Content uploaded by Xiongzhi Wang
Author content
All content in this area was uploaded by Xiongzhi Wang on Feb 07, 2021
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rsus20
Journal of Sustainable Tourism
ISSN: 0966-9582 (Print) 1747-7646 (Online) Journal homepage: https://www.tandfonline.com/loi/rsus20
A comparative study of relative roles and
sequences of cognitive and affective attitudes on
tourists’ pro-environmental behavioral intention
Xiongzhi Wang, Xiangru Qin & Yongbo Zhou
To cite this article: Xiongzhi Wang, Xiangru Qin & Yongbo Zhou (2020) A comparative
study of relative roles and sequences of cognitive and affective attitudes on tourists’ pro-
environmental behavioral intention, Journal of Sustainable Tourism, 28:5, 727-746, DOI:
10.1080/09669582.2019.1704297
To link to this article: https://doi.org/10.1080/09669582.2019.1704297
Published online: 24 Dec 2019.
Submit your article to this journal
Article views: 500
View related articles
View Crossmark data
Citing articles: 3 View citing articles
A comparative study of relative roles and sequences
of cognitive and affective attitudes on tourists’
pro-environmental behavioral intention
Xiongzhi Wang
a,b
, Xiangru Qin
b
and Yongbo Zhou
c
a
School of Communication and Arts, University of Queensland, Brisbane, Australia;
b
School of Tourism
Management, Sun Yat-sen University, Zhuhai, China;
c
School of Social Science, Soochow University,
Suzhou, China
ABSTRACT
Attitudinal interventions are commonly implemented to nudge pro-
environmental actions. However, the relative roles and sequences of
cognitive and affective attitudes on tourists’pro-environmental behav-
ioral intention (PEBI) and how PEBI variants impact the attitude-inten-
tion relationships require further investigation. Based on the theory of
planned behavior, this paper compares the differential effects and
sequences of cognitive and affective attitudes on tourists’low-effort and
high-effort PEBI. Structural equation modeling results of two studies
conducted in a nature-based and an urban destination showed a super-
ior role of affect, i.e., affective, but not cognitive attitude, guided tou-
rists’PEBI, irrespective of implementation difficulty variants. This finding
supports the automatic-processing rather than the controlled-processing
hypothesis of attitude-intention processes in the tourism context. The
results of sequential models further confirmed the above conclusion
and supported the “affective”model wherein affective attitude predicts
cognitive attitude when shaping subsequent PEBI. These studies
advance understandings of relationships between and sequences of atti-
tudes and PEBIs and challenge predominant cognitive interventions in
environmental communication practices.
ARTICLE HISTORY
Received 18 June 2019
Accepted 5 December 2019
KEYWORDS
Cognitive attitude; affective
attitude; theory of planned
behavior; pro-environmental
behavioral intention;
low-effort and high-effort
Introduction
Although individuals’pro-environmental behavioral intentions (PEBI) or actions play a central role
in fostering a destination’s sustainability (Lee, Jan, & Yang, 2013), Dolnicar and Leisch (2008)
pointed out the reluctance of tourists to protect the environment while vacationing. How to effect-
ively engage tourists in environmental protection has become a critical issue, and attitudinal inter-
ventions have been proposed as a potentially useful way of doing so (Alcock et al., 2017).
The theory of planned behavior (TPB) is an influential and widely used sociopsychological
model for examining the attitude-behavior/intention relationship (White, Smith, Terry,
Greenslade, & McKimmie, 2009). It assumes that an individual’s decision-making follows rational
rules and originally conceptualizes attitude in a cognitive sense (Ajzen, 1991), yet it has been
subsequently expanded to include both cognitive and affective components (Ajzen & Driver,
1991,1992). However, most recent conceptualizations of attitude focus predominately on the
cognitive aspect (Alcock et al., 2017; Huffman, Van Der Werff, Henning, & Watrous-Rodriguez,
CONTACT Yongbo Zhou zhouyongbo@live.cn School of Social Science, Soochow University, Suzhou, China
ß2019 Informa UK Limited, trading as Taylor & Francis Group
JOURNAL OF SUSTAINABLE TOURISM
2020, VOL. 28, NO. 5, 727–746
https://doi.org/10.1080/09669582.2019.1704297
2014), failing to tap the established cognitive-affective structure of attitude (Crites, Fabrigar, &
Petty, 1994; Eagly & Chaiken, 1993).
Environmental conservation efforts are not solely utilitarian but have an essential yet inad-
equately explored affective basis (Carrus, Passafaro, & Bonnes, 2008; Rhodes et al., 2015). Cognitive
and affective attitudes may work together to facilitate PEBI (Kazeminia, Hultman, & Mostaghel,
2016), yet their relative roles in inducing tourists’PEBI are still unclear (Carrus et al., 2008;Koenig-
Lewis, Palmer, Dermody, & Urbye, 2014; Malhotra, 2005; Taufik, Bolderdijk, & Steg, 2016). Some
attempts have been made in this direction using, for example, different proxies to compare the dif-
ferentiated predictive utility of cognitive and affective attitudes (e.g., Cottrell, 2003; Teng, Wu, &
Liu, 2015;Wangetal.,2016). However, the findings of these comparative attempts are inconsistent
(e.g., Kraft, Rise, Sutton, & Røysamb, 2005; Rhodes et al., 2015;Zhang,Moyle,&Jin,2018).
Two main reasons underline these inconclusive comparisons. First, proxies for cognitive and
affective attitudes lack semantic equivalence since they are defined at different levels of specifi-
city (van der Linden, 2014). Second, some measurements of these proxies demonstrate low con-
tent validity, as they are not aligned with related constructs (e.g., Ru, Wang, Chen, & Yan, 2018;
Hinds & Sparks, 2008; Wan, Shen, & Choi, 2017). Few comparative investigations—free of seman-
tic and measurement problems—examine the relative influences of cognitive and affective atti-
tudes in shaping PEBI (Keer, van den Putte, & Neijens, 2010; Malhotra, 2005). In addition to their
inconsistent parallel roles, understandings also diverge concerning the sequential impacts of cog-
nitive and affective attitudes on PEBI, as two competing schools of thought support the
“affective primacy”or “cognitive primacy”, respectively (e.g., Zajonc, 1984; Lazarus, 1981,1984).
PEBI variants matter in the attitude-intention relationship (Bagozzi, Yi, & Baumgartner, 1990).
There are different types of pro-environmental behavior (PEB) (Stern, 2000) that vary in terms of
implementation difficulty (Thøgersen, 2004). They could be categorized as high-effort and low-
effort, with the former referring to demanding behaviors that require investments of more phys-
ical actions or resources than the latter, which are relatively easy to implement (Coelho, Pereira,
Cruz, Sim~
oes, & Barata, 2017). As the attitude-intention link is contingent on the types of
behavior (Ajzen, 1991), the relative roles of cognitive and affective attitudes can be deemed
behavior-specific (Andersson & von Borgstede, 2010; Verplanken, Hofstee, & Janssen, 1998). That
is, cognitive or affective attitude may lead to low-effort PEBI but not necessarily to high-effort
PEBI (Gosling & Williams, 2010), and vice versa. However, few studies have compared the compet-
ing roles of attitudes on low-effort and high-effort PEBI. How different attitudinal components
influence PEBI variants of implementation difficulty thus remains unresolved.
In summary, three gaps exist. First, ineffective comparisons of the relative roles of cognitive and
affective attitudes in PEBI are diagnosed. Second, the sequences of cognitive and affective attitudes
on PEBI are mixed. And third, the role of PEBI variants in the attitude-intention relationship has
been under-scrutinized. This paper thus aims to test and compare the relative predictive power and
sequences of cognitive and affective attitudes on tourists’low-effort and high-effort PEBI. It asks: (1)
which is more effective in eliciting PEBI? (2) does cognition predict affect or affect drive cognition
when guiding subsequent PEBI? (3) would PEBI variants lead to variations in the model results? It
contributes theoretically to nuanced understandings of the components of attitude, PEBI, and intri-
cate relationships among them, thus providing a more accurate picture of the attitude-intention
relationships and attitudinal sequences of PEBI. It also offers practical implications to strengthen
pro-environmental communication practices with better informed intervening strategies.
Literature review and hypotheses development
Theory of planned behavior and dimensions of attitude
Numerous psychological theories have been developed to explain PEBs. Among these, the
theory of reasoned action (TRA), and its extended version, the theory of planned behavior (TPB), the
728 X. WANG ET AL.
value-belief-norm (VBN) theory, and its precursor, the norm activation model (NAM) are the most
commonly used theories (Kl€
ockner, 2013). These four theories represent two major schools in under-
standing tourists’pro-environmental decision-making, namely, the rational and moral approaches (Li
&Wu,2019). The former two model PEB as a utilitarian cost-benefit assessment of consequences,
whereas the latter two underline the activation of morality as its uppermost antecedent.
The TPB postulates that an individual’s behavior is determined by intention, which in turn is
predicted by three independent variables: attitude, subjective norm, and perceived behavioral
control (Ajzen, 1991). It is selected as this paper’s theoretical foundation for three reasons. First,
the TPB receives a wider application and is concluded as a better model in explaining PEBs than
the NAM or VBN (Chan & Bishop, 2013). Second, this paper focuses on the relative roles of atti-
tudes in eliciting tourists’PEBIs. In this respect, the TPB is an influential model in examining atti-
tude-intention links (White et al., 2009), while the NAM or VBN is irrelevant. And third, most
applications of the TPB are too rational, failing to tap the established cognitive-affective structure
of attitude and to compare the competing roles of cognitive and affective attitudes in fostering
tourists’PEBIs, which then serve as starting points of the present inquiry.
