Documente Academic
Documente Profesional
Documente Cultură
A R T I C L E I N F O A B S T R A C T
Keywords: The current study sought to extend the theory of planned behavior (TPB), which is rooted in the theory of
Green hotel reasoned action (TRA), to more comprehensively explain the formation of customers intention to revisit
Extended theory of planned behavior a green hotel. In particular, the extended TPB incorporates the critical constructs in the consumer
Revisit intention behavior and marketing literature (i.e., service quality, customer satisfaction, overall image, and
Decision-making
frequency of past behavior) into the original TPB model. Results of a structural analysis revealed that the
new model provides a better t with the data, and explains signicantly greater amounts of variance in
revisit intention in comparison to the TRA and TPB. Added constructs in the new model considerably
contribute to improve our understanding of the complicated process of green hotel customers decision-
making. In this study, all relationships appeared to be signicant as conceptualized according to the
theory. In addition, a mediating effect of satisfaction and attitude was found. The article includes
discussions on theoretical and managerial implications.
2010 Elsevier Ltd. All rights reserved.
0278-4319/$ see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijhm.2010.01.001
660 H. Han, Y. Kim / International Journal of Hospitality Management 29 (2010) 659668
2006; Han et al., 2010). Given this, the current study employed the
theory of planned behavior (TPB), an extension of the theory of
reasoned action (TRA) (Ajzen, 1991; Ajzen and Fishbein, 1980), in
order to gain a better understanding of green hotel customers
post-purchasing decision-making formation.
According to Ajzen (1991) and Perugini and Bagozzi (2001),
modifying the TPB model by altering paths and including
additional critical constructs in a certain context often contribute
to enhancing our understanding of the theoretical mechanism of
the model and increasing the prediction power for individuals
intention/behavior in that specic context. That is, the theory can
be broadened and deepened through such a process (Ajzen, 1991;
Perugini and Bagozzi, 2001). There is growing empirical evidence
that service quality, customer satisfaction, overall image, and
frequency of past behavior are critical to the decision-making
process and powerful predictors of post-purchasing decisions (e.g.,
Barsky, 1992; Han and Ryu, 2006; Morgan and Hunt, 1994; Oh and
Hsu, 2001). These studies stressed the signicance of the variables
in explaining customer post-purchase behaviors. Thus, the current
study also attempted to extend the TPB model by including such
Fig. 1. Proposed extended TPB model for green hotel repurchasing behavior.
constructs, which are important in a hospitality context, as service
quality, customer satisfaction, overall image, and frequency of past
behavior, and by altering the path in the model to improve our Jauhari, 2007; Wolfe and Shanklin, 2001). That is, a growing
ability to predict intention and understanding of green hotel number of hotels are joining the green movement to reduce
customers decision formation. harmful impacts on the environment, and thereby in doing so
eventually increasing their protability (e.g., cost savings and
1.1. Purpose of the study customer attraction/retention) (Pizam, 2009; Wolfe and Shanklin,
2001). The term green is alternatively known as eco-friendly,
The TPB has been applied in various contexts, but little research environmentally friendly, or sustainable (Han et al., 2009;
has employed the TPB to explain hotel customers decision-making Pizam, 2009). Wolfe and Shanklin (2001) indicated that green
process. In addition, no research has yet compared the predictive refers to actions that decrease negative impacts on our environ-
abilities of the TRA, TPB, and extended TPB, including additional ment (e.g., recycling, eco-purchasing). Similarly, according to the
constructs in a hotel setting, particularly in a green hotel context. Green Hotel Associations (GHA) (2009) and DEP (2001), a green
Further, while the signicance of service quality, customer hotel is an eco-friendly lodging property that has implemented
satisfaction, overall image, and past behavior in explaining post- various green practices and institutes sound and environmentally
purchasing behaviors have been stressed in various contexts, to the friendly programs to protect the environment and reduce
best of our knowledge no research has integrated these variables operational costs. In particular, in green hotels, the following
into the TPB model to predict customer intentions/behaviors. are quite commonly used, practiced, and served: durable service
Therefore, the purpose of this study was to develop an extended items, cotton towels and linens for air quality, donations to charity,
TPB model by taking such variables as service quality, customer well-educated staff about green practices, energy conservation,
satisfaction, overall image, and frequency of past behavior into environmental cleaning, eco-friendly/organic foods, fresh air,
account in order to better predict green hotel customers revisit water recycling/conservation, recycling bins, towel re-use pro-
intention. The specic objectives were to compare the predictive gram, etc. (GHA, 2009; DEP, 2001).
ability of the TRA, TPB, and modied TPB models, to investigate the
relationships among the proposed study constructs, and to 2.2. Theory of planned behavior
examine the mediating role of customer satisfaction, attitude,
subjective norms, and perceived behavioral control. Ajzen and Fishbein (1980), across a number of publications,
developed and explicated the TRA in order to account for
1.2. Proposed research model mechanisms of human behaviors in decision processes (e.g., Ajzen,
1985; Ajzen and Fishbein, 1980; Fishbein and Ajzen, 1975). The
The modied TPB model is presented in Fig. 1. The model TRA was specically designed to predict human behaviors under
includes the original variables in the TPB and new constructs (i.e., complete volitional control. That is, this theory assumed that most
service quality, customer satisfaction, overall image, and frequency of individuals decisions/behaviors are derived from the intensity
of past behavior). The bold lines indicate the newly added paths on of volitional efforts for the specic decisions/behaviors. According
the original TPB model. to this theory, individuals are rational and motivation-based in
their decision-making processes, and make a reasoned choice
2. Review of the literature among various alternatives (Fishbein and Ajzen, 1975). In their
investigation of the effectiveness of TRA, which involved use of a
2.1. Green hotels meta-analysis, Sheppard et al. (1988) veried the predictive power
of TRA. Their ndings indicated that the model accurately
Environmental protection continually attracts public attention predicted individuals decisions and behaviors when applied in
(Chan and Wong, 2006). Individuals are becoming more and more any situation or to any activity. Because of its accuracy in
aware of the environmental damage caused by various business predicting human behaviors and effectiveness in explaining
activities (Manaktola and Jauhari, 2007). This increasing public psychological processes in decision-making, the TRA is widely
concern is stimulating the implementation of environmentally utilized in comprehending customers decision-making processes
responsible management in the hotel industry (Manaktola and in various contexts.
