Liu Et Al (2020) How Does Enviormental Knowledge Translate Into Pro-Envionmental Behaviors

Science of the Total Environment 728 (2020) 138126
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
How does environmental knowledge translate into pro-environmental

behaviors?: The mediating role of environmental attitudes and
behavioral intentions
Pihui Liu a,1, Minmin Teng b,1, Chuanfeng Han a,⁎
a
School of Economics and Management, Tongji University, Shanghai 200092, China
b
School of Economics and Management, Shanghai University of Electric Power, Shanghai 200090, China
H I G H L I G H T S G R A P H I C A L A B S T R A C T
• Assessing a causal chain from environ-

mental knowledge (EK) to pro-
environmental behaviors (PEB).
• Assessing the mediating role of environ-
mental attitudes and behavioral inten-
tions in transition from EK to PEB.
• EK itself does not work, only once the af-
fective system is activated does protec-
tive action become probable.
• The multigroup SEM showed that vari-
ous demographic variables have diverse
effects on the model in this study.
a r t i c l e i n f o a b s t r a c t
Article history: This paper explores how general environmental knowledge shapes environmental attitudes, behavioral inten-
Received 8 January 2020 tions, and pro-environmental behaviors in relation to the surrounding environment. Structural equation model-
Received in revised form 29 February 2020 ing (SEM) with bootstrapping estimation was conducted to assess a causal chain from environmental knowledge
Accepted 20 March 2020
to pro-environmental behaviors with a nationally representative sample of 2824 respondents from China. The re-
Available online 23 March 2020
sults show that environmental knowledge has a significant positive effect on environmental attitudes, environ-
Editor: Damia Barcelo mental attitudes have a significant positive effect on environmental behavioral intentions and pro-
environmental behaviors, and environmental behavioral intentions have a significant positive effect on pro-
Keywords: environmental behaviors. We also concluded that although environmental knowledge has no significant direct
Environmental knowledge effects on pro-environmental behaviors, it is a crucial distal variable whose significant effect is fully mediated
Pro-environmental behaviors by environmental attitudes and environmental behavioral intentions. The results of the multigroup SEM showed
Environmental attitudes that the moderating effect of external context on the relationship between environmental behavioral intentions
Behavioral intentions and pro-environmental behaviors and various demographic variables (e.g., gender, urban vs. rural residence, ed-
Structural equation modeling
ucation level, and region) have diverse effects on the model in this study.
© 2020 Elsevier B.V. All rights reserved.
⁎ Corresponding author.
E-mail address: [email protected] (C. Han).
1
Pihui Liu and Minmin Teng have contributed equally to this study and shared first authorship.
https://doi.org/10.1016/j.scitotenv.2020.138126
0048-9697/© 2020 Elsevier B.V. All rights reserved.
2 P. Liu et al. / Science of the Total Environment 728 (2020) 138126
1. Introduction scholars regard knowledge as a prerequisite for any attitudes; thus, in-
tention can also mediate the causal relationship between knowledge
Environmental issues have dominated the World Risk List for the and behaviors. Therefore, there are grounds to assume that environ-
third year in a row, just as the latest Global Risk Report issued by the mental behavioral intentions may play an intermediary role in the rela-
World Economic Forum in 2019 warns, “of all the risks, environmental tionship between environmental knowledge and pro-environmental
risks are making the world sleepwalking into disaster.” Many sociolo- behaviors. Furthermore, environmental attitudes and environmental
gists believe that environmental problems are ultimately the objective behavioral intentions may play multiple intermediary roles in the rela-
result of humans' bad environmental behavior (Williams and Cary, tionship between environmental knowledge and pro-environmental
2002). This awareness around ecological and environmental problems behaviors. These findings raise an important question about pro-
relatively emphasizes the role of each individual in addressing the cur- environmental behaviors and how they are affected by personal envi-
rent eco-crisis. As a result, there has been extensive research on increas- ronmental knowledge. Therefore, it is important and necessary to
ing public attention to environmental sustainability problems and the study in depth about how environmental knowledge affects attitudes
effect of human behaviors on the natural ecological environment (Barr and intentions.
et al., 2011; Hynes and Wilson, 2016). This fact also raises the issue of This paper presents empirical research based on a survey of 2824
pro-environmental behaviors as a priority in public debates (Li et al., Chinese residents and attempts to use the structural equation model
2019). (SEM) to evaluate the causal relationship from environmental knowl-
Pro-environmental behaviors can be defined as an act causing min- edge to pro-environmental behaviors. The premise of this study is that
imal damage to the environment, even beneficial to the environment environmental attitudes and environmental behavioral intentions may
(Steg and Vlek, 2009). Environmental behaviors can be carried out mediate the relationship of environmental knowledge between pro-
through direct and indirect means to mitigate environmental damage environmental behaviors. The study focuses on evaluating how envi-
and improving the environment individually or collectively (Jensen, ronmental knowledge shapes environmental attitudes, environmental
2002). Pro-environmental behaviors are impacted by internal and ex- behavioral intentions, and more environmentally friendly behaviors in
ternal factors (Mainieri et al., 1997; Vicente-Molina et al., 2013; Blok relation to the surrounding environment. This research can inform
et al., 2015; Juvan and Dolnicar, 2017). Environmental knowledge was strategies to reduce environmental damage through interventions
one of the most potent predictors of environmentally friendly behaviors such as the cultivation of environmental attitudes and emotions that
in Hines et al. (1987) classic meta-analysis. One is unlikely to con- aim to influence individuals' pro-environmental behaviors.