Within social psychology, one of the particularly established and widespread conceptualiza-
tions of attitude is the tripartite model of attitudes, which posits that an attitude consists of
three distinct components: cognitive, affective, and conative attitude (Eagly & Chaiken, 1993).
Cognitive approaches dominate individuals’understanding of attitude formation and concern a
rational evaluation of behavior-related personal outcomes (Koenig-Lewis et al., 2014); affect also
constitutes an underpinning and involves affective feelings about attitude objects (Fielding,
Hornsey, & Swim, 2014); conative attitude usually refers to behavioral intentions or actual behav-
iors (Farley & Stasson, 2003). The TPB is an influential sociopsychological model for examining
the relationship between attitude and PEBI (Chan & Bishop, 2013). It originally conceptualizes
attitude as a global tendency to evaluate a behavior with a certain degree of favor or disfavor,
in a cognitive/utilitarian/instrumental sense (Ajzen, 1991). However, scholars have come to realize
the significance of considering distinct features of attitude rather than its global evaluative
nature (Crites et al., 1994).
Accordingly, Ajzen, who proposed the TPB, has expanded the attitude construct to include both
cognitive and affective attitudes (Ajzen & Driver, 1991,1992), arguing that evaluating attitude in a
single factor may fail to provide an accurate representation as negative cognitive attitude and
positive affective attitude may offset each other, and vice versa (Ajzen & Driver, 1992). Besides, due
to the distinct features of cognitive and affective attitudes, they should be measured in separate
constructs (Wan et al., 2017). In short, attitude can be subdivided into cognitive and affective atti-
tudes, and they have been methodologically confirmed (e.g., Crites et al., 1994; Breckler & Wiggins,
1989; Verplanken et al., 1998) and empirically supported (e.g., Jun & Arendt, 2016; Keer, Putte, &
Neijens, 2010;Rhodesetal.,2015;Ruetal.,2018) as distinct components of attitude.
PEBI variants of implementation difficulty
PEB refers broadly to any behavior during traveling in destinations that not only causes little
damage to the natural environment but may even benefit it (Steg & Vlek, 2009). As a complex
construct, PEB evidences typological variety (Stern, 2000), with differences in the implementation
difficulty being a crucial dimension (Lee et al., 2013). PEBs differ in terms of the required invest-
ment of resources (time, money, and energy) or physical actions (Coelho et al., 2017; Thøgersen,
2004) and could be termed as low-effort and high-effort. The literature focuses predominantly
on less demanding behaviors (Gosling & Williams, 2010). For this, Halpenny (2010) identified two
place-specific PEBI factors, with the second one being more difficult to perform. Ramkissoon,
Smith, and Weiler (2013a,2013b) further validated it necessary to capture PEBI as two factors:
high-effort and low-effort, as their model results varied across these two dimensions.
JOURNAL OF SUSTAINABLE TOURISM 729
The magnitude of the attitude-intention relationship varies across behaviors (Ajzen, 1991),
and PEBI variants of implementation difficulty play a significant role in it (Bagozzi et al., 1990).
Cognitive or affective attitude is thus behavior-specific and has differentiated roles wherein each
attitudinal component relates differently to and has different implications for PEBI variants
(Andersson & von Borgstede, 2010; Verplanken et al., 1998). Some initial studies have focused on
the relationship between attitudes and PEBI characterized by varying levels of implementation
difficulty. For example, the promotion of painless PEBs does not necessarily lead to the adoption
of other significant yet difficult ones (Thøgersen & Crompton, 2009). Gosling and Williams (2010)
found that emotional association with nature was related only to simpler conservation behaviors
like recycling. Chen et al. (2011) also found that environmental attitude exerted a significant
impact on simple behaviors but not on demanding ones. However, these studies ignore the cog-
nitive-affective structure of attitude and are thus unable to examine the competing roles of atti-
tudes on low-effort and high-effort PEBI, which provides a need for further inquiry.
Relationships between attitudes and PEBIs
Cognitive attitude refers to an individual’s general evaluation of behavioral outcomes, such as
the benefits, functions, or consequences of given behaviors. It implicitly postulates that an indi-
vidual’s intention to perform certain behaviors is a rational outcome of precise calculations of
costs and benefits (Ajzen, 1991). Therefore, pro-environmental decision-making results from a
utilitarian assessment of consequences, that is, whether the behavior in question yields more
benefits or advantages than it does costs (Li & Wu, 2019). Cognitive approaches predominate in
environmental communication practices aimed at offsetting information deficits or an awareness-
involvement gap through the provision of ecological knowledge (Koenig-Lewis et al., 2014)or
the conduct of awareness campaigns (Barr & Gilg, 2007).
Affective attitude is defined as the extent to which an individual develops affective feelings
regarding concerned behaviors (Jun & Arendt, 2016). Previous research on PEBs has confirmed
that environmental participation is not based solely on rational cost-benefit analysis; affective
feelings also play an essential role (Carrus et al., 2008; Rhodes et al., 2015), although these affect-
ive approaches have not received adequate exploration in pro-environmental studies (Koenig-
Lewis et al., 2014; Taufik et al., 2016). The cognitive and affective components of attitude could
both contribute to tourists’PEBI (Kazeminia et al., 2016), possibly in different ways.
The MODE model
1
proposed by Fazio (1990) maintains that there are two basic classes of
processes that link different components of attitude to various behaviors. One is a deliberative
processing mode that is planned, effortful or controlled. It is characterized by considerable cogni-
tive labor and involves rational analysis of positive or negative features of behaviors. The other is
a spontaneous processing mode that is automatic and more assessable (Farley & Stasson, 2003).
These two modes can also be referred to as the controlled-processing hypothesis and the auto-
matic-processing hypothesis, respectively (Fazio, 1990). The controlled-processing hypothesis is
compatible with the rational assumption of the TPB. It assumes that an individual consciously
analyzes the instrumental functions or consequences of a particular behavior and, in so doing,
deliberately reflects on the cognitive attitude before enacting that behavior. On the other hand,
the automatic-processing hypothesis assumes that an affective attitude guides individuals’
behavior automatically or spontaneously without effortful calculation and tradeoff. In short, cog-
nitive and affective attitudes both exert influence on PEBI variants of implementation difficulty
yet they do so through distinct processes (Fazio, 1990; Farley & Stasson, 2003). In addition, some
empirical studies have verified the significance of the cognitive attitude (Alcock et al., 2017; Han,
2015; Huffman et al., 2014) and affective attitude (Kazeminia et al., 2016; Coelho et al., 2017;Ru
et al., 2018; Wan et al., 2017) on various sustainable behaviors, such as recycling and green lodg-
ing. Thus, regarding PEBI in the tourism context, the following hypotheses are proposed:
730 X. WANG ET AL.
H1a: Cognitive attitude positively affects tourists’low-effort PEBI.
H1b: Cognitive attitude positively affects tourists’high-effort PEBI.
H2a: Affective attitude positively affects tourists’low-effort PEBI.
H2b: Affective attitude positively affects tourists’high-effort PEBI.
Relative roles of cognitive and affective attitudes in eliciting PEBIs
Affect plays a vital role in the relationship between human beings and the natural environment
(Hinds & Sparks, 2008). Some empirical studies have indicated that engendering positive affects
toward or emotional affinity with nature can predict PEBs (Perkins, 2010). These results echo
Wilson’s(1993) biophilia hypothesis, which posits that individuals are inherently attuned to
nature. Accordingly, compared with cognitive attitude, affective attitude is potentially a more
powerful predictor of PEBI.
The system of affective attitude is more primary and intrinsic than that of cognitive attitude
and is therefore considered as a default pattern of attitudinal influence (Farley & Stasson, 2003).
Also, affective attitude is associated with faster response times—as it is not subjected to verifi-
cation tests (Verplanken et al., 1998)—than cognitive attitude, which is slower and more con-
trolled (Zajonc, 1980). Consequently, affective attitude is more accessible than cognitive
attitude (Van den Berg, Manstead, van der Pligt, & Wigboldus, 2006). In sum, affective attitude
precedes a cognitive evaluation (Zajonc, 1984), thereby exerting a stronger effect on actions or
decisions (Farley & Stasson, 2003; Breckler & Wiggins, 1989). This view has been endorsed within
various empirical studies. For example, in studies by Jun and Arendt (2016) and Keer et al.
(2010), affective attitude was found to have stronger effects on intention and willingness than
cognitive attitude. Affective attitude is often identified as a stronger predictor of intention com-
pared with cognitive attitude (Kraft et al., 2005). Some studies on PEBs have suggested that
affective feelings are more influential than cognitive evaluations in cultivating PEBI (Kazeminia
et al., 2016; Koenig-Lewis et al., 2014;Wangetal.,2016). Accordingly, this paper proposes
further hypotheses:
H3a: Compared with cognitive attitude, affective attitude is more effective in eliciting tourists’
low-effort PEBI.
H3b: Compared with cognitive attitude, affective attitude is more effective in eliciting tourists’
high-effort PEBI.