H. Han, Y. Kim / International Journal of Hospitality Management 29 (2010) 659668 661
satisfaction, and intention. They found that service quality is a studies across diverse elds, and found strong evidence for the
signicant predictor of customer satisfaction in forming behavioral critical role of frequency of past behavior in forming intention and
intention. Ting (2004) examined customer behaviors in a service behavior. In addition, in testing the competing models that use the
setting and found that service quality predicts customer satisfac- TRA, TPB, and extended TPB, Lee and Back (2009) veried that past
tion in the decision-making process. Consistent with these behavior is an important determinant of intention and behavior.
ndings, in a restaurant setting, Han and Ryus (2006) ndings Further, Perugini and Bagozzi (2001) developed a modied model
indicated that service quality affects satisfaction, and satisfaction of the TPB, a model of purposive behavior, and found that
fully mediates the effect of quality on intention to revisit and be frequency of past behavior plays a signicant role in decision-
engaged in positive word-of-mouth behavior. In addition, in their making processes along with other added constructs in the original
integrative approach to understanding attitude-behavior relation- TPB model. Specically, their ndings showed that frequency of
ships, Bansal and Taylor (1999), in developing a service provider past behavior contributes directly to desires and behavioral
switching model, found that service quality is an important intention in both body weight regulation and studying effort
inuence on attitude toward switching and satisfaction, such that settings. Ajzen (1991) insisted that frequency of past behavior can
including service quality and satisfaction in their model was be utilized to assess the completeness/sufciency of a model
essential to the precise prediction of an individuals intention. In designed to predict decision/behavior. According to Ajzen, if
their study, both quality and satisfaction (mediator) contributed to frequency of past behavior explains a decision/behavior in a
a decrease in intention to switch. Given this evidence, it seems certain setting after the original model variables have been
appropriate to incorporate service quality and satisfaction into a accounted for, this construct may be employed to better predict
model designed to explain the formation of repurchasing intention such a decision/behavior in a particular setting. Indeed, many
in order to ensure a sounder theoretical underpinning. researchers in various elds have attempted to include this
variable as a determinant of intention and enhanced the predictive
2.3.2. Overall image power of the TPB (e.g., Lee and Back, 2009; Oh and Hsu, 2001;
The denition of overall image varies in the marketing and Perugini and Bagozzi, 2001; Ryu and Jang, 2006).
consumer behavior literature. According to one of the clearest
denitions, individuals total perceptions of a product (or service) 2.3.4. Relationship between subjective norm and attitude
and its salient attributes are generated by processing information A possible link between subjective norm and attitude toward the
from various sources (Assael, 1984; Bloemer and Ruyter, 1998). behavior was proposed and tested in previous studies (e.g., Chang,
Researchers generally agree that the formation of an overall image 1998; Han et al., 2010; Ryu and Jang, 2006). Chang (1998) developed
about a product/service is based on a cognitive and perceptual a modied model of the TPB and found that adding the path from
process (Assael, 1984; Bloemer and Ruyter, 1998; Han et al., 2009). subjective norm to attitude was essential to improving the structural
Consistent with these studies, the concept of overall image in the model t. In a hospitality setting, Ryu and Jang (2006) found that
current research is lodging customers overall perceptions of a perceived subjective norm is positively associated with ones
green hotel and its signicant attributes as shaped by information attitude toward a certain behavior. Similarly, in examining hotel
from diverse sources and by prior or explicit knowledge about customers eco-friendly behaviors, Han et al. (2010) veried that
them (Han et al., 2009). attitude toward a green hotel is a positive function of a subjective
Overall image has increasingly attracted the attention of norm; this added path in the original TPB signicantly improved the
academic and industry professionals and is regarded as playing t of the model. Given this evidence, the current study assumed that
an important role in predicting a consumers favorable/unfavor- ones perceived social pressure from salient referents to engage in a
able decisions. Many researchers across various elds have certain behavior induces his/her favorable/unfavorable evaluation
identied the signicance of overall image in customers deci- of a behavior. Thus, the path from subjective norm to attitude was
sion-making processes (Han et al., 2009; Prendergast and Ho, added to the proposed model.
2002; Ryu et al., 2007). Prendergast and Ho (2002) investigated the In summary, a modication of the TPB is associated with the
impact of store image, and found that a rms image is a critical introduction of additional variables explaining how they inuence
driving force in intention. In their empirical investigation of the intention in relation to existing predictors and with the revision of
role of overall image in a hospitality setting, Ryu et al. (2007) stated any path in the original model. Accordingly, the proposed extended
that a positive overall image of a hospitality rm increases model completes the systematic integration of four key factors into
customers willingness to revisit and spread favorable word-of- the TPB, considering their possible relations with the existing TPB
mouth. In addition, Han et al.s (2009) recent research on hotel variables, and included a specic path in order to better comprehend
customers behaviors indicated that ones overall perception of a green hotel customers complicated decision-making process. In the
green hotel (overall image) is positively associated with his/her current study, the following 12 hypotheses were developed:
intention to buy a green hotel product, to recommend it, and to pay
for it. Considering these empirical studies, it seems likely that H1. Behavioral beliefs have a positive inuence on attitude.
overall image may play an important role in predicting a green H2. Normative beliefs have a positive inuence on subjective
hotel customers intention to revisit the hotel when this construct norms.
is integrated into the TPB model.
H3. Control beliefs have a positive inuence on perceived behav-
2.3.3. Frequency of past behavior ioral control.
Frequency of past behavior is believed to be another signicant
predictor of intention/behavior (e.g., Lee and Back, 2009; Oh and H4. Attitude has a positive inuence on revisit intention.
Hsu, 2001; Quellette and Wood, 1998; Perugini and Bagozzi, 2001;
H5. Subjective norm has a positive inuence on revisit intention.