sciously care about the environment issues or act consciously in a pro- The contributions of our research are as follows. First, we expand
environmental manner if one knows nothing about the problems upon the theory of planned behavior by using the structural equation
(Gifford and Nilsson, 2014). Saripah et al. (2013) and Varela- approach in assessing the mediated or indirect effect of environmental
Candamio et al. (2018) argued that the government should attach im- knowledge on pro-environmental behaviors. Second, most empirical
portance to environmental education for the public and enhance the studies on environmental behavior focused on specific regions and pop-
level of public environmental knowledge to promote more pro- ulations with small sample sizes, however, due to personal characteris-
environmental behaviors. Hungerford and Volk (1990) believed that tics, economic level, local environmental issues, religious beliefs and
environmental education was different from other general educational other factors, pro-environmental behaviors may have significant differ-
behaviors because consciousness does not necessarily promote behav- ences. (Casaló and Escario, 2018). Our use of a nationally representative
iors. Moreover, Frick et al. (2004) and Bartiaux (2008) did not found multistage stratified sample can well reflect the pro-environmental be-
that environmental knowledge was related to ecological behaviors. haviors of Chinese people and make up for the shortcomings of previous
Subsequent studies have also shown that, while environmental studies. Third, the results indicate that environmental attitudes and en-
knowledge is important for behaviors, knowledge is insufficient to pro- vironmental behavioral intentions play an important intermediary role
mote environmentally friendly actions (Wiek et al., 2011; Paço and in the translation of general environmental knowledge into environ-
Lavrador, 2017), emphasizing the roles of sentiments and beliefs (de mentally friendly actions. Fourth, the multiple-group analysis indicates
Miranda Coelho et al., 2016). This finding implies that knowledge can the moderating effect of external context on the relationship between
promote pro-environmental behaviors only when it stimulates senti- environmental behavioral intentions and pro-environmental behaviors
ment and is absorbed and internalized by individuals. Schultz (2010) in- and various demographic variables (e.g., gender, urban vs. rural resi-
vestigated whether more environmental-friendly actions will be taken dence, education level, and region) have different influences on the re-
when individuals have much environmental knowledge and argued lationship between environmental knowledge, attitudes, behavioral
that “the changes that do result are typically limited to people who al- intentions, and pro-environmental behaviors.
ready cared about the topic.” Therefore, Carmi et al. (2015) pointed The rest of this paper as follows. Section 2 describes the theories and
out that the effect of environmental knowledge on environmentally research hypotheses. Section 3 presents data sources and research
friendly behaviors is indirect. There are other probable intermediaries methods, including the research population and sample, measure-
in a causal relationship from environmental knowledge to environmen- ments, and analytical strategy. Section 4 presents the results of this em-
tally friendly actions that may be contributing to help us understand the pirical study, including model fit testing, structural results, mediating
role of knowledge in pro-environmental behaviors (Carmi et al., 2015). effect analysis, and multigroup analysis. Section 5 discusses the study
Kaiser et al. (1999), Flamm (2009), Lee et al. (2015), Casaló and results. Finally, we provide a conclusion and issues for further research.
Escario (2018) and Casaló et al. (2019) concluded that knowledg can
foster environmental attitudes, and environmental attitudes can pro- 2. Theories and research hypotheses
mote pro-environmental behavior. Iozzi (1989) also argued that envi-
ronmental attitudes are directly related to behavior, not knowledge. 2.1. Environmental knowledge
Therefore, there are good reasons to assume that environmental atti-
tudes play an intermediary role between environmental knowledge Environmental knowledge is a major variable in explaining pro-
and behaviors. Scholars have developed various theories to illustrate environmental behaviors, indicating individuals' awareness of environ-
the correlation between attitudes and behaviors. Perhaps the most mental issues and “general knowledge of facts, concepts, and relation-
widely used is the theory of planned behavior (TPB; Ajzen and ships concerning the natural environment and its major ecosystems”
Madden, 1986), which holds that attitude affects behavioral intent, (Fryxell and Lo, 2003). Environmental knowledge in the classical sense in-
while behavioral intent in turn shapes behavior. TPB suggests that an in- cludes factual knowledge about ecosystem structures, functions, and eco-
dividual's behavioral intentions can predict behaviors well. Most system processes (Schahn and Holzer, 1990). Boerschig and De Young
P. Liu et al. / Science of the Total Environment 728 (2020) 138126 3
(1993) pointed out that environmental knowledge should include three H2. Individuals' environmental knowledge has a significant positive ef-
dimensions: issues, strategies, and action skills to address environmental fect on environmental attitudes.
problems. Jensen (2002) divided environmental knowledge into four di-
The theory of planned behavior argues that intentions play an indi-
mensions: the impact of environmental problems, the causes of environ-
rect role in predicting the behavior of many of the factors (Ajzen,
mental problems, environmental action strategies, and the relationship
1991). Thus, the causal relationship between factual knowledge and be-
between the expected environment and human beings.
haviors is also mediated by intention. Aman et al. (2012) and Wang
The measurement of environmental knowledge can be traced back
et al. (2014) mentioned that environmental knowledge and concern
to some quantitative studies in the field of environmental social science
play a vital role in eco-friendly purchasing intentions among customers.
in the early 1970s. The ecology scale was proposed earlier by Maloney
Vicente-Molina et al. (2013) also believed that ecological knowledge
and Ward (1973) and involved the systematic measurement of environ-
can promote environmental behavioral intentions and behaviors
mental knowledge. German scholars Schahn and Holzer (1990) be-
through the study of College students. Steg & Mostafa (2009) also
lieved that the environmental knowledge scale of Maroni et al.
noted that knowledge is one of the critical variables that influence con-
measures an “abstract knowledge” that has little impact on action.
sumers' willingness and behaviors to purchase environmentally
Some studies have measured the environmental knowledge of specific
friendly products. Effectiveness and social knowledge also significantly
environmental issues. Arcury et al. (1987) designed a 12-item scale
influence the formation of attitudes towards sustainable fashion prod-
for environmental knowledge about acid rain to measure respondents'
ucts (Kong et al., 2016). Visits to the aquarium and the distribution of
knowledge of the causes and consequences of acid rain. Similarly,
marine protection information manuals to give visitors more knowl-
some scholars examined the respondents' understanding of energy ap-
edge about marine conservation can drastically improve the marine en-
plications (Trotta, 2018), waste disposal (Saratale et al., 2018), and au-
vironmental protection intentions of tourists (Wyles et al., 2013). Thus,
tomotive environmental impacts (Flamm, 2009). However, the
the following hypothesis was proposed:
specific environmental knowledge emphasizes scientificity and accu-
racy and is usually limited to professionals, while the general environ- H3. Individuals' environmental knowledge has a significant positive ef-
mental knowledge is more accessible to the majority of citizens, so it fect on environmental behavioral intentions.
is of more universal significance to consider it. The Chinese scholars
Dayong and Chenyang (2007) designed a 10-item scale for general en-
vironmental knowledge to measure Chinese people's understanding of 2.2. Environmental attitudes
general environmental knowledge and found that the scale had good in-
ternal consistency. Attitude is a psychological tendency that is expressed by evaluating
Many scholars give different answers about whether environmental a particular entity with some degree of favor or disfavor (Eagly and
knowledge influences pro-environmental behaviors Levine and Strube Chaiken, 1993), and it is deemed to be a key variable in predicting be-
(2012) claimed that general ecological knowledge could significantly havior. Therefore, attitude plays a core role in multiple theories that
predict pro-environmental behaviors. Consumers with more long- aim to understand human behaviors, such as the theory of reasoned ac-
term environmental knowledge are more like to spend more money tion (TRA) and TPB (Ajzen, 1991). Furthermore, a great number of stud-
on green products (Ogbeide et al., 2015) and are willing to purchase en- ies pay special attention to people's attitudes towards the environment
vironmentally friendly products (Brosdahl and Carpenter, 2010). How- and has pointed out that, according to these theories, such attitudes
ever, additional studies argued there is no obvious relationship between should contribute to the development of pro-environmental behaviors
ecological knowledge and pro-environmental behaviors. Laroche et al. (e.g., Shen and Saijo, 2008; Duarte et al., 2017).