Sequences of cognitive and affective attitudes in directing PEBI
Understandings diverge concerning the sequential impacts of cognitive and affective attitudes
on behavioral intentions. Scholars in the “affective primacy”school treat affective attitude as a
driver of cognitive attitude. They assume, based on the information processing theory
(Kobbeltved, Brun, Johnsen, & Eid, 2005), that it is more efficient to rely on the affect heuristic
than to cognitively analyze all available information, such that affective attitude is a precursor of
cognitive attitude (Slovic, Finucane, Peters, & Macgregor, 2004,2007; Zajonc, 1980,1984) when
shaping subsequent intentions (i.e., the “affective”model). Whereas for scholars in the “cognitive
primacy”school, they draw on the appraisal theory to argue that things are cognitively appraised
before engendering affective reactions (Keller et al., 2012). That is, affect as a post-cognitive pro-
cess, or cognitive evaluations give rise to affective response (Lazarus, 1981,1984) and then facili-
tate subsequent intentions (i.e., the “cognitive”model). These two schools of thought both have
a sound theoretical basis and substantial empirical support (van der Linden, 2014; Lai, Hagoort,
& Casasanto, 2012). Thus, concerning the sequences, this paper proposes two compet-
ing hypotheses:
JOURNAL OF SUSTAINABLE TOURISM 731
H4a: The “cognitive”model is superior in explaining tourists’PEBI.
H4b: The “affective”model is superior in explaining tourists’PEBI.
Method
Measures
Measurement scales were adopted from established studies and were modified to suit the pre-
sent context. Specifically, to avoid semantic and measurement problems and to enable valid
comparison of cognitive and affective attitudes, both constructs were measured using three
items adopted from Jun and Arendt (2016). Perceived behavioral control was assessed using
four items derived from Shi, Fan, and Zhao (2017), Han (2015), and Han, Jae, and Hwang
(2016). Subjective norm contained four items based on Shi et al. (2017) and White et al. (2009).
The two dependent variables, namely low-effort and high-effort PEBI, each included six items
adopted from Halpenny (2010) and Ramkissoon et al. (2013a). These adoptions were based on
two considerations. Firstly, the indicators of low-effort and high-effort PEBI are comprehensive
as they cover all the recognized six dimensions (civil, educational, financial, legal, physical, and
persuasive action) (Lee et al., 2013). Secondly, they help test whether and how the magnitude
of the attitude-intention relationship is contingent on PEBI variants of implementa-
tion difficulty.
All of the items on the questionnaire were back-translated into Chinese by three research-
ers fluent in Mandarin and English, and feedback was obtained from nine experts. Cognitive
and affective attitudes are usually measured using bipolar semantic scales (e.g., harmful-bene-
ficial or unpleasant-pleasant) (Jun & Arendt, 2016). However, semantic opposites do not neces-
sarily constitute a bipolar continuum; instead, they may associate with different causes and
effects (Herzberg, 1966). Bipolar scales could also prompt attitude ambivalence (Costarelli &
Colloca, 2004), leading to ambiguous results of attitudes. Accordingly, all items were posi-
tioned along a 7-point Likert-type unipolar scale ranging from 1 (strongly disagree) to 7
(strongly agree).
Data analysis
SPSS was employed to conduct item analysis, determine common method bias, and calculate
the reliability and normality of variables and their intercorrelations. Meanwhile, structural equa-
tion modeling (SEM) via AMOS was utilized to confirm the structure of attitude and PEBI, con-
duct construct validation, and test the hypotheses. SEM via maximum likelihood was conducted,
after confirming normality, to compare path coefficients in the proposed model. By doing so, the
differentiated roles of cognitive and affective attitudes in fostering tourists’PEBIs were tested.
Besides, a pair of SEMs was executed to compare fit indices of the “cognitive”and “affective”
models to examine their sequences. Considering the destination itself may influence tourists’
PEBI (Line & Hanks, 2016), this paper tested the model in both nature-based and urban destina-
tions to enable robust comparison.
Study 1
Study site and participants
Wulingyuan, known for its striking natural landscape and ecological resources, was designated as
China’s first national forest park and subsequently as a World Natural Heritage. This site attracts
over 3 million tourists annually and is thus a typical representative of nature-based tourism desti-
nations (Gao, Zhang, & Huang, 2018). However, it suffers serious environmental problems and
732 X. WANG ET AL.
has received two official warnings from UNESCO as a result of tourism overdevelopment and
tourists’irresponsible behavior.
An on-site survey was conducted by five trained research assistants, from July 22 to 28, 2017,
including both off-peak (weekdays) and peak times (weekends) to ensure a diverse range of par-
ticipants. Self-administrated questionnaires were conveniently distributed to domestic tourists
based on four criteria: (1) they should be adult to have a mature and independent judgment; (2)
they ought not to be residents to fit the definition that a tourist indicates; (3) they shall had
already visited the site for at least one hour to gain a basic understanding of the environment;
(4) they should be relaxed and not in a hurry to leave, so as to increase the chances and quality
of participation. Thus, research assistants approached those potential tourists who were resting
in major scenic spots rather than at the entrance or exit and proceeded if they were adults but
not residents, and had stayed for more than one hour. The research purpose, anonymity, volun-
tary participation, guidelines, and the contextualized definition of the PEB presented on the top
of the questionnaire, was again stressed by the assistants. Those willing tourists completed the
survey under the supervision of assistants, who stepped back about 2 meters to reduce social
desirability bias from researchers. The assistants would ask the participant to refill the question-
naire if many missing answers existed. They then marked the validity based on on-site observa-
tions of whether the questionnaire was carefully completed.
Out of 150 distributed questionnaires, 143 valid ones were retrieved. This sample size can be
regarded as adequate as it meets a minimum sample size of 50 (Iacobucci, 2010) or the criteria
proposed by Tinsley and Tinsley (1987), suggesting a ratio of about 5 to 10 subjects for each
indicator. Seventy-three men (51%) and seventy women (49%) took part. The participants were
mainly young and middle-aged tourists, with 91.7% aged between 18 and 45. Most participants
were well-educated, with 74.1% having obtained a college/university degree or above. The par-
ticipants were from all walks of life, and the monthly incomes of 42% of the participants ranged
between 3001 and 7000 RMB. The sample demographics were similar to a study by Gao et al.
(2018), in which most tourists visiting Wulingyuan were younger, well-educated, and had average
incomes.
Results
Preliminary analyses
Item analysis results showed that PEBI1 and PEBI2 performed poorly in terms of communalities
(<0.25), factor loading (<0.50), and corrected item-total correlation (<0.40), and were therefore
deleted (Churchill, 1979). The absolute value of skewness and kurtosis coefficient for each indica-
tor was less than 3 and the conservative threshold of 10, respectively (Table 1), demonstrating
that the data were normally distributed (Kline, 2010).
Table 1. Pearson’s correlations among variables (N ¼143).
Mean SD CATT AATT PBC SN Low-effort PEBI High-effort PEBI
CATT 6.88 0.35 0.85
AATT 6.48 0.79 0.44 0.79
PBC 5.92 1.05 0.08 0.32 0.69
SN 6.32 0.88 0.26 0.41 0.27 0.85
Low-effort PEBI 5.88 0.96 0.14 0.29 0.180.30 0.66
High-effort PEBI 4.64 1.49 0.03 0.28 0.27 0.28 0.60 0.81
Skewness 2.61 1.46 1.01 1.34 0.36 0.00
Kurtosis 5.77 1.23 0.97 1.23 0.77 0.90
Note: p<0.05, p<0.01, p<0.001. CATT: cognitive attitude; AATT: affective attitude; PBC: perceived behavioral con-
trol; SN: subjective norm; PEBI: pro-environmental behavioral intention; diagonally positioned data in bold denotes the
square roots of AVE.
JOURNAL OF SUSTAINABLE TOURISM 733
Common method bias may undermine the validity of the results, given the use of a mono-
methodology of self-reported questionnaires (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). The
results of Harman’s single-factor test indicated that all measurement items were extracted onto
six factors with eigenvalues above 1, which collectively explained 73.37% of the total variance.
The first factor was below the threshold value of 40%, explaining only 30.08% of the total vari-
ance. These results suggested that common method bias was unlikely a problem.
The structure of attitude and PEBI was examined by performing CFA. The results indicated an
acceptable model fit for the two-component structure of attitude (Lee et al., 2013): CMIN/df ¼
15.310/8 ¼1.914, IFI ¼0.983, TLI ¼0.968, CFI ¼0.983, RMSEA ¼0.080. CFA results also con-
firmed a good model fit for the low-effort and high-effort PEBI structure (Lee et al., 2013): CMIN/
df ¼41.401/29 ¼1.428, IFI ¼0.986, TLI ¼0.978, CFI ¼0.986, RMSEA ¼0.055.
Reliability and validity
Cronbach’s alpha for all variables was above the recognized level of 0.7, whereas the composite
reliability (CR) results indicated that they all exceeded the cut-off value of 0.6 (Lee et al., 2013),
indicating reliable measurement scales.
The convergent validity of constructs is evidenced when their AVE values exceed 0.5, and factor
loadings of their indicators are higher than the cut-off value of 0.4 (Chen, Bao, & Huang, 2014).