Ryu and Jang, 2006). While past behavior was not originally
included in the TRA/TPB, numerous researchers have identied the H6. Perceived behavioral control has a positive inuence on revisit
need to add this construct equivalently to other predictors of intention.
intention or future behavior (e.g., Lee and Back, 2009; Oh and Hsu,
2001; Quellette and Wood, 1998; Perugini and Bagozzi, 2001). In H7. Service quality has a positive inuence on customer
their meta-analysis, Quellette and Wood (1998) examined 64 satisfaction.
H. Han, Y. Kim / International Journal of Hospitality Management 29 (2010) 659668 663
H8. Service quality has a positive inuence on attitude. 3.1.3. Construct measures for other study variables
Items on attitude, subjective norm, perceived behavioral
H9. Customer satisfaction has a positive inuence on revisit in- control, revisit intention, service quality, customer satisfaction,
tention. overall image, and frequency of past behavior were based on the
existing validated measures from the previous literature (e.g.,
H10. Overall image has a positive inuence on revisit intention.
Ajzen, 1988, 1991; Ajzen and Fishbein, 1980; Baloglu and
H11. Frequency of past behavior has a positive inuence on revisit McCleary, 1999; Han and Ryu, 2006; Lam and Hsu, 2004, 2006;
intention. Lee and Back, 2007; Taylor and Baker, 1994). The measures were
slightly modied for use in a questionnaire suitable for a green
H12. Subjective norm has a positive inuence on attitude. hotel setting. In particular, 6 items and a 7-point semantic
differential scale were employed to assess attitude (e.g., For me,
revisiting this green hotel when traveling to the same location next
3. Methodology time is 1 = extremely bad, 7 = extremely good). The measure of
frequency of past behavior was assessed using 5 frequency
3.1. Questionnaire development categories (e.g., Which statement best describes how many times
you have stayed at this hotel 1 = 1 time, 2 = 23 times, 3 = 45
3.1.1. Elicitation study times, 4 = 610 times, 5 = more than 10 times). Three items and a
According to Ajzen (2009) and Ajzen and Fishbein (1980), there 7-point Likert-type scale were used to measure overall image
is no standard questionnaire for TPB. With regard to belief (e.g., Overall image for staying at this green hotel is 1 = very
constructs, they insisted that formative research (i.e., elicitation negative, 7 = very positive). Other study constructs were assessed
study) and validation of the theorys belief constructs were needed using a 7-point Likert-type scale ranging from strongly disagree (1)
prior to construction of the nal questionnaire. They indicated that to strongly agree (7). In addition, multi-item scales were used to
such an endeavor helps researchers construct a questionnaire that assess subjective norm (3 items) (e.g., Most people who are
is adequate for a specic behavior and population of interest. important to me think I should stay at this green hotel when
Accordingly, an elicitation study and pilot-test were conducted in traveling to the same location next time), perceived behavioral
the present study. As an elicitation method, a focus group was control (3 items) (e.g., Whether or not I stay at this hotel when
employed (Cheng et al., 2006; Lam and Hsu, 2004; Lee and Back, traveling to the same location next time is completely up to me),
2007). The focus group participants were green hotel customers, revisit intention (3 items) (e.g., I am willing to revisit this hotel
hospitality academics, and hotel industry professionals. To obtain when traveling to the same location next time), service quality (2
the new set of belief items, salient beliefs and referents were items) (e.g., I believe this green hotel offers excellent service),
discussed in the focus group, and an open-ended eliciting and customer satisfaction (3 items) (e.g., I am very satised with
questionnaire was completed. The initial questionnaire, which the overall service at this green hotel).
included 12 items for beliefs (6 items for behavioral beliefs; 3 items
for normative beliefs; and 3 items for control beliefs) and 12 items 3.2. Data collection
for evaluative components, was generated through this process
and a literature review. The generated questionnaire was rened An Internet-based survey was used to collect data. A question-
through hospitality experts reviews. The salient referents identi- naire was sent to 4500 general hotel customers in the U.S. through
ed were: family/relatives, friends, and colleagues/co-workers. As a survey system available via an on-line market research company.
a next step, a pilot-test was performed with 62 hotel customers. In the opening instructions of the survey, a detailed description of a
The results indicated that the instrument had a sufcient level of green hotel was provided along with a list of green hotels in each
reliability and validity. state certied according to well-recognized and -respected
national/regional organizations (e.g., U.S. Green Building Council
3.1.2. Construct measures for beliefs Leadership in Energy and Environmental Design (LEED) and the
A six-item scale was used with a 7-point Likert-type scale Green Seal). The hotels on the list are believed to actively follow
ranging from strongly disagree (1) to strongly agree (7) for measures green guidelines and ideas. In addition, before completing the
of behavioral beliefs (e.g., Staying at this green hotel when questionnaire, two screening questions were given to the
traveling to the same location next time would enable me to be participants (i.e. (1) Have you ever stayed at a green hotel?;
more socially responsible.) and very unimportant (1) to very (2) If yes, when was the most recent stay at a green hotel?). Those
important (7) for measures of outcome evaluations (e.g., Being respondents who had stayed at a green hotel within the last 6
more socially responsible is 1 = very unimportant, 7 = very months were asked to complete the questionnaire. According to
important). Three items with a 7-point Likert-type scale ranging Keaveney (1995), the time frame (6 months) is recent enough to
from very false (1) to very true (7) for measures of normative beliefs ensure that customers will reliably recall their service experi-
(e.g., My family (or relatives) thinks I should stay at this hotel ences. The participants who had stayed at a green hotel not on the
when traveling to the same location next time) and extremely list were also free to ll out the questionnaire. The respondents
unlikely (1) to extremely likely (7) for measures of motivation to were also asked to indicate the name of that green hotel. A total of
comply (e.g., Generally speaking, how likely are you to do what 469 responses were received from the survey participants, for a
your family (or relatives) thinks you should do?) were utilized. response rate of 10.42%. An Internet-based survey is considered a
Finally, a three-item measurement with 7-point Likert-type scale more immediate and effective means of response (Kim, 2001).
was used to assess control beliefs (e.g., Staying at this hotel is With the development of the Internet, increasing numbers of
expensive) and each items control power (e.g., The price/cost of researchers have used web-based survey (e.g., Asperin, 2007; Han
staying at a green hotel would inuence my decision for hotel et al., 2009; Kim and Ok, 2009; Kim et al., 2009; Yang and Peterson,
selection.), ranging from strongly disagree (1) to strongly agree (7). 2004). The response rates of these recent studies employing an on-
In order to gain an overall level of three belief constructs, items for line survey range from 8.1% to 17.1%. The response rate of the
each belief construct were multiplicatively combined with their present study is reasonable given the rates predicted by these
evaluative components (SBBiOEi, SNBjMCj, and SCBkPPk) (Ajzen, recent studies. A panel of experts from hotels and academic
1991, 2009; Ajzen and Fishbein, 1980). institutions reviewed the hotels named in the completed ques-
664 H. Han, Y. Kim / International Journal of Hospitality Management 29 (2010) 659668
Table 1
Construct summary statistics (N = 434).