(2002) claimed that environmental knowledge did not predict green Essentially, there are two typical environmental attitudes used to
behaviors well. Some studies claimed that only a small part of pro- predict pro-environment behaviors: 1 attitudes towards the environ-
environmental behaviors is related to ecological knowledge (Kollmuss ment or some aspect of it (e.g., water quality) and 2 attitudes towards
and Agyeman, 2002). Although knowledge is no longer a reliable pre- pro-environment behaviors (e.g., waste classification, recycling and uti-
dictor of behavior, it is a necessary means of overcoming psychological lization) (Hines et al., 1987). In recent years, many scholars have
barriers such as unconsciousness, fear or misrepresentation (Kaiser adopted a new ecological paradigm (NEP) to measure people's environ-
and Gutscher, 2003; Kaiser et al., 2008). It is difficult to make wise envi- mental attitudes (Halkos and Matsiori, 2017; Biasutti and Frate, 2017).
ronmental choices if one has incorrect or no knowledge (Levine and NEP mainly contains the conflict and balance among human beings
Strube, 2012; Fielding and Head, 2012). Thus, the following hypothesis and nature, human rights in nature, human self-limitation and other as-
was proposed: pects, and there are 5 facets: limits to growth, rejection of exemptions,
anti-anthropocentrism, the possibility of an ecocrisis, and the fragility
H1. Individuals' environmental knowledge has a significant positive ef- of nature's balance (Dunlap et al., 2000). Stern et al. believed that the
fect on pro-environmental behaviors. NEP has the same connotation as ecological values. The NEP is an inter-
Some studies published in recent years have shown that individuals nal cognitive structure and ecological world view that affects people's
with much environmental knowledge are more positive about environ- more specific beliefs, attitudes, norms, behavioral intentions, and be-
mental issues (Hines et al., 1987; Pickett‐Baker and Ozaki, 2008). haviors. Hawcroft et al. conducted a meta-analysis of studies using the
Bamberg et al. (2003) found that environmental knowledge and atti- NEP scale from the past 30 years and found that there were many
tudes are highly correlated, and they reinforce each other, especially ways to use the NEP scale. Dunlap et al. believed that whether the NEP
in information-seeking about environmental issues. Flamm (2009) scale should be divided into two or more dimensions should be deter-
and Barber et al. (2009) observed that environmental attitude changes mined based on the results of specific studies. In view of different social
positively with the improvement of environmental knowledge and backgrounds and cultural traditions, it is understandable that the NEP
that knowledge and associated attitudes are the catalysts that drive scale has different dimensions in different sample studies. The results
green purchasing behaviors. Polonsky et al. (2012) also agreed that con- of the exploratory factor analysis show that there are two dimensions
sumers' attitudes towards green products are likely to change when in the NEP scale in China: the new environmental paradigm and the
they get more information about green products. In fact, knowledge human exceptionalism paradigm. The two-dimensional model is sup-
has proven to be one of the main predictors in various theoretical ported by confirmatory factor analysis, and the fit indices of the models
models of the relationship between behaviors and attitudes, indicating are all above 0.90. The internal concordance coefficient of the NEP scale
that knowledge has the ability to influence behaviors through attitudes. is 0.810, and the split-half reliability is 0. 822, which both meet the psy-
Thus, the following hypothesis was proposed: chometrics standard (Wang and Cheng, 2017).
Some previous research has confirmed the correlation between en- H6. Individuals' environmental behavioral intentions have a significant
vironmental attitudes and behaviors. Kollmuss and Agyeman (2002) positive effect on pro-environmental behaviors.
believed that only by changing an individual's environmental value, in-
trinsic motivation, and relevant environmental attitude could individ- To summarize the above hypotheses, the research model is shown in
uals produce positive environmental behaviors. Pe and Yavetz's Fig. 1.
research showed that environmental attitudes can positively influence
environmental behavior. Laroche et al. (2001) found that customers 3. Data and methods
with a positive attitude were more probable to buy green and energy-
saving products and would not find it inconvenient to buy green prod- 3.1. The research population and sample
ucts. In a study in India, Malik and Singhal (2017) also found that clients
with more environmental attitudes prefer to purchase environmentally The data are taken from the 2010 Chinese General Social Survey
friendly products. Turaga et al. (2010) reminded us that there are many (CGSS2010),2 which was associated collected by the Renmin University
factors, such as personal motivation and external factors, that may have of China and Hong Kong University of Science and Technology in 2010.
an impact on behaviors. However, most research models showed that Using the method of multistage stratified sampling, N10,000 individual
individuals need to change their attitudes towards the environment be- respondents covering 30 provinces in China (excluding Xinjiang and
fore they can further change their behaviors to the environment. Thus, Hainan) were interviewed in person. The respondents ranged from 18
the following hypothesis was proposed: to 72 years old. The questionnaires elicited information about
sociodemographics, living styles, social networks, personal cognitive
H4. Individuals' environmental attitudes have a significant positive ef- ability, social security, and environmental behavior (as shown in the
fect on pro-environmental behavior. Appendix A). A total of 2428 questionnaires about environmental be-
haviors were obtained. The descriptive statistics for the demographic
According to the TPB, attitude towards behavior is a remarkable pre-
variables are shown in Table 1. Xiao and Hong (2010, 2017) use this
dictor that can explain and promote behavioral intention (Ajzen, 1991).
data to study gender differences in concerns for the environment the
Kalafatis measured the appropriateness of the TPB to determine cus-
Chinese public. Wang and Cheng (2017) use this data to examine the re-
tomers' intentions to buy green products in two different cultural con-
lationships between perceived severity of environmental problems.
texts, Greece and the United Kingdom. Yadav and Pathak (2016a)
This is the first time that we have used this data to examine how general
showed that the TPB partly supports the attitude of young consumers
environmental knowledge shapes environmental attitudes, environ-
towards deciding to purchase organic food. Wan et al. (2012) also
mental behavioral intention, and more environmental-friendly
pointed out that all constructions of the TPB, especially attitudes, have
behaviors.
a positive effect on the willingness of college students to recycle.