The factor loadings of all indicators ranged from 0.446 to 0.950, exceeding 0.4 (Table 2). Variables’
average variance extracted (AVE) were above the recommended threshold of 0.5 (Ru et al., 2018),
except for perceived behavioral control (0.48) and low-effort PEBI (0.44). Yet as pointed out by
Fornell and Larcker (1981), the adequacy of the given construct’s convergent validity could be
based on composite reliability (CR) alone since the AVE is a conservative measure. Therefore, even
if the AVE values for PBC and low-effort PEBI were below 0.5, they could still be regarded as con-
vergently valid when considering their CR values. The discriminant validity was also examined. The
square root of AVE for each variable was greater than its correlation with other variables (Table 1),
which demonstrated clear evidence of discriminant validity (Lee et al., 2013).
Hypothesis testing
As an acceptable measurement model was obtained, SEM followed to examine whether the the-
oretical relationships specified between constructs were supported by the data (Figure 1). The
goodness-of-fit statistics indicated that the proposed model fit the data satisfactorily: CMIN/df ¼
1.702, IFI ¼0.919, TLI ¼0.903, CFI ¼0.917, RMSEA ¼0.070. Standardized path coefficients of
the structural model suggested that affective attitude (b¼0.244, p<0.1) was the second stron-
gest predictor, while cognitive attitude exerted an insignificant influence on low-effort PEBI.
Meanwhile, affective attitude (b¼0.277, p<0.05) emerged as the strongest predictor of high-
effort PEBI, yet cognitive attitude (b¼0.238, p<0.05) had a significant negative impact. In sum,
affective attitude effectively predicted low-effort and high-effort PEBI, while cognitive attitude
was ineffective for low-effort PEBI and had a significant negative influence on high-effort PEBI.
As a result, H2a,H2b,H3a, and H3b were supported, while H1a and H1b were not supported,
contrary to our theoretical assumption.
Concerning the sequences of cognitive and affective attitudes, overall fit indices (Figure 2)
suggested that the data fit the “affective”model (CMIN/df ¼1.607, IFI ¼0.930, TLI ¼0.916, CFI
¼0.928, RMSEA ¼0.065) better than the “cognitive”model (CMIN/df ¼1.718, IFI ¼0.917, TLI ¼
0.901, CFI ¼0.915, RMSEA ¼0.071). As these models were non-nested, the parsimony fit indices
were also compared, with lower values indicating better fit. The results (DAIC¼26.136) were in
strong favor of the “affective”model (AIC ¼507.340), rather than the “cognitive”one (AIC ¼
533.476), thus supporting H4b, rather than H4a. Mediation analysis revealed that affective
734 X. WANG ET AL.
Table 2. Measurement model.
Study 1 (N ¼143) Study 2 (N ¼466)
Construct and items FL CR AVE aFL CR AVE a
Cognitive attitude (CATT) (Jun & Arendt, 2016) 0.89 0.72 0.88 0.87 0.69 0.87
CATT1: For me, environmental protection is good 0.848 0.866
CATT2: For me, environmental protection is beneficial 0.845 0.844
CATT3: For me, environmental protection is wise 0.853 0.771
Affective attitude (AATT) (Jun & Arendt, 2016) 0.83 0.63 0.79 0.88 0.71 0.82
AATT1: For me, environmental protection is pleasant 0.718 0.914
AATT2: For me, environmental protection is enjoyable 0.945 0.873
AATT3: For me, environmental protection is interesting 0.699 0.728
Perceived behavioral control (PBC) (Shi et al., 2017; Han, 2015; Han et al., 2016) 0.78 0.48 0.76 0.86 0.61 0.77
PBC1: It is easy for me to participate in environmental protection 0.672 0.775
PBC2: I am confident that if I want, I can protect the environment when traveling 0.859 0.846
PBC3: Whether or not I protect the environment while traveling is completely up to me 0.606 0.728
PBC4: I have the resources, time, and opportunities to protect the environment while traveling 0.600 0.762
Subjective norm (SN) (Shi et al., 2017; White et al., 2009) 0.92 0.73 0.92 0.94 0.80 0.92
For those who are important to you (like relatives, friends or colleagues),
SN1: they would say/think I should act pro-environmentally when traveling 0.772 0.893
SN2: they think that engaging in environmental protection while traveling is something that one
ought to do
0.849 0.902
SN3: they would prefer/approve of behaving in an environmentally friendly way while traveling 0.887 0.890
SN4: they expect me to participate in environmental protection when traveling 0.910 0.894
Low-effort pro-environmental behavioral intention (Low-effort PEBI) (Halpenny, 2010; Ramkissoon
et al., 2013a)
0.75 0.44 0.75 0.83 0.55 0.77
PEBI3: Sign petitions in support of Wulingyuan/Kulangsu and similar protected areas 0.446 0.605
PEBI4: Learn more about Wulingyuan’s/Kulangsu’s natural environment 0.634 0.717
PEBI5: Pick up litter at Wulingyuan/Kulangsu or other parks left by other visitors 0.783 0.924
PEBI6: Tell my friends not to feed the animals in Wulingyuan/Kulangsu or similar destinations 0.747 0.694
High-effort pro-environmental behavioral intention (High-effort PEBI) (Halpenny, 2010; Ramkissoon
et al., 2013a)
0.92 0.65 0.92 0.89 0.58 0.89
PEBI7: Pay increased ticket fees if they were introduced and used for environmental programs 0.566 0.527
PEBI8: Participate in a public meeting about managing Wulingyuan/Kulangsu or similar
destinations
0.867 0.791
PEBI9: Write letters in support of Wulingyuan/Kulangsu and similar destinations 0.875 0.816
PEBI10: Volunteer my time to projects that help Wulingyuan/Kulangsu or similar destinations 0.950 0.854
PEBI11: Encourage others to reduce waste and pick up their litter when they are at Wulingyuan/
Kulangsu or similar destinations
0.804 0.799
PEBI12: Contribute donations to ensure protection of destinations like Wulingyuan/Kulangsu 0.735 0.751
Note: FL ¼factor loading, CR ¼composite reliability, AVE ¼average variance extracted, a¼Cronbach’s alpha; PEBI1 and PEBI2 were deleted as they failed in the item analysis.
JOURNAL OF SUSTAINABLE TOURISM 735
attitude worked as a full and partial mediator on the effects of cognitive attitude on low-effort
and high-effort PEBI, respectively. However, cognitive attitude only mediated the link between
high-effort PEBI and affective attitude, yet in a negative way. These mediation results suggested
a superior role of affective attitude.
Discussion
Study 1 constituted the first step to test the hypotheses in a nature-based destination. The CFA
results confirmed the two-dimensionality of attitude and PEBI. SEM results revealed that affective
attitude consistently predicted tourists’PEBI variants of implementation difficulty, while cognitive
attitude was ineffective for low-effort PEBI, even having a reversed significant impact on high-
Figure 1. Results of the structural model (N ¼143).
Note:
þ
p<0.1, p<0.05, p<0.01, p<0.001.
Figure 2. Results of sequential models (N ¼143).
Note:
þ
p<0.1, p<0.05, p<0.01, p<0.001.
736 X. WANG ET AL.
effort PEBI. The “affective”model was supported wherein affective attitude was a precursor of
cognitive attitude.
The destination itself, however, plays a part in determining tourists’PEBI-related goals.
Scholars have argued that tourists who visit nature-based destinations demonstrate hyperaware-
ness of environmental protection through immersion in natural beauty (Line & Hanks, 2016).
Natural elements are less salient in urban destinations by comparison. Since it was unclear
whether findings for nature-based and urban destinations would differ, the first study was repli-
cated in an urban destination, discussed below.
Study 2
Study site and participants
Another round of data collection was conducted in Kulangsu, which suffers from environmental
problems similar to those at Wulingyuan. As a tiny island listed as a World Heritage Site,
Kulangsu faces the metropolitan city of Xiamen by the 600-meter-wide Lujiang Strait and simi-
larly attracts over three million tourists annually (Qiu, Zhang, & Zheng, 2018). UNESCO also has
expressed great concern for Kulangsu’s sustainable development and warned against the
threat of the tourism boom to its environment. A recent official report
2
reveals that many
residents have fled the island because of the dirty environment. Kulangsu has become an over-
commercialized and typical urban tourism destination (Hu, 2017).
An on-site survey was conducted by six trained research assistants, from August 11 to 20,
2018, on off-peak (weekdays) and peak times (weekends). Self-administrated questionnaires were
conveniently distributed to domestic tourists, based on the same judgment criteria and following
the same sampling process as in Study 1. Among the five hundred distributed questionnaires, a
total of 466 questionnaires was deemed usable, after excluding incomplete questionnaires (miss-
ing values per indicator exceeded 15%), questionnaires with similar response pattern, or those
that had been answered halfheartedly (marked by research assistants). Descriptive statistics sug-
gested no outliers and several missing values (less than 5% of values per indicator), which were
then replaced by their serial mean. The proportions of men and women respondents were
almost equal, and nearly 90% aged between 18 and 45. Most of them were well-educated,
75.8% had a college/university education or higher, and held various jobs. The incomes of 43.9%
of the respondents were at an average level. This profile is similar to that of the sample used by
Qiu et al. (2018) in Kulangsu.