Constructs SQ BB NB CB CS AT SN PBC OI RI
SQ 1.000
BB .634 (.402) 1.000
NB .452 (.204) .387 (.150) 1.000
CB .084 (.007) .227 (.052) .106 (.011) 1.000
CS .537 (.288) .366 (.134) .442 (.195) .147 (.022) 1.000
AT .627 (.393) .582 (.339) .357 (.127) .129 (.017) .404 (.163) 1.000
SN .529 (.280) .443 (.196) .738 (.545) .042 (.002) .509 (.259) .498 (.248) 1.000
PBC .438 (.192) .422 (.178) .223 (.050) .186 (.035) .248 (.062) .273 (.075) .233 (.054) 1.000
OI .789 (.623) .715 (.511) .394 (.155) .160 (.026) .508 (.258) .683 (.466) .498 (.248) .371 (.138) 1.000
RI .685 (.469) .675 (.456) .495 (.245) .147 (.022) .569 (.324) .669 (.448) .638 (.407) .465 (.216) .783 (.613) 1.000
Mean 5.295 35.225 16.015 32.679 4.045 5.543 3.941 5.545 5.677 5.303
SD 1.322 6.190 5.399 5.003 1.222 1.231 1.167 1.196 1.322 1.327
Composite reliability .900 .842 .837 .660 .949 .823 .913 .644 .937 .744
AVE .895 .829 .824 .601 .947 .779 .909 .584 .934 .712
Alpha .912 .967 .933 .814 .959 .956 .967 .787 .974 .888
Note 1. SQ: service quality; BB: behavioral beliefs; NB: normative beliefs; CB: control beliefs; CS: customer satisfaction; AT: attitude; SN: subjective norm; PBC: perceived
behavioral control; OI: overall image; RI: revisit intention.
Note 2. Frequency of past behavior (FPB) was not included in the table in that FPB was measured by using a single item; the mean and standard deviation of FPB were 3.136 and
.984, respectively.
a
Correlation coefcients are estimates from AMOS 5. Model measurement t: x2 = 1228.954, df = 512, p < .001, x2/df = 2.400, RMSEA = 0.057; CFI = 0.959; NFI = 0.931.
tionnaires. Of the 464 responses, 30 were disregarded because the average variance extracted (AVE), and the correlation between
panel determined the hotels named in these questionnaires were constructs. As shown in Table 1, all AVE values were above .50,
not considered to be green. In addition, some extreme cases indicating adequate convergent validity. In addition, the AVE value
identied through a test for multivariate outliers were eliminated. for each factor exceeded the squared value of each correlation
Specically, with the use of a p < .001 criterion for Mahalanobis between constructs, providing evidence of discriminant validity
distance, ve extreme multivariate outliers among the cases were (Fornell and Larcker, 1981).
found (Mahalanobis D (42) > 76.054, p < .001) and removed from
further analyses, retaining a nal sample of 434 cases. 4.3. Structural model
55.7% of the variance in revisit intention (see Fig. 4). Finally, the
extended TPB model exhibited a good t to the data
(x2 = 1492.186, df = 563, p < .001, x2/df = 2.650, RMSEA = 0.062;
CFI = .957; NFI = .938), and six antecedent variables explained
about 72.0% of the total variance in revisit intention (see Fig. 5).
Overall, the results of model evaluations implied that the TRA, TPB,
Fig. 5. Extended TPB model results for green hotel repurchasing behavior (N = 434).
and proposed extended model well predicted hotel customers
intention to revisit a green hotel.
As a next step, the TRA, TPB, and extended models were
compared to assess their ability to explain revisit intention and t 4.3.2. Hypotheses testing
statistics. Findings indicated that the extended model had superior Hypotheses 1, 2, and 3 were tested. The regression paths from
explanatory power to that found in the TRA and TPB model behavioral, normative, and control beliefs to attitude (b = .253,
(adjusted R2 for intention in the extended TPB = .720 vs. adjusted t = 4.917, p < .01), subjective norm (b = .745, t = 17.751, p < .01),
R2 for intention in the TRA = .519 and adjusted R2 for intention in and perceived behavioral control (b = .211, t = 3.640, p < .01)
the TPB = .557). Specically, the extended TPB model improved R2 were signicant. Thus, hypotheses 1, 2, and 3 were supported.