Yadav and Pathak's (2016b) findings support the feasibility of the TPB
in predicting eco-friendly buying intentions, with attitudes significantly 3.2. Measures
influencing intentions. There is a knowledge gap in the interaction be-
tween subjective norms and attitudes, which affects the willingness to 3.2.1. Environmental knowledge
recycle. Furthermore, Relevant research has confirmed that the environ- Environmental knowledge was assessed using ten items adopted
mental attitudes of visitors can predict their environmental behavior in- from the Chinese environmental knowledge scale (CEKS) (Dayong and
tentions (Untaru et al., 2014; Miller et al., 2015). Thus, the following Chenyang, 2007; Xiao and Hong, 2010). Previous research shows that
hypothesis was proposed: the scale possesses good content validity, maintaining acceptable and
stable internal consistency across items. Accordingly, the scale can act
H5. Individuals' environmental attitudes have a significant positive ef- as a fundamental instrument for studying Chinese public environmental
fect on environmental behavioral intentions. knowledge (Dayong and Ye Chao, 2016).
This ten-item measure has two dimensions: (Ajzen, 1991) general
daily environmental knowledge (e.g., “Acid rain has nothing to do
2.3. Environmental behavioral intentions with coal burning”) and (Ajzen and Madden, 1986) general professional
environmental knowledge (e.g., “A single-species forest is more likely to
Behavioral intention is defined as a strong internal stimulus and is lead to pests and diseases”). Table A1 reports these items. Respondents
often understood as the cause of behaviors (Moisander, 2007). The the- obtain one point for each right answer. Wrong answers and the “I don't
ory of planned behavior argues that intentions play an indirect role in know” option scored zero. Due to the sum of general daily environmen-
predicting the behavior of many of the factors (Ajzen, 1991). As a direct tal knowledge items was 7, the scale ranged from 0 to 7, and the general
predictor variable of behaviors, intentions play an obvious role in the re- environmental professional knowledge scale ranged from 0 to 3. In the
lationship among attitude, perceived behavior control, subjective norm, current sample, the internal consistencies for the 7-item general daily
and pro-environmental behaviors (Untaru et al., 2014). Environmental environmental knowledge scale and 3-item general professional envi-
behavioral intention is the perceived subjective of a person participating ronmental knowledge scale in the current sample were 0.71 and 0.69,
in environmental behaviors, reflecting the disposition of the person to respectively.
participate in a specific environmental behavior (Kaiser and Gutscher,
2003). 3.2.2. Environmental attitudes
Environmental behavioral intention (EBI) is strongly or, at worst, Environmental attitudes were assessed using six items adopted from
moderately related to pro-environmental behaviors (Auhagen and the NEP scale (Dunlap et al., 2000) on a five-point Likert-type scale. Pre-
Neuberger, 1994) (Hines et al., 1987; Diekmann and Franzen, 1995). vious research using exploratory factor analysis showed that there are
Previous studies have mentioned that the EBI of visitors can be used two dimensions in the NEP scale in China: (Ajzen, 1991) the new envi-
to predict their pro-environmental behavior well (Mobley et al., 2010; ronmental paradigm (e.g., “The balance of nature is very delicate and
Lee et al., 2015). Morren and Grinstein (2016), in a study of cross- easily upset”) and (Ajzen and Madden, 1986) the human exceptional-
border environmental behavior, found that environmental behavioral ism paradigm (e.g., “Humans are severely abusing the environment”).
intention is more likely to translate into actual behavior in developed
and individualistic countries. Thus, the following hypothesis was 2
The database is public and the site for obtaining the data is http://cgss.ruc.edu.cn/
proposed: index.php?r=index/publication
Fig. 1. Research model.
Table A1 reports these items. In the current sample, the internal consis- and Gifford, 2013; Vicente-Molina et al., 2018; Zhang et al., 2019; Huang
tency was 0.7. and Ge, 2019), we included these four variables as control variables.
3.2.3. Environmental behavioral intentions 3.2.6. External context

Many scholars believe that in order to capture people's pro- Guagnano et al. (1995) proposed the ABC (Attitude-Behavior-Con-
environmental intentions, items should represent clear commitments text) theory to explain environmental behaviors, which believed that
to act pro-environmentally (e.g. “I have the intention to buy organic environmental behaviors (B) was generated from the interaction be-
food”). Table A1 reports these items. However, there are so many tween individual behavioral attitude (A) and external context
kinds of pro-environment behaviors in daily life that it is difficult to dis- (C) (Black et al., 1985; Guagnano et al., 1995; Olli et al., 2001). To ex-
tinguish specific intentions and behaviors of individuals. In fact, if indi- plore the influence of context on the proposed model, we taken the ex-
viduals are willing to accept monetary burde for pro-environmental ternal context as the control variable. External context was recorded
behavior, they can be considered to have a strong environmental behav- using a binary scale to ask respondents how to evaluate the environ-
ioral intentions. This research environmental behavioral intentions mental quality in their area (0 = high environmental quality, 1 = low
were assessed using three items adapted from the studies of environmental quality).
Tabernero and Hernández (2011) and Bock et al. (2005) on a five-
point Likert-type scale. The item measures included such statements 3.3. Analytical strategies
as “I would give part of my income if I were certain that the money
would be used to prevent environmental pollution” In the current sam- In this analysis, we accepted the recommendations of Anderson and
ple, the internal consistency was 0.83. Gerbing (1988) to validate the hypothetical model through a two-step
strategy. Firstly, the measurement model is validated by confirmatory
factor analysis (CFA), and then the fitting coefficient and path coefficient
3.2.4. Pro-environmental behaviors of the hypothetical model was measured by structural equation model-
Pro-environmental behaviors were assessed using four items ing. We used SPSS 17.0 and AMOS 25.0 for SEM analysis.
adapted from the research of Goldman et al. (2006) and Xiao and The reliability and consistency of the questionnaire data were tested
Hong (2010) on a four-point Likert-type scale ranging from 1, “never,” before the SEM analysis. According to the results, all Cronbach's alpha
to 4, “always.” Our measurements include green consumerism, water (a) coefficients exceed standard the value of 0.6, as shown in Table 2, in-
and energy use, and waste recycling (e.g., “Do you often specifically clas- dicating that the scale was reliable.
sify glass, aluminum cans, plastics or newspapers for easy recycling? ”). Confirmatory factor analysis (CFA) is part of the structural equation
Table A1 reports these items. In the current sample, the internal consis- model. Thompson (2007) proposed that the CFA should be completed
tency was 0.81. to test the structural validity before analyzing the structural equation
model. Structural validity consists of convergence validity and discrim-
3.2.5. Demographic characteristics ination validity. In this study, CFA analysis was carried out for all facets.
Due to previous research has indicated that a person's gender, urban As shown in Table 2, the factor loadings were between 0.5 and 0.9, the
vs. rural residence, region and education level may have a potential im- combined reliability (CR) was between 0.55 and 0.9, and all average var-
pact on human pro-environmental behaviors (Arcury and Christianson, iance extracted (AVE) values were between 0.35 and 0.7. All the values
1993; Rauwald and Moore, 2002; Luchs and Mooradian, 2012; Scannell exceed the standard value (Fornell and Larcker, 1981). Therefore, the
questionnaire had appropriate convergent validity.