Results
Preliminary analyses
The data were normally distributed since the absolute value of skewness and kurtosis for each
indicator was less than 3 and 10 (Kline, 2010), respectively (Table 3). Common method bias was
assessed using Harman’s single-factor test. The results indicated that all measurement items
were extracted onto six factors, which together accounted for 70.49% of the total variance. The
first factor explained 27.59% of the total variance, below the threshold value of 40%. Thus, com-
mon method bias did not arise.
The structure of attitude and PEBI was again examined. CFA results indicated a good model
fit for the two-component structure of attitude (Lee et al., 2013): CMIN/df ¼17.522/7 ¼2.503, IFI
¼0.993, TLI ¼0.985, CFI ¼0.993, RMSEA ¼0.057. A good model fit was also confirmed for the
structure of low-effort and high-effort PEBI (Lee et al., 2013): CMIN/df ¼61.016/21 ¼2.906, IFI ¼
0.983, TLI ¼0.964, CFI ¼0.983, RMSEA ¼0.064.
JOURNAL OF SUSTAINABLE TOURISM 737
Reliability and validity
As Table 2 shows, Cronbach’s alpha for all variables was above the acknowledged level of 0.7,
while the CR results indicated that they all exceeded the cut-off value of 0.6 (Lee et al., 2013),
suggesting good reliability. Their AVE values were all above 0.50, and their factor loadings were
all greater than 0.4, thus supporting convergent validity (Chen et al., 2014). The square root of
AVE for each variable exceeded its correlation with other variables (Table 3), thus supporting dis-
criminant validity (Lee et al., 2013).
Hypothesis testing
After obtaining a satisfactory measurement model, SEM was performed to test the proposed
model, with excellent model fit indices (Figure 3): CMIN/df ¼1.645, IFI ¼0.975, TLI ¼0.970, CFI
¼0.975, RMSEA ¼0.037. Inspections of standardized path coefficients revealed that affective
attitude (b¼0.167, p<0.05) was the second strongest predictor for low-effort PEBI while cogni-
tive attitude was insignificant. Affective attitude also exerted a significant impact on high-effort
PEBI (b¼0.140, p<0.05). Nonetheless, cognitive attitude was again a significant negative pre-
dictor for high-effort PEBI (b¼0.168, p<0.01), and its influence was much stronger than that
Table 3. Pearson’s correlations among variables (N ¼466).
Mean SD CATT AATT PBC SN Low-effort PEBI High-effort PEBI
CATT 6.85 0.40 0.83
AATT 6.49 0.75 0.52 0.84
PBC 5.95 0.99 0.16 0.25 0.78
SN 6.30 0.89 0.24 0.32 0.28 0.89
Low-effort PEBI 5.77 0.97 0.18 0.28 0.22 0.34 0.74
High-effort PEBI 4.51 1.33 0.01 0.14 0.23 0.23 0.55 0.76
Skewness 2.47 1.62 1.18 1.50 0.41 0.06
Kurtosis 5.87 2.14 2.13 2.89 0.70 0.29
Note: p<0.05, p<0.01, p<0.001. Diagonally positioned data in bold denotes the square roots of AVE.
Figure 3. Results of the structural model (N ¼466).
Note:
þ
p<0.1, p<0.05, p<0.01, p<0.001.
738 X. WANG ET AL.
of affective attitude. In sum, affective attitude positively and significantly predicted low-effort
and high-effort PEBI, whereas cognitive attitude had an insignificant impact on low-effort PEBI
and a significant negative influence on high-effort PEBI. Thus, H2a and H2b were reinforced while
H1a and H1b were not supported. Furthermore, H3a was explicitly supported while H3b was
implicitly supported, as cognitive attitude impeded tourists’high-effort PEBI despite having a sig-
nificantly greater path influence.
Overall fit indices (Figure 4) suggested that the data fit the “affective”model (CMIN/df ¼
2.346, IFI ¼0.948, TLI ¼0.938, CFI ¼0.948, RMSEA ¼0.054) better than the “cognitive”model
(CMIN/df ¼2.500, IFI ¼0.942, TLI ¼0.931, CFI ¼0.942, RMSEA ¼0.057). The parsimony fit indi-
ces (DAIC¼35.892) were also in strong favor of the “affective”model (AIC ¼681.002), rather
than the “cognitive”one (AIC ¼716.894), thus supporting H4b, rather than H4a. Bootstrap results
(done with 10,000 resamples) further suggested that affective attitude fully and partially medi-
ated the impact of cognitive attitude on low-effort PEBI (z¼0.317, 95% Boot. CI ¼[0.190, 0.468])
and high-effort PEBI (z¼0.339, 95% Boot. CI ¼[0.165, 0.554]), respectively. Whereas, cognitive
attitude did not mediate the impact of affective attitude on low-effort PEBI (z¼0.036, 95% Boot.
CI ¼[0.022, 0.098]) but negatively on high-effort PEBI (z¼0.107, 95% Boot. CI ¼[0.200,
0.026]). All these results again confirmed a superior role of affective attitude.
Discussion
Study 2, conducted in an urban destination, produced the same results as Study 1. These results
also confirmed a two-component structure of attitude (cognitive and affective) and PEBI (low-
effort and high-effort). They also demonstrated that affective attitude had consistent predictive
utility for both low-effort and high-effort PEBI while the role of cognitive attitude in tourists’PEBI
was a negative one. Furthermore, the “affective”model was again supported.
General discussion
The relative roles of cognitive and affective attitudes in eliciting tourists’PEBI are obscure (Carrus
et al., 2008; Malhotra, 2005; Taufik et al., 2016). Despite some preliminary attempts, the lack of
semantic equivalence and low content validity (Costarelli & Colloca, 2004) undermine the consist-
ency of their findings (e.g., Teng et al., 2015; Koenig-Lewis et al., 2014; Zhang et al., 2018).
Besides, the sequences of cognitive and affective attitudes on PEBI are mixed, advocating the
“affective primacy”or “cognitive primacy”(e.g., Zajonc, 1984; Lazarus, 1981,1984), respectively.
Figure 4. Results of sequential models (N ¼466).
Note:
þ
p<0.1, p<0.05, p<0.01, p<0.001.
JOURNAL OF SUSTAINABLE TOURISM 739
Furthermore, the role of PEBI variants of implementation difficulty in the attitude-intention rela-
tionship is crucial (Bagozzi et al., 1990; Ajzen, 1991) yet under-scrutinized. Therefore, this paper
makes a comparative study on the differing roles and sequences of cognitive and affective atti-
tudes on tourists’PEBI and further examines whether the extensions of PEBI would influence the
model results.
The results of both studies suggest a superior role of affective attitude in driving tourists’
PEBI. Specifically, affective attitude is consistently the second strongest predictor of low-effort
PEBI, and its impact on high-effort PEBI is robustly significant. By contrast, cognitive attitude is
an invalid predictor for low-effort PEBI, and it has a significant unintended negative impact on
high-effort PEBI, even demonstrating a stronger opposite path influence than affective attitude.
The results also confirm a superior role of the “affective”model wherein affective attitude pre-
dicts cognitive attitude when shaping tourists’subsequent low-effort and high-effort PEBI.
Although destination itself may influence tourists’PEBI (Line & Hanks, 2016), the findings of the
two studies are the same, validating the robustness of the above conclusions.
Compared with studies that assume an equal influence of cognitive and affective attitudes on
sustainable behaviors/intentions (Kazeminia et al., 2016; Rhodes et al., 2015; Ru et al., 2018), our
results reveal a different pattern: affective, but not cognitive attitude, has a primary impact on
tourists’PEBI irrespective of variants. These results confirm the findings of Kraft et al. (2005) and
Koenig-Lewis et al. (2014), in which affect, rather than cognition, was identified as a substantial
predictor of intentions. They also accord with studies that emphasize the role of affective feel-
ings. For example, affective connections significantly predict intentions to engage with nature
(Hinds & Sparks, 2008). Taufik et al. (2016) argued that feelings outweighed calculations in facili-
tating pro-environmental intentions. Wang et al. (2016) prioritized affective appeals, arguing that
pro-environmental consumption intention rested primarily on environmental affect rather than
cognition. Our findings also accord with those of Huffman et al. (2014) that cognitive attitude
was insignificant for self-reported and observed recycling behaviors. The results, however, contra-
dict those of particular studies. For example, Wan et al. (2017) found that the impact of experien-
tial attitude on recycling intention was insignificant. Additionally, Chen and Tung (2010) and
Cottrell (2003) found that individuals’cognitive but not affective attitude predicted their recy-
cling intentions. Yet, these contradictions are untenable considering the low content validity (i.e.,
a misalignment between the construct and its intended measures) of their scales of attitudes.
The results of sequential models support the “affective primacy”(Zajonc, 1980,1984) rather
than the “cognitive primacy”(Lazarus, 1981,1984) in the tourism context where the affect heuris-
tic (Slovic et al., 2004,2007) proceeds cognition when shaping tourists’subsequent PEBIs. These
results contradict Kobbeltved et al.’s(2005) argument that individuals unlikely allow their affect-
ive impressions to guide risk judgments, and indicate a primary, intrinsic, and spontaneous/auto-
matic pattern of affective attitude (Farley & Stasson, 2003; Van den Berg et al., 2006).