from .519 to .720 and from .519 to .720 in comparison with the TRA These ndings were consistent with expectancy-value theory and
and TPB models, respectively. In addition, the extended TPB model previous studies in various settings (e.g., Ajzen, 1991; Han et al.,
(x2/df = 2.650, RMSEA = 0.062, CFI = .957, NFI = .938) had relatively 2010; Lam and Hsu, 2004; Lee and Back, 2009). The ndings imply
better t statistics than the TRA (x2/df = 4.965, RMSEA = 0.076, that customers perceived probability of an expected outcomes
CFI = .914, NFI = .896) and TPB models (x2/df = 3.275, occurrence, perceived behavioral expectations of their important
RMSEA = 0.072, CFI = .942, NFI = .919). For decades, many referents, and perceived presence of resources and opportunities
researchers have attempted to rene and extend the TRA and contribute to building their favorable attitude toward revisiting a
TPB models by modifying the paths or by including new green hotel, inducing their perceived social pressure to perform
construct(s) to better understand a wider range of human the behavior, and stimulating their perception of the ease in
behaviors in various settings (Bansal and Taylor, 1999; Chang, engaging in the behavior, respectively. Hypotheses 4, 5, and 6
1998; Han et al., 2010; Oliver and Bearden, 1985; Ryu and Jang, were tested. The ndings indicated that attitude (b = .147,
2006; Taylor and Todd, 1995; Vallerand et al., 1992). Consistent t = 3.480, p < .01), subjective norm (b = .263, t = 7.047, p < .01),
with these studies, the current study ndings revealed that the and perceived behavioral control (b = .134, t = 3.901, p < .05)
proposed extended model, which included signicant constructs in were positively and signicantly associated with revisit intention,
the consumer decision-making process (service quality, customer supporting hypotheses 4, 5, and 6. These ndings aligned with
satisfaction, overall image, and frequency of past behavior), previous studies (e.g., Ajzen, 1991; Lam and Hsu, 2004, 2006; Lee
increased our prediction of intention and appeared to be superior and Back, 2009) implying that an increase in favorable attitude,
to the TRA and TPB models. Accordingly, this extended TPB model subjective norm, and perceived behavioral control will result in an
was used in hypotheses testing. increase in the likelihood of revisiting a green hotel. Subjective
norm was found to have the greatest direct effect on intention
among these three variables. That is, the intention to revisit a
green hotel was associated with hotel customers perceived social
pressure from their signicant referents. Hypotheses 7 and 8 were
tested. Results showed that service quality had a positive
inuence on satisfaction (b = .563, t = 12.140, p < .01) and
attitude (b = .412, t = 4.360, p < .01). Thus, hypotheses 7 and 8
were supported. This result indicates that as hotel customers
perceived service quality increases, so do their satisfaction levels,
as well as attitudes toward revisiting a green hotel. The regression
path from customer satisfaction and intention (b = .157, t = 4.347,
p < .01) was signicant. As expected, overall image was found to
be a signicant predictor of revisit intention (b = .285, t = 3.587,
p < .01). Therefore, hypotheses 9 and 10 were supported. The
ndings imply that increasing customers satisfaction level and
favorable overall image contribute to building strong intention to
Fig. 4. TPB model results (N = 434). revisit a green hotel. This result was in line with previous studies
666 H. Han, Y. Kim / International Journal of Hospitality Management 29 (2010) 659668
Table 3 5. Discussion
Standardized parameter estimatesextended TPB (N = 434).
Hypothesized paths Standardized estimates t-Values Results Little research has focused on green hotel customers decision-
H1: BB ! AT .253 4.917 **
Supported
making process. The current study sought to provide a deeper
H2: NB ! SN .745 17.751** Supported understanding of hotel customers intention to revisit a green hotel
H3: CB ! PBC .211 3.640** Supported by incorporating four critical constructsservice quality, satisfac-
H4: AT ! RI .147 3.480** Supported tion, overall image, and frequency of past behaviorinto the TPB
H5: SN ! RI .263 7.047** Supported
model. The extended model was tested using SEM, and there was
H6: PBC ! RI .134 3.901** Supported
H7: SQ ! CS .563 12.140** Supported strong support for the model. Specically, the study results
H8: SQ ! AT .412 4.360** Supported indicated that the proposed model had a satisfactory t to the data
H9: CS ! RI .157 4.347** Supported and the inclusion of these variables signicantly increased the
H10: OI ! RI .285 3.587** Supported predictive power of customers intention to revisit a green hotel.
H11: FPB ! RI .257 2.930** Supported
H12: SN ! AT .171 4.134** Supported
All twelve hypotheses in the study model were supported. The
ndings also indicated that satisfaction and attitude acted as
R2 (CS) .317
mediators between service quality and intention. Overall, the
R2 (AT) .470
R2 (SN) .556 current study achieved all study objectives.
R2 (PBC) .145
R2 (RI) .720 5.1. Theoretical and practical implications
Goodness-of-t statistics: x2 = 1492.186, df = 563, p < .001 (x2/df = 2.650)
RMSEA = .062 Study ndings hold both theoretical and practical implications.
CFI = .957 First, the prediction of green hotel customers revisit intention
NFI = .938
under the extended TPB model was well supported. The results of
Indirect impact on RI the modeling comparisons implied that both TRA and TPB are
SQ = .149** insufcient for comprehending intention formation in a green
BB = .037
hotel context, and that the extended TPB model represents a
NB = .105
SN = .025 substantial improvement over the TRA and TPB. Ajzen (1991, 2009)
CB = .028 stated that the modication of the TPB should occur cautiously,
Note 1. SQ: service quality; BB: behavioral beliefs; NB: normative beliefs; CB:
and provided some criteria. The extended model generally met the
control beliefs; CS: customer satisfaction; AT: attitude; SN: subjective norm; PBC: criteria for revising the TPB: similar to existing predictors of the
perceived behavioral control; OI: overall image; FPB: frequency of past behavior; RI: theory, the proposed variables are behavior-specic in compliance
revisit intention. with the principle of compatibility; the factors are conceptually
**
p < .01.
independent of existing constructs of the theory; the constructs are
regarded as casual factors that determine decisions; and nally,
the variables are potentially applicable to a broad range of
in different contexts (e.g., Bansal and Taylor, 1999; Han et al., behaviors in different settings (Ajzen, 1991, 2009). In particular,
2009; Han and Ryu, 2006; Taylor and Baker, 1994). In addition, added constructs in the current research (i.e., service quality,
frequency of past behavior was positively associated with revisit satisfaction, overall image, and frequency of past behavior)
intention (b = .257, t = 2.930, p < .01), supporting hypothesis 11. apparently provide sufcient impetus for the post-purchase
This nding is also consistent with the results of previous studies decision-making process, and a simultaneous analysis of these
(e.g., Lee and Back, 2009; Oh and Hsu, 2001; Quellette and Wood, constructs and original determinants of the TPB considerably
1998; Perugini and Bagozzi, 2001). Ajzen (1991) insisted that improved our understanding of the intricate process of green hotel
frequency of past behavior plays a critical role in decision-making customers revisit intention formation. Accordingly, beyond the
processes, especially when individuals deliberately make a simple application of the TPB to hospitality product purchasing
decision. Thus, in the present context, it would be true that activity, researchers should add variables as critical factors as they
customers consciously form an intention to revisit a green hotel. develop any theory that explains customers decision-making
Finally, the relationship between subjective norm and attitude process in both green hotel and hospitality contexts.
