doi:10.1348/014466604X16219
 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012 61
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black,
W. C. (1998). Multivariate data analysis (5th ed.).
Upper Saddle River, NJ: Prentice Hall.
Kink, N., & Hess, T. (2008). Search engines as
substitutes for traditional information sources? An
investigation of media choice. The Information Soci-
ety, 24(1), 18–29. doi:10.1080/01972240701771630
Lam, T., & Hsu, C. H.-C. (2006). Predicting behav-
ioral intention of choosing a travel destination. Tour-
ism Management, 27(4), 589–599. doi:10.1016/j.
tourman.2005.02.003
Lee, C.-K., Song, H.-J., Bendle, L. J., Kim, M.-J., &
Han, H. (2012). The impact of non-pharmaceutical
interventions for 2009 H1N1 influenza on travel
intentions: A model of goal-directed behavior.
Tourism Management, 33(1), 89–99. doi:10.1016/j.
tourman.2011.02.006
Manstead, A. S. R., & Parker, D. (1995). Evaluat-
ing and extending the theory of planned behavior.
European Review of Social Psychology, 6, 69–96.
doi:10.1080/14792779443000012
Norman, P., Conner, M., & Bell, R. (2000). The
theory of planned behavior and exercise: Evidence
for the moderating role of past behavior. Brit-
ish Journal of Health Psychology, 5, 249–261.
doi:10.1348/135910700168892
Nunnally, J. C., & Bernstein, I. H. (1994). Psycho-
metric theory (3rd ed.). New York, NY: McGraw-Hill.
Peterson, R. A. (2001). On the use of college students
in social science research: Insights from a second-
order meta-analysis. The Journal of Consumer
Research, 28(3), 450–461. doi:10.1086/323732
Quintal, V. A., Lee, J. A., & Soutar, G. N. (2010).
Risk, uncertainty and the theory of planned behavior:
A tourism example. Tourism Management, 31(6),
797–805. doi:10.1016/j.tourman.2009.08.006
Reber, A. S., Allen, R., & Reber, E. S. (2009). The
Penguin dictionary of psychology. London, UK.
Penguin Books.
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Sheeran, P., & Abraham, C. (2003). Mediator of
moderators: temporal stability of intention and
the intention-behavior relation. Personality and
Social Psychology Bulletin, 29(2), 205–215.
doi:10.1177/0146167202239046
Simpson, M. C., Gössling, S., Scott, D., Hall, C. M.,
& Gladin, E. (2008). Climate change adaptation and
mitigation in the tourism sector: Frameworks, tools
and practices. Paris, France: UNEP, University of
Oxford, UNWTO, WMO.
Sönmez, S. F., & Graefe, A. R. (1998). De-
termining future travel behavior from past
travel experience and perceptions of risk and
safety. Journal of Travel Research, 37(2), 171–177.
doi:10.1177/004728759803700209
Sparks, B. (2007). Planning a wine tourism vaca-
tion? Factors that help to predict tourist behavioral
intentions. Tourism Management, 28(5), 1180–1192.
doi:10.1016/j.tourman.2006.11.003
Steenkamp, J. M., & Baumgartner, H. (1998). As-
sessing measurement invariance in cross-national
consumer research. The Journal of Consumer Re-
search, 25, 78–90. doi:10.1086/209528
Stratford, E., & Davidson, J. (2002). Capital assets
and intercultural borderlands: Socio-cultural chal-
lenges for natural resource management. Journal
of Environmental Management, 66, 429–440.
doi:10.1006/jema.2002.0597
Yang-Wallentin, F., Schmidt, P., Davidov, E., &
Bamberg, S. (2003). Is there an interaction effect
between intention and perceived behavioral control?
Methods of Psychological Research, 8(2), 127–157.
Yoo, B. (2002). Cross-group comparisons: a caution-
ary note. Psychology and Marketing, 19, 357–368.
doi:10.1002/mar.10014
Yuksel, A., & Yuksel, F. (2002). Measurement of tour-
ist satisfaction with restaurant services: A segment-
based approach. Journal of Vacation Marketing, 9(1),
52–68. doi:10.1177/135676670200900104
62 International Journal of Technology and Human Interaction, 8(4), 45-62, October-December 2012

Nae-Wen Kuo is a professor of the Department of Geography, National Taiwan Normal Univer-
sity. He received his PhD in Environmental Management from the National Taiwan University in
2000. His research expertise is about environmental planning and management, Environmental
Impact Assessment (EIA), Health Impact Assessment (HIA), Industrial Ecology, Material Flow
Analysis, Life Cycle Assessment (LCA), sustainable tourism management, and ecotourism. He
has many academic performances, which were published in SCI/SSCI/EI journals and other
international journals, such as Journal of Cleaner Production, Environmental Monitoring and
Assessment, Renewable Agriculture and Food Systems, Tourism Management, and Environmental
Impact Assessment Review. He was also selected as Marquis Who’s Who in the World, 2009-
2012 (Marquis), Marquis Who’s Who in Asia, 2007-2012 (Marquis), and Leading Engineers of
the World, 2007 (IBC).

You-Yu Dai is a doctoral candidate of the Graduate Institute of Recreation, Tourism, and
Hospitality Management, National Chiayi University. He holds a MS in Graduate Institute of
Tourism and Health Science, National Taipei University of Nursing and Health Sciences; and a
BS in Department of Psychology, National Chung Cheng University. His research interests are
in sustainable tourism management, human resource management in hospitality, and tourism
and health sciences. He had attended and practiced many academic and education programs
about ecotourism, low-carbon tourism, and Life Cycle Assessment (LCA) in Taiwan. You-Yu
Dai has some conference theses be announced in United States, Taiwan, Australia, and China.
He also has published in journals such as Journal of Health Promotion and Health Educa-
tion (in Chinese), Journal of Sport, Leisure and Hospitality Research (in Chinese), Journal of
Management Practices and Principles (in Chinese), Journal of Management Practices and Prin-
ciples (in Chinese), and others.

Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.