Table 1 Discriminant validity verifies whether there are significant differ-
Demographic profile of respondents. ences between two different dimensions. As shown in Table 3, the diag-
onal values indicating the average variance of each latent variable are
Respondents' Item Frequency Percentage
characteristics (n = 2428) (%) larger than the off-diagonal values representing the square of the corre-
lation coefficient of each potential variable, which indicates that the dis-
Gender Male 1174 48.35%
Female 1254 51.65%
criminant validity of the scale data is better (Fornell and Larcker, 1981).
Urban vs. rural Urban residence 1032 42.50% Therefore, the reliability and validity of this study are excellent and can
residence be further analyzed.
Rural residence 1396 57.5%
Education level Primary education and 636 26.19%
4. Results
below
Secondary education 1286 52.97%
Higher education 506 20.84% 4.1. Model fit tests
Region Eastern China 1120 46.13%
Central China 722 29.74% When SEM is used to validate theoretical models, a satisfactory
Western China 586 24.13%
model fit is a necessary condition (Byrne, 2010). The better the
Table 2
Internal and convergent reliability.
Variable Parameter Estimation Convergent Validity Reliability
Constructs Indicators Factor load S.E. C.R. (t-value) P Standardized Factor load SMC C.R. AVE Cronbach's alpha (a)
New Environmental Paradigm (NEP) NEP 1 1.000 0.573 0.328 0.653 0.388 0.716
NEP 2 1.026 0.066 15.539 *** 0.581 0.338
NEP 3 0.449 0.095 15.231 *** 0.705 0.497
Human Exceptionalism Paradigm (HEP) NEP 4 1.000 0.624 0.389 0.651 0.383 0.732
NEP 5 0.896 0.047 16.468 *** 0.583 0.340
NEP 6 1.017 0.059 17.360 *** 0.649 0.421
Environmental Attitude (EA) NEP 1.000 0.603 0.364 0.550 0.381 0.703
HEP 1.516 0.138 11.023 *** 0.633 0.401
Environmental Knowledge (EK) EK1 1.000 0.541 0.293 0.660 0.518 0.66
EK2 3.051 0.206 14.813 *** 0.862 0.743
Environmental Behavior Intention (EBI) EBI 1 1.000 0.820 0.672 0.841 0.641 0.831
EBI 2 1.084 0.028 38.554 *** 0.893 0.797
EBI 3 0.839 0.025 33.803 *** 0.673 0.453
Pro-environmental Behavior (PEB) PEB 1 1.000 0.569 0.324 0.811 0.522 0.805
PEB 2 1.320 0.051 26.122 *** 0.798 0.637
PEB 3 1.305 0.050 26.030 *** 0.785 0.616
PEB 4 1.151 0.046 24.978 *** 0.715 0.511
model fit is, the closer the model matrix is to the sample matrix. 4.3. Mediating effect analysis
Following McDonald and Ho (2002), Boomsma (2000), Hoyle and
Panter (1995), the values of x2 /df, GFI, RMSEA, CFI, IFI, NFI, and There are many ways to test the mediation effect. The causal steps
AGFI were adopted to estimate model fit. As shown in Table 4, all approach is the most widely used method (Baron and Kenny, 1986).
of these indices exceed the standard value. But there are some defects in causal-step method; thus, another
method, the Sobel test, is becoming popular. In fact, the Sobel test is
often used as a supplement to causal-step method, not as a substitute
4.2. Structural results for it. The Sobel test has a drawback; it requires that the sampling distri-
bution of indirect effects is assumed to be normal. However, the sam-
The structural results are presented in Fig. 2 and Table 5. The pling distribution tends to be asymmetric. (Bollen and Stine, 1990;
results indicated that four hypotheses were supported, and two Stone and Sobel, 1990). Both causal steps and the Sobel test have limita-
hypotheses were rejected. The results of the positive and direct ef- tions when conducting multiple mediation analysis. Our research model
fect of environmental knowledge (EK) on environmental attitudes involves multiple mediators. Therefore, we used bootstrapping (Bollen
(EA) (standardized direct effect β = 0.74, p b 0.05) were statisti- and Stine, 1990) to test the mediating variable effects. MacKinnon
cally significant, so hypothesis 2 was thus supported. The results et al. (2007) and Williams and MacKinnon (2008) claimed that the me-
of the positive and direct effect of environmental attitudes on en- diation effect analysis with bootstrapping is more accurate than that
vironmental behavior intentions (EBI) (standardized direct effect with causal steps and the Sobel test.
β = 0.28, p b 0.05) were statistically significant, so hypothesis 4 We performed bootstrapping at a 95% confidence interval with 5000
was thus supported. Both the EA (standardized direct effect β = samples (Taylor et al., 2008). We calculated the asymptotic critical ratio
0.20, p b 0.05) and EBI (standardized direct effect β = 0.28, (Z) and the confidence interval of the lower and upper bounds (95% BC,
p b 0.05) had significant positive impacts on pro-environmental 95% percentile) to test whether the indirect effects are significant
behaviors (PEB), so hypotheses 5 and 6 were thus supported. (Preacher and Hayes, 2008). When Z N 0 and a 95% confidence interval
There were no significant relationships between EK and PEB (stan- does not contain zero, there is an indirect effect. First, we test whether
dardized direct effect β = 0.043, p N 0.05) or EK and EBI (stan- the total indirect effect is significant. As shown in Table 4, the asymp-
dardized direct effect β = 0.002, p N 0.05); therefore, hypotheses totic critical ratio for the total indirect effect of environmental knowl-
1 and 3 were rejected. edge on pro-environmental behaviors is 3.518 N 1.96, and its 95%
As shown in Fig. 2, although there is no significant direct effect of EK confidence interval does not contain zero, which leads to a rejection of
and PEB, they may have some indirect effect. There may be three poten- the null hypothesis that the total indirect effect is zero.