was tested. The hypothesized link was positive and signicant Second, this study found that service quality, satisfaction, and
(b = .171, t = 4.134, p < .01). Therefore, hypothesis 12 was overall image have a signicant positive association with revisit
supported. Based on this nding, it can be inferred that ones intention. From a practical standpoint, the ndings suggested that
social pressure from critical referents is important in explaining green hotel marketers should pursue the excellent attributes and
customers eco-friendly purchasing behaviors. This nding also services that induce customers positive evaluations and stimulate
supported empirical studies, which demonstrated that subjective the formation of favorable attitudes toward visits to green hotels.
norm has a direct impact on attitude toward a behavior (e.g., In addition, the ndings suggested that marketers should develop
Chang, 1998; Han et al., 2010; Randall and Gibson, 1991; Ryu and efcient strategies for enhancing their image by advertising their
Jang, 2006). Overall, all hypothesized paths were supported. The environment-friendly practices (e.g., donation of usable furniture/
ndings are summarized in Table 3. equipment to charitable organizations, recycling, conservation of
As a next step, the mediating roles of customer satisfaction, water and energy) to visiting and potential customers using
attitude, subjective norms, and perceived behavioral control were multiple information sources. These various efforts would
tested by examining the indirect effect of service quality, contribute to building customers strong intention to repurchase
behavioral beliefs, normative beliefs, subjective norms, and control a green hotel product.
beliefs on revisit intention. The ndings revealed that only service Third, frequency of past behavior was found to be a signicant
quality signicantly affected revisit intention through customer predictor of intention. That is, customers intention to revisit a green
satisfaction and attitude (bSQ-CS&AT-RI = .149, p < .01). This nding hotel becomes stronger with an increase in the number of prior visits
indicated that customer satisfaction and attitude have a signicant to the hotel. Ajzen (1991) which indicated that frequency of past
mediating role in the relationship between service quality and behavior should be utilized if the inclusion of this construct in the
revisit intention. TPB contributes to better prediction of decision-making in a specic
H. Han, Y. Kim / International Journal of Hospitality Management 29 (2010) 659668 667
setting after the original TPB constructs have been accounted for. current study ndings to other types of hotels (e.g., regular hotels)
The current study result veried the critical role of this construct in should occur cautiously. Future research should apply the
explaining green hotel customers post-purchase decision-making extended model to other types of hotels. Second, this study used
process. From a practical point-of-view, customers who have had a Web-based survey method to reach and sample a broader range
favorable experiences at a specic green hotel on a number of visits of hotel customers. Future studies should examine green hotel
are more willing to repurchase the eco-friendly hotel product. customers post-decision-making process in an actual green hotel
Accordingly, for those marketing a green hotel, the development of purchase setting to reduce extraneous variance and increase
more effective service strategies is essential to improving their internal validity. That is, a replication of the study in a eld setting
customers favorable experiences while staying at a hotel. is required in future studies. The third limitation lies in the
Fourth, as shown in Table 3, the present study result indicated response rate. While it is frequent to have a relatively low response
that among the four mediator variables in the proposed model, rate, such as 10%, when using a web-based survey (e.g., Asperin,
only customer satisfaction and attitude signicantly mediated the 2007; Han et al., 2009; Kim and Ok, 2009; Kim et al., 2009; Yang
impact of service quality on revisit intention (bSQ-CS&AT-RI = .149, and Peterson, 2004), increasing the response rate would enhance
p < .01). This nding implied that service quality shapes satisfac- the generalizability of the ndings of this study. Thus, future
tion and attitude toward a behavior, and these variables enhance studies should include more effective ways to induce Internet
green hotel customers favorable decisions. Accordingly, green users active responses. Finally, while the current study satisfacto-
hotel operators should seek to increase the level of customers rily extended the original TPB model, a broader variety of variables
attitudes and satisfaction in order to take full advantage of the (e.g., personal characteristics, value, self-concepts) should be
impact of service quality on decisions to repurchase a green hotel included in this study model. Such efforts would ensure more
product. It should also be noted that green attributes in a hotel comprehensive understanding of green hotel customers decision-
should not be seen as an alternative to quality of service. It may be making process.
true that green credentials can partially and occasionally
compensate for a poor quality product or service performance. Acknowledgement
Indeed, empirical evidences indicated that environmentally
responsible customers are willing to sacrice a little convenience This study was supported by research funds from Dong-A
on occasion (e.g., Dalton et al., 2008; Diekmann and Preisendorfer, University (No. 20090270).
2003). However, they are not prepared to experience consistent
inconveniences, continuously accepting lower product and service
References
performance levels (e.g., Dalton et al., 2008; Diekmann and
Preisendorfer, 2003; Manaktola and Jauhari, 2007). Thus, for green Ajzen, I., 1985. From intentions to actions: a theory of planned behavior. In: Kuhl,
hotel operators, maintaining a level of service quality in their J., Beskmann, J. (Eds.), Action Control: From Cognition to Behavior. Springer,
hotels that is at least comparable to that in non-green hotels is New York, pp. 1139.
Ajzen, I., 1988. Attitude, Personality, and Behavior. Dorsey Press, Chicago, IL.
critical for retaining customers. Ajzen, I., 1991. The theory of planned behavior. Organizational Behavior and Human
Lastly, as hypothesized, the subjective norm exerted a Decision Processes 50 (2), 179211.
signicant inuence on attitude toward revisiting a green hotel. Ajzen, I., 2009. A Theory of Planned Behavior. Retrieved from http://people.
umass.edu/aizen/tpb.html on June 15, 2009.