In short, the results of both studies consistently show a superiority of the “affective”model
and affective attitude in inducing tourists’low-effort and high-effort PEBI, in both nature-based
and urban destinations. The reasons may include the lower accessibility and relatively indirect
effect of cognitive attitude, coupled with the hedonic and transient features of the tourism con-
text and the altruistic element of PEB. First, cognitive attitude is less accessible than affective atti-
tude (Van den Berg et al., 2006), which is more automatic, intrinsic, and primary (Zajonc, 1984)
and requires no validation (Verplanken et al., 1998). As cognition follows the affect heuristic
(Slovic et al., 2007), its impact on intention is weaker (Farley & Stasson, 2003). Second, cognitive
attitude has a relatively indirect effect, as its influence on PEBI would be mediated by other vari-
ables. For example, Keer et al. (2010) found that affective feelings directly influence intention,
whereas the influence of cognitive evaluation is partially mediated by overall evaluation. Third,
the role of tourism contexts matters. Tourists tend to seek relaxation, enjoyment, and fun while
vacationing in destinations (Dolnicar & Leisch, 2008), such that their behaviors are often less pro-
environment in tourism destinations than in home environments (Alcock et al., 2017; Miao &
740 X. WANG ET AL.
Wei, 2013). Tourists would rationally assess the perceived costs and benefits or weigh potential
inconveniences incurred during hard-won hedonic vacations given the altruistic features of PEBs.
Consequently, tourists loosely associate the benefits of environmental actions or the consequen-
ces of inactions with their immediate personal interests (Li & Wu, 2019), or put otherwise, it is
cognitively none of their business. Besides, cognitive attitude fails or even backfires as environ-
mental appeals via rational cost-benefit analysis or infusing ecological knowledge adds cognitive
load and time burden to tourists (Taufik et al., 2016; Steg, 2016) when they are transitorily in an
unfamiliar destination for hedonic pleasure.
Conclusion
Theoretical contributions
This paper advances a broader picture of relationships between attitudes and PEBI variants
and nuanced understandings of attitude-intention processes and sequences related to pro-
environmental efforts, making four critical contributions to the literature.
Firstly, the results of two studies in this paper robustly confirm a superior role of affective atti-
tude in motivating tourists’PEBI irrespective of implementation difficulty. By contrast, cognitive
attitude has a superficial or even negative effect on PEBI in the tourism context, as it has lower
accessibility (Zajonc, 1980) and a relatively indirect effect (Keer et al., 2010). This finding validates
the notion that cognitive and affective attitudes are behavior-specific (Andersson & von
Borgstede, 2010; Verplanken et al., 1998) in guiding PEB decision-making in tourism contexts.
Secondly, this paper tests the debate of the “affective primacy”(Zajonc, 1980,1984) vs. the
“cognitive primacy”(Lazarus, 1981,1984) and suggests the superiority of the “affective”model
wherein affective attitude predicts cognitive attitude when shaping tourists’subsequent PEBI.
Therefore, it advances the understanding of attitude-intention sequences related to pro-
environmental efforts in tourism contexts.
Thirdly, this paper furthers the understandings of the attitude-intention processes of PEBs.
The controlled-processing hypothesis is not supported due to the unique nature of the tourism
context. Meanwhile, the automatic-processing hypothesis (Fazio, 1990) is confirmed as affective
attitude effectively guides tourists’PEBI regardless of implementation difficulty.
Lastly, this paper supports a two-component structure of PEBI (Ramkissoon et al., 2013a,
2013b) and attitude (Ajzen & Driver, 1991,1992), and makes an initial attempt to test the role of
PEBI variants of implementation difficulty on the processes and sequences of the attitude-
intention relationships beyond a direct and straightforward link posited in the TPB.
Practical implications
Several practical implications can be derived from the conclusions of a constructive (destructive)
role of affective (cognitive) attitude in nudging tourists’low-effort and high-effort PEBI and a
superior role of the “affective”model in explaining tourists’PEBI.
Cognition-based attitudinal interventions are quite common in tourism destinations. For
example, both Wulingyuan and Kulangsu persuade tourists to take away their waste or volunteer
their time or money to environmental actions via mathematical calculations of the benefits for
nature. These interventions assume knowledge or awareness deficits underlying environmental
inactions, and the deficits could be solved by infusing environmental facts or improving aware-
ness (Koenig-Lewis et al., 2014; Barr & Gilg, 2007). However, this paper, in line with Schultz
(2014), implies that persuading tourists, in the name of being good or beneficial for the environ-
ment, to act pro-environmentally can be invalid as these interventions may direct tourists to an
instrumental evaluation of “costs”against their hedonic goals and add cognitive load while on
vacation. Besides, when they are transitorily in an unfamiliar destination, the benefits
JOURNAL OF SUSTAINABLE TOURISM 741
(consequences) of environmental actions (inactions) may be cognitively none of their business.
Consequently, attitudinal interventions that involve or may explicitly raise the effort of rational
cost-benefit analysis of environmental protection should be reduced or used with caution, espe-
cially when appealing demanding PEBs in the tourism context.
Quite the contrary, attitudinal interventions that incorporate experiential feelings should be
prioritized in environmental campaigns. Destinations can thus use affective appeals, such as com-
municating vivid images of protecting nature to strengthen tourists’affective feelings or crafting
messages to trigger tourists’views that environmental protection is pleasant/enjoyable/interest-
ing. These strategies are useful as they work through an automatic-processing mode without
adding tourists’cognitive load and are compatible with their hedonic goals while on vacation.
Specifically, destinations can integrated these strategies into tourist interpretation systems by
highlighting the positive (affective) consequences of engaging in PEBs vividly through visual
images (like the Earth smiles with green coat, nature flourishes with plants and trees, and ani-
mals live happily in their habitat) to communicate the experiential features (e.g., pleasant and
enjoyable) of engaging in PEBs. They can also organize attractive activities to stimulate tourists’
positive affective feelings regarding environmental protection.
Limitations and future research
However, the results of this paper should be interpreted with caution as it has four limitations.
First, this paper is unable to tackle the differences in attitude-PEBI relationships between resi-
dents and non-local tourists, as it systematically excludes residents in the sampling process to fit
the conventional definition of “tourists”. However, individuals’attitudes and behaviors are often
more pro-environment in their home environments than in tourism destinations (Alcock et al.,
2017; Miao & Wei, 2013). Contrary to tourists, cognitive attitude may be potent for residents in
home environments, as the benefits or costs of environmental actions or the consequences of
inactions are closely related to residents’personal interests (like good everyday environment or
social reputation) (Li & Wu, 2019). Consequently, this paper’s conclusions are limited to tourists
only and could not be generalized to residents. Future research needs to further examine the dif-
ferences in attitude-PEBI relationships between residents and tourists.
Second, the feature of traveling is not directly reflected in the scales of attitudes. However, it
is underlined in the contextualized definition of the PEB presented at the top of the question-
naire. Besides, the research assistants also stress this definition before participants fill out the sur-
vey to ensure an understanding that all indicators related to PEBs (or colloquially environmental
protection) are rated in the tourism setting. Yet, admittedly, tourists may misunderstand or neg-
lect the fact that the attitudinal perceptions refer to attitudes in the tourism context but not in
their home environment. Consequently, this paper may present a biased picture of attitudinal
influences on tourists’PEBI. Future research should then make salient the traveling element and
clearly distinguish attitudes in home environments from that in tourism destinations.
Third, this paper’s results support a superior role of the “affective”model in explaining tou-
rists’PEBI, wherein affective attitude is a precursor of cognitive attitude. However, the “cognitive”
model in which cognitive attitude gives rise to affective attitude also has a reasonable model fit.
These results may point to a dual-process model (van der Linden, 2014; Kobbeltved et al., 2005;
Lai et al., 2012) where cognitive and affective attitudes reciprocally and simultaneously influence
each other in driving PEBI. However, this paper lacks suitable instrumental variables to estimate
this non-recursive SEM; future research can then further test the dual-process model.
Fourth, it assumes causal relationships based on the data collected from two single cross-sec-
tional surveys and may draw criticism from scholars with a different philosophy of causality. This
paper agrees that causal assumptions could be achieved through rigorous research design and
742 X. WANG ET AL.
theoretical reasoning rather than purely mathematical data analysis (van der Linden, 2014).
Future research can replicate this paper by using experimental manipulations or panel data.
Notes
1. It refers to the motivation (M) and opportunity (O) as determinants (DE) of which processing mode
(deliberative or spontaneous) is likely to work in a particular situation (Fazio, 1990, p.92).
2. GRT (Fuzhou) Data Research Co., LTD. (2017). Kulangsu World Cultural Heritage Research Report, 26.
Acknowledgments
We would like to thank the anonymous reviewers for their constructive comments, Prof. Chaozhi Zhang for his
support in survey arrangements of the two studies, and the research assistants for data collection work.
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This paper was supported by the Ministry of Culture and Tourism of the People’s Republic of China [grant numbers:
WMYC20171079] and the Ministry of Education of the People’s Republic of China [grant numbers: 14YJCZH229].
Notes on contributors
Xiongzhi Wang is a Ph.D. candidate in the School of Communication and Arts, The University of Queensland. He
gained his master’s degree in the School of Tourism Management, Sun Yat-sen University, China. His research inter-
ests include environmental psychology, sustainable tourism, and heritage tourism. Email: xiongzhi.wang@uq.net.au.
Xiangru Qin is a master student in the School of Tourism Management, Sun Yat-sen University, China. Her research
interests include green events, event tourism, social media, and environmental communication. E-mail:
qinxr3@mail2.sysu.edu.cn.