tial mediating effects in the research model. The first potential mediat- In the research model, we analyzed specific indirect effects between
ing effect is that EA may act as an intermediary between EK and PEB. environmental knowledge (EK) and pro-environmental behaviors
The second potential mediating effect is that EBI may act as an interme- (PEB), rather than the overall indirect effect. As shown in Table 6, spe-
diary between EK and PEB. The third potential mediating effect is that cific indirect effects are EEP 0.25 (through environmental attitudes
EA and EBI may play multiple intermediary roles in the relationship be- (EA)), EEP2 0.001 (through environmental behavioral intentions
tween EK and PEB. Therefore, we further analyze the above possible me- (EBI)), and EEEP 0.049 (through EA and EBI). After the examination of
diating effects. potential mediators, we can draw a conclusion that EA is the significant
Table 3 Table 4
Result of discriminant validity measures. Model fit summary for the proposed research model.
AVE EA PEB EK EBI Inedx X2/DF RMSEA NFI CFI GFI AGFI IFI
EA 0.38 0.62 Norm* b5 b0.05 N0.90 N0.95 N0.95 N0.95 N0.95

PEB 0.52 0.37 0.72 Value 3.905 0.035 0.967 0.975 0.983 0.975 0.976
EK 0.52 0.61 0.29 0.72 Judgment Yes Yes Yes Yes Yes Yes Yes
EBI 0.64 0.20 0.34 0.15 0.80
Note: *source: Schreiber et al., (2006).
Fig. 2. Path diagram of the structural equation model.
mediator of the EK → PEB relationship, with Z = 2.874 N 1.96, and the 4.4. Multigroup analysis
95% confidence interval does not contain zero. Therefore, greater EK
leads to greater EA, which in turn leads to greater PEB. EBI is not the sig- A multigroup structural equation model was used to explore the in-
nificant mediator of the EK → PEB relationship, with the Z = fluence path of pro-environmental behaviors among different control
0.0417 b 1.96 and a 95% confidence interval containing zero. Therefore, variables. Based on prior research, five control variables were proposed:
EBI cannot influence the relationship between EK and PEB. EA and EBI external context, gender, urban vs. rural residence, region and educa-
are the significant mediators of the EK → PEB relationship, with Z = tional. Group differences among the study measures are shown in
2.333 N 1.96, and the 95% confidence interval does not contain zero. Table 7.
Therefore, greater EK leads to greater EA, and greater EA leads to greater As shown in Table 7, the respondents in low-quality environ-
EBI, which in turn leads to greater PEB. We examined the specific indi- ments had higher environmental knowledge (EK), environmental
rect effects found that there are two mediating effects of “EK→ EA→ attitude (EA), environmental behavioral intentions (EBI), and pro-
PEB” and “EK→ EA→ EBI→ PEB.” As shown in Table 6, the examination environmental behavior (PEB) than the respondents in high-
of the contrasts of the indirect effects shows that the specific indirect ef- quality environments. Males had significantly more EK than females,
fect through EA is larger than the specific indirect effect through EA and while there was no difference in their EA, EBI, and PEB. The rural re-
EBI, with the Z = 2.434 N 1.96, and the 95% confidence interval does not spondents had higher EK, EA, and PEB than urban respondents. As
contain zero. expected, the respondents with higher education levels had higher
Although environmental knowledge does not exert obvious impacts EK, EA, and EBI, but there was no significant difference in PEB. Re-
on pro-environmental behaviors, it has significant indirect effects on spondents from different regions did not differ significantly in their
pro-environmental behaviors with environmental attitudes as an inter- EK and EBI. The respondents from eastern China had higher EA
mediary; furthermore, environmental knowledge has an indirect influ- than the respondents from central China and western China, while
ence on pro-environmental behaviors by taking advantage of the those in the central region had significantly higher EA than those in
double mediating effects arising from environmental attitudes and en- the western regions. The respondents from eastern China had signif-
vironmental behavioral intentions. Obviously, environmental attitudes icantly more PEB than central China and western China, while there
are of great significance to the relationship between environmental was no difference in PEB between the respondents from the central
knowledge and pro-environmental behaviors. and western regions.
Regarding the fit indices for the nested sequence of models in multi-
group analysis, as shown in Table 8, all of these indices were greater
Table 5 than the standard value, so that the further multigroup analysis
Results of the structural equation model and hypothesis testing. could be carried out. We conducted a series of multigroup SEM to
discover any changes in the measurement parameters and structural
Path Estimate S.E. C.R. P Hypothesis Supported
relationship patterns in the proposed model across the different de-
PEB b ---EK 0.043 0.051 0.079 0.641 H1 NO mographic samples. We followed the invariance routine suggested
EA b ---EK 0.744 0.033 12.59 *** H2 YES
EBI b ---EK 0.002 0.109 0.025 0.98 H3 NO
by Byrne et al. (1989) in which measurement parameters are initially
PEB b ---EA 0.282 0.177 3.431 *** H4 YES limited to be equivalent across samples, namely, the factor loadings
EBI b ---EA 0.197 0.234 2.541 0.011* H5 YES (FL), factor correlations (FC), and structural parameters (SP)
PEB b ---EBI 0.278 0.018 10.891 *** H6 YES representing the hypothesized relationships among the model
P: *p b 0.05, **p b 0.01, ***p b 0.001. constructs.
Table 6
Direct, indirect and total effects† of the SEM components, depicted in Fig. 2.
Effect source Point estimate Product of coef BCEBC 95% Percentile 95%
SE Z Lower Upper Lower Upper
Indirect Effects
TOTAL 0.299 0.085 3.518 0.173 0.502 0.167 0.496
PEB b ---EA b ---EK (EEP) 0.250 0.087 2.874 0.124 0.459 0.117 0.448
PEB b ---EBI b ---EK (EEP2) 0.001 0.024 0.0417 −0.049 0.047 −0.051 0.046
PEB b ---EBI b ---EA b ---EK (EEEP) 0.049 0.021 2.333 0.011 0.096 0.01 0.096
Direct Effects
PEB b ---EK 0.051 0.091 0.561 −0.162 0.207 −0.162 0.208
Total Effects
TOTAL 0.350 0.032 10.938 0.293 0.418 0.291 0.416
Contrasts
EEP vs EEEP 0.202 0.083 2.434 0.073 0.403 0.064 0.397
Note: 5000 bootstrap samples.