This nding indicated that attitudinal and normative factors of the Ajzen, I., Driver, B.L., 1992. Application of the theory of planned behavior to leisure
TPB were not independent. Some researchers argued for a possible choice. Journal of Leisure Research 24 (3), 207224.
link between these two variables in the TPB, which should be Ajzen, I., Fishbein, M., 1980. Understanding Attitude and Predicting Social Behavior.
Prentice-Hall, Englewood Cliffs, NJ.
considered when testing the theory in any context (e.g., Chang, Ajzen, I., Madden, T., 1986. Prediction of goal-directed behavior: attitude, inten-
1998; Han et al., 2010; Randall and Gibson, 1991; Ryu and Jang, tions, and perceived behavioral control. Journal of Experimental Social Psychol-
2006). The current study result is consistent with these research- ogy 22, 453474.
Anderson, E.W, Fornell, C., Lehmann, D.R., 1994. Customer satisfaction, market
ers claims. Thus, application of the theory in explaining customer share and protability. Journal of Marketing 58 (3), 5366.
decision-making in any context should occur cautiously by Asperin, 2007. Exploring Brand Personality Congruence: Measurement and Appli-
considering the relationship between attitudinal and normative cation in the Casual Dining Restaurant Industry. Unpublished doctoral disser-
tation, Kansas State University.
constructs of the theory. From a practical point-of-view, the
Assael, H., 1984. Consumer Behavior and Marketing Action. Kent, Boston.
ndings implied that hotel customers favorable perceptions of a Bagozzi, R.P., Yi, Y., 1988. On the evaluation of structural equation models. Journal of
green hotel led to their positive appraisal of the eco-friendly the Academy of Marketing Science 16, 7494.
decision. In other words, green hotel customers favorable/ Baloglu, S., McCleary, K.W., 1999. A model of destination image formation. Annals of
Tourism Research 26 (4), 868897.
unfavorable attitudes toward repurchasing a green hotel product Bansal, H.S., Taylor, S.F., 1999. The service provider switching model (SPSM): a
largely depend on the positive/negative way in which their salient model of consumer switching behavior in the service industry. Journal of
referents (i.e., family/relatives, friends, and colleagues/co-workers) Service Research 2 (2), 200218.
Barsky, J.D., 1992. Customer satisfaction in the hotel industry: meaning and
consider revisiting a green hotel. Thus, green hotel marketers measurement. Hospitality Research Journal 16 (1), 5173.
should actively nd ways to help such referents develop favorable Bitner, M.J., Hubbert, A.R., 1994. Encounter satisfaction versus overall satisfaction
perceptions of their hotel. Presenting the particular environmen- versus quality. In: Rust, R.T., Oliver, R.L. (Eds.), Service Quality: New Directions
in Theory and Practice. Sage, London, pp. 7294.
tally friendly attributes of their hotel (e.g., water- and energy- Bloemer, J., Ruyter, K., 1998. On the relationship between store image, store
efcient xtures/devices, durable service items, environmental satisfaction and store loyalty. European Journal of Marketing 32, 499513.
cleaning, fresh air exchange system, organic foods) to the public Chan, W., Wong, K., 2006. Estimation of weight of solid waste: newspapers in Hong
Kong hotels. Journal of Hospitality & Tourism Research 30 (2), 231245.
through various information sources may improve such referents
Chang, M.K., 1998. Predicting unethical behavior: a comparison of the theory of
favorable perceptions of a green hotel. reasoned action and theory of planned behavior. Journal of Business Ethics 17,
18251834.
Cheng, S., Lam, T., Hsu, C.H.C., 2006. Negative word-of-mouth communication
5.2. Limitations and avenues for future research
intention: an application of the theory of planned behavior. Journal of Hospi-
tality & Tourism Research 30 (1), 95116.
This studys several weaknesses should be pointed out. First, the Claver-Cortes, E., Molina-Azorin, J.F., Pereira-Moliner, J., 2007. The impact of
extended TPB model was tested in a green hotel context. Since the strategic behaviors on hotel performance. International Journal of Contempo-
rary Hospitality Management 19 (1), 620.
measurement items for the study constructs were particularly Cronin, J.J., Taylor, S.A., 1992. Measuring service quality: a reexamination and
designed to be adequate in a green hotel setting, generalizing the extension. Journal of Marketing 56, 5568.
668 H. Han, Y. Kim / International Journal of Hospitality Management 29 (2010) 659668
Dalton, G.J., Lockington, D.A., Baldock, T.E., 2008. A survey of tourist attitudes to Madden, T., Ellen, P.S., Ajzen, I., 1992. A comparison of the theory of planned
renewable energy supply in Australian hotel accommodation. Renewable En- behavior and the theory of reasoned action. Personality and Social Psychology
ergy 33, 21742185. Bulletin 18 (2), 39.
Diekmann, A., Preisendorfer, P., 2003. Green and Greenback: the behavioral effects Manaktola, K., Jauhari, V., 2007. Exploring consumer attitude and behavior towards
of environmental attitudes in low-cost and high-cost situations. Rationality and green practices in the lodging industry in India. International Journal of Con-
Society 15 (4), 441472. temporary Hospitality Management 19 (5), 364377.
DSouza, C., Taghian, M., 2005. Green advertising effects on attitude and choice Morgan, R.M., Hunt, S.D., 1994. The commitment-trust theory of relationship
of advertising themes. Asian Pacic Journal of Marketing and Logistics 17, marketing. Journal of Marketing 58, 2038.
5166. Oh, H., Hsu, C.H.C., 2001. Volitional degrees of gambling behaviors. Annals of
Department of Environmental Protection, 2001. Greener Accommodation: Saving Tourism Research 28 (3), 618637.
Money and the Environment. Retrieved from http://www.dep.state.pa.us/ Oliver, R.L., 1997. Satisfaction: A Behavioral Perspective on the Consumer. McGraw-
greenerAccom.pdf June 10, 2009. Hill, New York.
Environmentally Friendly Hotels, 2008, http://www.environmentallyfriendlyhotels. Oliver, R.L., Bearden, W.O., 1985. Crossover effects in the theory of reasoned action:
com on June 10, 2009. a moderating inuence attempt. Journal of Consumer Research 12 (3), 324340.