Yongbo Zhou, the corresponding author, is an associate professor in the School of Social Science, Soochow
University. His research interests include tourism marketing, destination image, and landscape of cultural heritage.
E-mail: zhouyongbo@live.cn. Address: School of Social Science, Dushu campus of Soochow University, Suzhou,
Jiangsu Province, China.
References
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes,50(2),
179–211. doi:10.1016/0749-5978(91)90020-T
Ajzen, I., & Driver, B. L. (1991). Prediction of leisure participation from behavioral, normative, and control beliefs: An
application of the theory of planned behavior. Leisure Sciences,13(3), 185–204. doi:10.1080/01490409109513137
Ajzen, I., & Driver, B. L. (1992). Application of the theory of planned behavior to leisure choice. Journal of Leisure
Research,24(3), 207–224. doi:10.1080/00222216.1992.11969889
Alcock, I., White, M. P., Taylor, T., Coldwell, D. F., Gribble, M. O., Evans, K. L., …Fleming, L. E. (2017). Green’on the
ground but not in the air: Pro-environmental attitudes are related to household behaviours but not discretionary
air travel. Global Environmental Change,42, 136–147. doi:10.1016/j.gloenvcha.2016.11.005
Andersson, M., & von Borgstede, C. (2010). Differentiation of determinants of low-cost and high-cost recycling.
Journal of Environmental Psychology,30(4), 402–408. doi:10.1016/j.jenvp.2010.02.003
Bagozzi, R. P., Yi, Y., & Baumgartner, J. (1990). The level of effort required for behaviour as a moderator of the atti-
tude-behaviour relation. European Journal of Social Psychology,20(1), 45–59. doi:10.1002/ejsp.2420200105
Barr, S., & Gilg, A. W. (2007). A conceptual framework for understanding and analyzing attitudes towards environ-
mental behaviour. Geografiska Annaler Series B: Human Geography,89(4), 361–379. doi:10.1111/j.1468-0467.2007.
00266.x
JOURNAL OF SUSTAINABLE TOURISM 743
Breckler, S. J., & Wiggins, E. C. (1989). Affect versus evaluation in the structure of attitudes. Journal of Experimental
Social Psychology,25(3), 253–271. doi:10.1016/0022-1031(89)90022-X
Carrus, G., Passafaro, P., & Bonnes, M. (2008). Emotions, habits and rational choices in ecological behaviours: The
case of recycling and use of public transportation. Journal of Environmental Psychology,28(1), 51–62. doi:10.
1016/j.jenvp.2007.09.003
Chan, L., & Bishop, B. (2013). A moral basis for recycling: Extending the theory of planned behaviour. Journal of
Environmental Psychology,36,96–102. doi:10.1016/j.jenvp.2013.07.010
Chen, G., Bao, J., & Huang, S. (2014). Developing a scale to measure backpackers’personal development. Journal of
Travel Research,53(4), 522–536. doi:10.1177/0047287513500392
Chen, X., Peterson, M. N., Hull, V., Lu, C., Lee, G. D., Hong, D., & Liu, J. (2011). Effects of attitudinal and sociodemo-
graphic factors on pro-environmental behaviour in urban China. Environmental Conservation,38(1), 45–52. doi:10.
1017/S037689291000086X
Chen, M. F., & Tung, P. J. (2010). The moderating effect of perceived lack of facilities on consumers’recycling inten-
tions. Environment and Behavior,42(6), 824–844. doi:10.1177/0013916509352833
Churchill, G. A. (1979). A paradigm for developing better measures of marketing constructs. Journal of Marketing
Research,16(1), 64–73. doi:10.1177/002224377901600110
Coelho, F., Pereira, M. C., Cruz, L., Sim~
oes, P., & Barata, E. (2017). Affect and the adoption of pro-environmental
behaviour: A structural model. Journal of Environmental Psychology,54, 127–138. doi:10.1016/j.jenvp.2017.10.008
Costarelli, S., & Colloca, P. (2004). The effects of attitudinal ambivalence on pro-environmental behavioural inten-
tions. Journal of Environmental Psychology,24(3), 279–288. doi:10.1016/j.jenvp.2004.06.001
Cottrell, S. P. (2003). Influence of sociodemographics and environmental attitudes on general responsible environ-
mental behavior among recreational boaters. Environment and Behavior,35(3), 347–375. doi:10.1177/
0013916503035003003
Crites, S. L., Fabrigar, L. R., & Petty, R. E. (1994). Measuring the affective and cognitive properties of attitudes:
Conceptual and methodological issues. Personality and Social Psychology Bulletin,20(6), 619–634. doi:10.1177/
0146167294206001
Dolnicar, S., & Leisch, F. (2008). An investigation of tourists’patterns of obligation to protect the environment.
Journal of Travel Research,46(4), 381–391. doi:10.1177/0047287507308330
Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Fort Worth, PA: Harcourt Brace Jovanovich.
Farley, S. D., & Stasson, M. F. (2003). Relative influences of affect and cognition on behavior: Are feelings more
related to blood donation intentions? Experimental Psychology,50(1), 55–62. doi:10.1027//1618-3169.50.1.55
Fazio, R. H. (1990). Multiple processes by which attitudes guide behavior: The MODE model as an integrated frame-
work. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 23, pp. 75–109). New York, NY:
Academic Press.
Fielding, K. S., Hornsey, M. J., & Swim, J. K. (2014). Developing a social psychology of climate change. European
Journal of Social Psychology,44(5), 413–420. doi:10.1002/ejsp.2058
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measure-
ment error. Journal of Marketing Research,66(6), 39–50. doi:10.2307/3151312
Gao, J., Zhang, C., & Huang, Z. (2018). Chinese tourists’views of nature and natural landscape interpretation: A gen-
erational perspective. Journal of Sustainable Tourism,26(4), 668–684. doi:10.1080/09669582.2017.1377722
Gosling, E., & Williams, K. J. (2010). Connectedness to nature, place attachment and conservation behaviour: Testing
connectedness theory among farmers. Journal of Environmental Psychology,30(3), 298–304. doi:10.1016/j.jenvp.
2010.01.005
Halpenny, E. A. (2010). Pro-environmental behaviours and park visitors: The effect of place attachment. Journal of
Environmental Psychology,30(4), 409–421. doi:10.1016/j.jenvp.2010.04.006
Han, H. (2015). Travelers’pro-environmental behavior in a green lodging context: Converging value-belief-norm the-
ory and the theory of planned behavior. Tourism Management,47,164–177. doi:10.1016/j.tourman.2014.09.014
Han, H., Jae, M., & Hwang, J. (2016). Cruise travelers’environmentally responsible decision-making: An integrative
framework of goal-directed behavior and norm activation process. International Journal of Hospitality
Management,53,94–105. doi:10.1016/j.ijhm.2015.12.005
Herzberg, F. (1966). Work and the nature of man. Oxford, UK: World Publishing.
Hinds, J., & Sparks, P. (2008). Engaging with the natural environment: The role of affective connection and identity.
Journal of Environmental Psychology,28(2), 109–120. doi:10.1016/j.jenvp.2007.11.001
Hu, X. (2017). Study on tourists’environmental protection awareness and its influencing factors: Taking Kulangsu as an
example (Master dissertation). Xiamen University, Xiamen.
Huffman, A. H., Van Der Werff, B. R., Henning, J. B., & Watrous-Rodriguez, K. (2014). When do recycling attitudes
predict recycling? An investigation of self-reported versus observed behavior. Journal of Environmental
Psychology,38, 262–270. doi:10.1016/j.jenvp.2014.03.006
Iacobucci, D. (2010). Structural equations modelling: Fit indices, sample size, and advanced topics. Journal of
Consumer Psychology,20(1), 90–98. doi:10.1016/j.jcps.2009.09.003
744 X. WANG ET AL.
Jun, J., & Arendt, S. W. (2016). Understanding healthy eating behaviors at casual dining restaurants using the
extended theory of planned behavior. International Journal of Hospitality Management,53, 106–115. doi:10.1016/
j.ijhm.2015.12.002
Kazeminia, A., Hultman, M., & Mostaghel, R. (2016). Why pay more for sustainable services? The case of ecotourism.
Journal of Business Research,69(11), 4992–4997. doi:10.1016/j.jbusres.2016.04.069
Keer, M., Putte, B., & Neijens, P. (2010). The role of affect and cognition in health decision making. British Journal of
Social Psychology,49(1), 143–153. doi:10.1348/014466609X425337
Keller, C., Bostrom, A., Kuttschreuter, M., Savadori, L., Spence, A., & White, M. P. (2012). Bringing appraisal theory to
environmental risk perception: A review of conceptual approaches of the past 40 years and suggestions for
future research. Journal of Risk Research,15(3), 237–256. doi:10.1080/13669877.2011.634523
Kline, R. B. (2010). Principles and practice of structural equation modeling (3rd Ed.). London: Guildford Press.
Kl€
ockner, C. A. (2013). A comprehensive model of the psychology of environmental behaviour-A meta-analysis.