Analysis of factorial invariance across external context. First, to test rural areas; (Aman et al., 2012) eastern and central China present stron-
the invariance of the factor loadings (FL), the FL of the two groups ger relationship between environmental knowledge and environmental
were limited to being equal. The chi-square value (CMIN) difference attitudes than western China; (Anderson and Gerbing, 1988) no differ-
test between the baseline model (BM) and the constrained model ence in the measurement parameters and structural relationships
(CM) was not noticeable (CMIN = 22.03, p N 0.05), showing that the across all the education level groups. Limited by the length of the
FL of the two groups were invariant. To test the invariance of the struc- paper, the detailed results of different demographic variables complete
tural parameters (SP), the CMIN difference test between the BM and the invariance model are presented in Appendix B.
CM was noticeable (CMIN = 14.51, p b 0.05), showing that the SP of the
two groups were variant. Finally, to test the invariance of factor correla- 5. Discussion
tions (FC), the CMIN difference test between the BC and the CM was not
noticeable (CMIN = 5.43, p N 0.05), showing that factor correlations of The results of this study make four contributions to the environmen-
the two groups were invariant. The parameter estimates of this com- tal psychology and sustainable development literature. First, in the Chi-
plete invariance model are presented in Table 9. nese context, nearly all of the relationships among environmental
Because the two groups of structural parameters are different, it is knowledge, environmental attitudes, environmental behavioral inten-
necessary to further analyze which paths are different. This study tions, and pro-environmental behaviors that have been found in previ-
followed Bentler's (1980) suggestion to use the critical value ratio (Z), ous research were replicated in this study. Many of the studies on
which is formed by dividing the standardized path coefficient by its environmental behaviors originate from developed countries such as
standard error to judge whether there is a difference in each hypothet- the USA, Sweden, Spain, and Britain (Diamantopoulos et al., 2003;
ical path between the two groups. Atone needs to examine the value of Casaló and Escario, 2018; Park, 2019; Chng et al., 2019). Despite some
the critical value ratio (Z), and if it is greater than±1.96, it is significant recent expansion in emerging countries, little is known about pro-
at the 0.05 level. As shown in Table 10, the path H6 difference test be- environmental behaviors in China (Zhou et al., 2018). This study is an
tween high environmental quality and low environmental quality was exploration of environmental behavior in the Chinese context. We
significant (Z = 2.312 N 1.96), indicating that the respondents in low- found that environmental knowledge significantly positively affects en-
quality environments present stronger relationships between environ- vironmental attitudes. This finding conforms to that of past studies by
mental behavioral intentions and pro-environmental behavior than Roberts (1996), who reported evidence that individuals with more gen-
the respondents in high-quality environments. eral environmental knowledge are more probable to be aware of envi-
Analysis of factorial invariance across the different demographic ronmental problems. Kaiser et al. (1999), Oreg and Katz-Gerro (2006),
samples. The results showed that (Ajzen, 1991) females present stron- and Zsóka et al. (2013) also affirmed that environmental knowledge is
ger relationships between environmental knowledge and environmen- the premise to produce positive environmental attitudes. We found
tal attitudes than males; (Ajzen and Madden, 1986) no difference in the that environmental attitudes significantly positively affect environmen-
measurement parameters and structural relationships across urban and tal behavioral intentions and pro-environmental behaviors. Consistent
Table 7
Group differences for measures of EK, EA, EBI, PEB.
Item EK EA EBI PEB
Mean SD p Mean SD p Mean SD p Mean SD p
High environmental quality 4.84 2.27 0.000 16.56 3.54 0.000 8.68 2.97 0.005 8.52 2.99 0.000
Low environmental quality 6.15 2.47 *** 19.14 3.84 *** 9.21 2.83 ** 9.48 2.95 ***
Male 6.07 2.55 0.000 18.75 3.87 9.10 2.94 9.18 2.97 0.32
Female 5.59 2.51 *** 18.48 3.80 0.47 8.92 2.78 0.11 9.24 3.01
Urban residence 4.87 2.42 0.000 17.56 3.66 0.000 8.90 2.95 0.10 8.30 2.77 0.001
Rural residence 6.53 2.39 *** 19.39 3.78 *** 9.09 2.80 9.88 2.96 ***
Primary education 4.27 2.35 0.000 16.86 3.34 0.013 8.62 3.00 0.000 8.25 2.96 0.16
Secondary education 5.92 2.40 *** 18.75 3.80 *** 9.04 2.88 *** 9.31 2.95
Higher education 7.512 1.89 20.47 3.56 9.42 2.56 10.17 2.77
Eastern China 6.32 2.50 0.51 19.26 3.81 0.018 9.03 2.74 0.06 9.92 3.06 0.001
Central China 5.47 2.44 18.28 3.49 *** 8.63 2.92 8.36 2.79 ***
Western China 5.30 2.56 17.64 4.14 9.46 2.94 8.80 2.72
P: *p b 0.05, **p b 0.01, ***p b 0.001.

Table 8 Table 10
Fit indices for nested sequence of models in multigroup analysis. Comparison of differences in path coefficients between different environmental quality
groups.
Inedx X2/DF RMSEA NFI CFI GFI AGFI IFI
Hypothesis Path Male Female Z
Norm* b5 b0.05 N0.90 N0.95 N0.95 N0.95 N0.95
External context 2.472 0.026 0.941 0.963 0.967 0.961 0.963 H1 PEB b ---EK 0.0180* 0.106 0.106
Gender 2.624 0.026 0.957 0.973 0.977 0.966 0.973 H2 EA b ---EK 0.672 0.720*** 1.120
Urban and rural 2.554 0.025 0.955 0.972 0.978 0.967 0.972 H3 EBI b ---EK 0.055 0.098 −0.759
Education 1.951 0.020 0.919 0.974 0.974 0.963 0.974 H4 PEB b ---EA 0.251 0.383 0.65
region 2.025 0.021 0.950 0.973 0.960 0.96 0.973 H5 EBI b ---EA 0.017 0.072 0.440
H6 PEB b ---EBI 0.028 0.281*** 2.312
with past research, environmental attitudes were found to be a strong

predictor of environmental behavioral intentions (Rajapaksa et al., as it might help to arouse our attention and concern, thus creating a
2019; Yadav and Pathak, 2016a; Taufique and Vaithianathan, 2018) starting point for a willingness to act.