Fishbein, M.A., Ajzen, I., 1975. Belief, Attitude, Intention and Behavior: An Intro- Parasuraman, A., Zeithaml, V.A., Berry, L.L., 1988. SERVQUAL: a multiple-item scale
duction to Theory and Research. Addison-Wesley, Reading, MA. for measuring consumer perceptions of service quality. Journal of Retailing 64
Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with unob- (1), 1240.
servable variables and measurement error. Journal of Marketing Research 18, Park, J., 2003. Understanding Consumer Intention to Shop Online. Unpublished
3950. doctoral dissertation. University of Missouri, Columbia, MO.
Green Hotel Association, 2009. What are Green Hotels? Retrieved from http:// Perugini, M., Bagozzi, R.P., 2001. The role of desires and anticipated emotions in
www.greenhotels.com on June 10, 2009. goal-directed behaviors: broadening and deepening the theory of planned
Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1998. Multivariate Data Analysis, behavior. British Journal of Social Psychology 40, 7998.
Eighth ed. Prentice Hall, Upper Saddle River, NJ. Pizam, A., 2009. Green hotels: a fad, ploy or fact of life? International Journal of
Han, H., Back, K., 2008. Relationships among image congruence, consumption Hospitality Management 28, 1 editorial.
emotions, and customer loyalty in the lodging industry. Journal of Hospitality Prendergast, G., Ho, W., 2002. The inuence of store image on store loyalty in Hong
& Tourism Research 32 (4), 467490. Kongs quick service restaurant industry. Journal of Foodservice Business
Han, H., Ryu, K., 2006. Moderating role of personal characteristics in forming Research 5 (1), 4559.
restaurant customers behavioral intentions: an upscale restaurant setting. Quellette, J.A., Wood, W., 1998. Habit and intention in everyday life: the multiple
Journal of Hospitality & Leisure Marketing 15 (4), 2553. processes by which past behavior predicts future behavior. Psychology Bulletin
Han, H., Hsu, L., Lee, J., 2009. Empirical investigation of the roles of attitudes toward 124, 5474.
green behaviors, overall image, gender, and age in hotel customers eco-friendly Randall, D.M., Gibson, A.M., 1991. Taking stock: can the theory of reasoned action
decision-making process. International Journal of Hospitality Management 28, explain unethical conduct? Journal of Business Ethnics 8 (10), 873882.
519528. Roberts, J.A., 1996. Green consumers in the 1990s: prole and implications for
Han, H., Hsu, L., Sheu, C., 2010. Application of the theory of planned behavior to advertising. Journal of Business Research 36, 217231.
green hotel choice: testing the effect of environmental friendly activities. Ryu, K., Jang, S., 2006. Intention to experience local cuisine in a travel destination:
Tourism Management 31, 325334. the modied theory of reasoned action. Journal of Hospitality & Tourism
Keaveney, S.M., 1995. Customer switching behavior in service industries: an Research 30 (4), 507516.
exploratory study. Journal of Marketing 59, 7182. Ryu, K., Han, H., Kim, T., 2007. The relationships among overall quick-casual
Kim, S., 2001. E-mail survey response rates: a review. Journal of Computer Mediated restaurant image, perceived value, customer satisfaction, and behavioral inten-
Communication 6 (2), 120. tions. International Journal of Hospitality Management 27 (3), 459469.
Kim, W., Ok, C., 2009. The effects of relational benets of favorable inequity, Sheppard, B.H., Hartwick, J., Warshaw, P.R., 1988. The theory of reasoned action: a
affective commitment, and repurchase intention in full-service restaurants. meta-analysis of past research with recommendations for modications and
Journal of Hospitality & Tourism Research 33 (2), 227244. future research. Journal of Consumer Research 15, 325343.
Kim, W., Ok, C., Canter, D.D., 2009. Contingency variables for customer share of visits Taylor, S.A., Baker, T.L., 1994. An assessment of the relationship between service
to full-service restaurant. International Journal of Hospitality Management 29, quality and customer satisfaction in the formation of consumers purchase
136147. intentions. Journal of Retailing 70 (2), 163178.
Kirk, D., 1995. Environmental management in hotels. International Journal of Taylor, S., Todd, P., 1995. Assessing IT usage: the role of prior experience. MIS
Contemporary Hospitality Management 7 (6), 38. Quarterly 19, 561568.
Kotler, P., 1991. Marketing Management: Analysis, Planning, Implementation and Ting, D.H., 2004. Service quality and satisfaction perceptions: curvilinear and
Control. Prentice-Hall, Englewood Cliffs, NJ. interaction effect. International Journal of Bank Marketing 22 (6), 407420.
Lam, T., Hsu, C.H.C., 2004. Theory of planned behavior: potential travelers from Vallerand, R.J., Deshaies, P., Cuerrier, J., Pelletier, L.G., Mongeau, C., 1992. Ajzen and
China. Journal of Hospitality & Tourism Research 28 (4), 463482. Fishbeins theory of reasoned action as applied to moral behavior: a conrma-
Lam, T., Hsu, C.H.C., 2006. Predicting behavioral intention of choosing a travel tory analysis. Journal Personality and Social Psychology 62, 98109.
destination. Tourism Management 27, 589599. Vandermerwe, S., Oliff, M.D., 1990. Customers drive corporations green. Long Range
Lee, M.J., Back, K., 2007. Association members meeting participation behaviors: Planning 23, 1016.
development of meeting participation model. Journal of Travel & Tourism Wolfe, K.L., Shanklin, C.W., 2001. Environmental practices and management con-
Marketing 22 (2), 1533. cerns of conference center administrations. Journal of Hospitality & Tourism
Lee, M.J., Back, K., 2009. Association meeting participation: a test of competing Research 25 (2), 209216.
models. Journal of Travel Research 46, 300310. Yang, Z., Peterson, R.T., 2004. Customer perceived value, satisfaction, and loyalty:
Lewis, R.C., Chambers, R.E., 2000. Marketing Leadership in Hospitality. Wiley, New the role of switching costs. Psychology & Marketing 21 (10), 799822.
York. Yesawich, P., 1997. The myth of the loyal quest. Lodging Hospitality 53 (11), 18.