Global Environmental Change,23(5), 1028–1038. doi:10.1016/j.gloenvcha.2013.05.014
Kobbeltved, T., Brun, W., Johnsen, B. H., & Eid, J. (2005). Risk as feelings or risk and feelings? A cross-lagged panel
analysis. Journal of Risk Research,8(5), 417–437. doi:10.1080/1366987042000315519
Koenig-Lewis, N., Palmer, A., Dermody, J., & Urbye, A. (2014). Consumers’evaluations of ecological packaging-Rational
and emotional approaches. Journal of Environmental Psychology,37,94–105. doi:10.1016/j.jenvp.2013.11.009
Kraft, P., Rise, J., Sutton, S., & Røysamb, E. (2005). Perceived difficulty in the theory of planned behaviour: Perceived
behavioural control or affective attitude? British Journal of Social Psychology,44(3), 479–496. doi:10.1348/
014466604X17533
Lai, V. T., Hagoort, P., & Casasanto, D. (2012). Affective primacy vs. cognitive primacy: Dissolving the debate.
Frontiers in Psychology,3, 243–243. doi:10.3389/fpsyg.2012.00243
Lazarus, R. S. (1981). A cognitivist’s reply to Zajonc on emotion and cognition. American Psychologist,36(2),
222–223. doi:10.1037/0003-066X.36.2.222
Lazarus, R. S. (1984). On the primacy of cognition. American Psychologist,39(2), 124–129. doi:10.1037/0003-066X.39.2.124
Lee, T. H., Jan, F. H., & Yang, C. C. (2013). Conceptualizing and measuring environmentally responsible behaviors
from the perspective of community-based tourists. Tourism Management,36, 454–468. doi:10.1016/j.tourman.
2012.09.012
Li, Q. C., & Wu, M. Y. (2019). Rationality or morality? A comparative study of pro-environmental intentions of local
and nonlocal visitors in nature-based destinations. Journal of Destination Marketing & Management,11, 130–139.
doi:10.1016/j.jdmm.2019.01.003
Line, N. D., & Hanks, L. (2016). The effects of environmental and luxury beliefs on intention to patronize green
hotels: The moderating effect of destination image. Journal of Sustainable Tourism,24(6), 904–925. doi:10.1080/
09669582.2015.1091467
Malhotra, N. K. (2005). Attitude and affect: New frontiers of research in the 21st century. Journal of Business
Research,58(4), 477–482. doi:10.1016/S0148-2963(03)00146-2
Miao, L., & Wei, W. (2013). Consumers’pro-environmental behavior and the underlying motivations: A comparison
between household and hotel settings. International Journal of Hospitality Management,32, 102–112. doi:10.
1016/j.ijhm.2012.04.008
Perkins, H. E. (2010). Measuring love and care for nature. Journal of Environmental Psychology,30(4), 455–463. doi:
10.1016/j.jenvp.2010.05.004
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral
research: A critical review of the literature and recommended remedies. Journal of Applied Psychology,88(5),
879–903. doi:10.1037/0021-9010.88.5.879
Qiu, M., Zhang, J., & Zheng, C. (2018). Exploring tourists’soundscape emotion and its impact on sustainable tourism
development. Asia Pacific Journal of Tourism Research,23(9), 862–879. doi:10.1080/10941665.2018.1494614
Ramkissoon, H., Smith, L. D. G., & Weiler, B. (2013a). Relationships between place attachment, place satisfaction and
pro-environmental behaviour in an Australian national park. Journal of Sustainable Tourism,21(3), 434–457. doi:
10.1080/09669582.2012.708042
Ramkissoon, H., Smith, L. D. G., & Weiler, B. (2013b). Testing the dimensionality of place attachment and its relation-
ships with place satisfaction and pro-environmental behaviours: A structural equation modelling approach.
Tourism Management,36, 552–566. doi:10.1016/j.tourman.2012.09.003
Rhodes, R. E., Beauchamp, M. R., Conner, M., de Bruijn, G. J., Kaushal, N., & Latimer-Cheung, A. (2015). Prediction of
depot-based specialty recycling behavior using an extended theory of planned behavior. Environment and
Behavior,47(9), 1001–1023. doi:10.1177/0013916514534066
Ru, X., Wang, S., Chen, Q., & Yan, S. (2018). Exploring the interaction effects of norms and attitudes on green travel
intention: An empirical study in eastern China. Journal of Cleaner Production,197, 1317–1327. doi:10.1016/j.jcle-
pro.2018.06.293
Schultz, P. W. (2014). Strategies for promoting proenvironmental behavior. European Psychologist,19(2), 107–117.
doi:10.1027/1016-9040/a000163
JOURNAL OF SUSTAINABLE TOURISM 745
Shi, H., Fan, J., & Zhao, D. (2017). Predicting household PM2.5-reduction behavior in Chinese urban areas: An inte-
grative model of theory of planned behavior and norm activation theory. Journal of Cleaner Production,145,
64–73. doi:10.1016/j.jclepro.2016.12.169
Slovic, P., Finucane, M. L., Peters, E., & Macgregor, D. G. (2004). Risk as analysis and risk as feelings: Some thoughts
about affect, reason, risk, and rationality. Risk Analysis : An Official Publication of the Society for Risk Analysis,
24(2), 311–322. doi:10.1111/j.0272-4332.2004.00433.x
Slovic, P., Finucane, M. L., Peters, E., & Macgregor, D. G. (2007). The affect heuristic. European Journal of Operational
Research,177(3), 1333–1352. doi:10.1016/j.ejor.2005.04.006
Steg, L. (2016). Values, norms, and intrinsic motivation to act proenvironmentally. Annual Review of Environment
and Resources,41(1), 277–292. doi:10.1146/annurev-environ-110615-085947
Steg, L., & Vlek, C. (2009). Encouraging pro-environmental behaviour: An integrative review and research agenda.
Journal of Environmental Psychology,29(3), 309–317. doi:10.1016/j.jenvp.2008.10.004
Stern, P. C. (2000). Toward a coherent theory of environmentally significant behaviour. Journal of Social Issues,
56(3), 407–424. doi:10.1111/0022-4537.00175
Taufik, D., Bolderdijk, J. W., & Steg, L. (2016). Going green? The relative importance of feelings over calculation in
driving environmental intent in the Netherlands and the United States. Energy Research & Social Science,22,
52–62. doi:10.1016/j.erss.2016.08.012
Teng, Y. M., Wu, K. S., & Liu, H. H. (2015). Integrating altruism and the theory of planned behavior to predict
patronage intention of a green hotel. Journal of Hospitality & Tourism Research,39(3), 299–315. doi:10.1177/
1096348012471383
Thøgersen, J. (2004). A cognitive dissonance interpretation of consistencies and inconsistencies in environmentally
responsible behavior. Journal of Environmental Psychology,24(1), 93–103. doi:10.1016/S0272-4944(03)00039-2
Thøgersen, J., & Crompton, T. (2009). Simple and painless? The limitations of spillover in environmental campaign-
ing. Journal of Consumer Policy,32(2), 141–163. doi:10.1007/s10603-009-9101-1
Tinsley, H. E., & Tinsley, D. J. (1987). Uses of factor analysis in counseling psychology research. Journal of Counseling
Psychology,34(4), 414–424. doi:10.1037//0022-0167.34.4.414
VandenBerg,H.,Manstead,A.S.,vanderPligt,J.,&Wigboldus,D.H.(2006). The impact of affective and cognitive
focus on attitude formation. Journal of Experimental Social Psychology,42(3), 373–379. doi:10.1016/j.jesp.2005.04.009
van der Linden, S. (2014). On the relationship between personal experience, affect and risk perception: The case of
climate change. European Journal of Social Psychology,44(5), 430–440. doi:10.1002/ejsp.2008
Verplanken, B., Hofstee, G., & Janssen, H. J. (1998). Accessibility of affective versus cognitive components of atti-
tudes. European Journal of Social Psychology,28(1), 23–35. doi:10.1002/(SICI)1099-0992(199801/02)28:1<23::AID-
EJSP843>3.0.CO;2-Z
Wan, C., Shen, G. Q., & Choi, S. (2017). Experiential and instrumental attitudes: Interaction effect of attitude and
subjective norm on recycling intention. Journal of Environmental Psychology,50,69–79. doi:10.1016/j.jenvp.2017.
02.006
Wang, X., Tu, M., Yang, R., Guo, J., Yuan, Z., & Liu, W. (2016). Determinants of pro-environmental consumption
intention in rural China: The role of traditional cultures, personal attitudes and reference groups. Asian Journal
of Social Psychology,19(3), 215–224. doi:10.1111/ajsp.12142
White, K. M., Smith, J. R., Terry, D. J., Greenslade, J. H., & McKimmie, B. M. (2009). Social influence in the theory of
planned behaviour: The role of descriptive, injunctive, and in-group norms. British Journal of Social Psychology,
48(1), 135–158. doi:10.1348/014466608X295207
Wilson, E. O. (1993). Biophilia and the conservation ethic. In S. R. Kellert, & E. O. Wilson (Eds.), The biophilia hypoth-
esis (pp. 31–41). Washington, DC: Island Press.
Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist,35 (2), 151–175.
doi:10.1037/0003-066X.35.2.151
Zajonc, R. B. (1984). On the primacy of affect. American Psychologist,39(2), 117–123. doi:10.1037/0003-066X.39.2.117
Zhang, Y., Moyle, B. D., & Jin, X. (2018). Fostering visitors’pro-environmental behaviour in an urban park. Asia
Pacific Journal of Tourism Research,23(7), 691–702. doi:10.1080/10941665.2018.1487457
746 X. WANG ET AL.