and pro-environmental behaviors (Mobley et al., 2010; Paço et al., Third, this research found that there are two full mediating effects
2010). Goldman and Yavetz's research shows that environmental atti- between environmental knowledge and pro-environmental behaviors,
tudes can positively influence environmental behavior. Laroche et al. “EK→ EA→ PEB (EEP)” and “EK→ EA→ EBI→ PEB (EEEP)”, and the medi-
(2001) mentioned that customers with higher environmental attitudes ating effect through environmental attitude is larger than the mediating
will buy more green products. This finding conforms to the TPB; attitude effect through environmental attitudes and environmental behavior in-
towards behavior is an effective predictor that can explain and promote tention. Therefore, environmental attitudes play a central role in the
behavioral intention (Ajzen, 1991; Untaru et al., 2014). We found that transformation of environmental knowledge into behaviors. Likewise,
environmental behavioral intentions significantly positively affect pro- consistent with past research (e.g., Kaiser et al., 1999; Barber et al.,
environmental behaviors. The results are the same as the TPB; behav- 2009), environmental attitudes are the core intermediary variable be-
ioral intentions are direct predictor variables of behaviors (Ajzen, tween EK and PEB. Indeed, even after considering Other significant var-
1991). Likewise, consistent with past research, environmental behav- iables, we claimed that environmental attitudes were a powerful and
ioral intention is strongly related to pro-environmental behaviors significant predictor of pro-environmental behaviors. It is generally be-
(Auhagen and Neuberger, 1994; Schahn and Holzer, 1990). lieved that decision making depends on two side-by-side disposal sys-
Second, our results show that more general environmental knowl- tems: a cognizant analysis system that makes logical decisions based
edge does not necessarily lead to more positive environmental behav- on evidence and an affective system. The influence of emotional systems
ioral intentions and behaviors. In past studies, environmental is predominant, especially in the case of indeterminacy, which is typical
knowledge was regarded as the most direct predictor of environmental of most environmental issues (Weber, 2006). This fact also explains
behavior (Mostafa, 2007; Polonsky et al., 2012). However, many other why environmental knowledge had an insignificant impact on behav-
studies have shown that the direct relationship between environmental iors. As mentioned earlier, recognition itself does not motivate us to en-
knowledge and pro-environmental behaviors is weak (Laroche et al., gage in pro-environmental behaviors, but once the emotional system is
2002). Kollmuss and Agyeman (2002) concluded that it is no obvious inspired, protective action is more probable to occur (Weber, 2006). En-
direct relationship between general environmental knowledge and vironmental attitudes and environmental behavioral intentions play the
PEB. These results suggested despite individuals holding some environ- role of environmental emotions to a certain extent. As we discuss the ur-
mental knowledge, this factor alone does not drive more positive pro- gency and necessity of pro-environment behaviors, Jane Goodall beauti-
environmental intentions and behaviors. The fact that environmental fully expounded the sequence of knowledge-emotion-action: “Only if
knowledge by itself does not lead to pro-environmental behaviors we understand can we care. Only if we care will we help. Only if we
may be caused by many factors. These factors will now be discussed help shall we be saved” (Denys and Holmes, 1998).
here: 1. Traditional environmental knowledge, because it is taught in Finally, the multiple-group analysis shows that external context and
schools, is not action-oriented. 2. Environmental education in schools different demographic variables have different effects on the model in
has traditionally focused on the transfer of knowledge to students. this study. We found the moderating effect of external context on the
Thus, environmental education fails to provide students with the oppor- relationship between EBI and PEB, which supports the ABC theory to a
tunity to actively integrate such knowledge. 3. As Scannell and Gifford certain extent. When the environmental damage in the residential
(2013) noted, “the changes that do result are typically limited to people area are serious, individuals will be urged to turn EBI into PEB. There-
who already cared about the topic.” Therefore, environmental knowl- fore, to encourage more PEB, citizens should be made aware of the se-
edge can only drive pro-environmental behaviors for those who already verity of the current environmental problems. We found that females
cared about environmental issues. In the past, Chinese people were gen- present stronger positive impacts of environmental knowledge on envi-
erally considered to lack environmental awareness. Although great ronmental attitudes than males. Females are more likely to feel uneasy
progress has been made in recent years, many Chinese people still about environmental damage (Gambro and Switzky, 1999; Levine and
need to improve their environmental awareness. Therefore, it is not dif- Strube, 2012). When females learn more about the environment, they
ficult to understand why environmental knowledge does not lead to en- have a more positive environmental attitude than males. We found
vironmental behavior in the Chinese context. Although general that rural respondents displayed significantly more PEB than urban re-
environmental knowledge cannot directly predict environmental ac- spondents. Consistent with previous research, In Britain, students who
tions, its indirect and mediating effects should not be underestimated, grew up in the countryside tend to be more natural than those who
grew up in the city (Hinds and Sparks, 2008). People from rural areas
are closer and more dependent on nature, prompting them to engage
in more pro-environmental behaviors. We found that respondents
Table 9
with higher education levels had higher EK, EA, and PEB. There is no
Invariance tests across external context.
doubt that the more educated a person is, the greater his or her environ-
Model comparison Df diff CMIN diff P-value mental knowledge must be (Manzanal and Tudela, 2015). Our results
Unconstrained vs Measurement weights 10 22.03 0.15 show that there are significant differences in the EA and PEB of respon-
Measurement weights vs Structural weights 7 14.51 0.015** dents from different regions. Overall, the EA and PEB of respondents in
Structural weights vs Structural covariances 1 5.43 0.063 eastern and central China were higher than those from western respon-
P: *p b 0.05, **p b 0.01, ***p b 0.001. dents. China is a typical large country, so there are evident disparities in
economic and social development among different areas. Eastern China longitudinal data or causal experiments should be used in future studies
has the highest level of economic and social development and quality of to clarify he claims from the present study. Second, the data for this
life, followed by central China and western China. Therefore, it is not dif- study came from a comprehensive national survey. Therefore, the
ficult to understand why eastern and central respondents have higher scale used to measure variables in this study is not developed for our re-
EA and PEB rates than western respondents. An unexpected finding search purpose. Although we have tested the internal consistency reli-
was that the respondents in western China present a stronger positive ability and structural validity of each questionnaire, future studies
impact of environmental knowledge on environmental attitudes than may need to adopt more standardized measurement tools to further
respondents in eastern and central China. This result may be because verify these results. Third, part of the theoretical model of this study is
people in western China are more dependent on nature for production based on the theory of planned behavior, but due to the limitations of
through farming and grazing. Although the original environmental the data obtained, we did not include subjective norm and perceived be-
knowledge and attitudes of people in western China are low, when havioral control into the model for analysis. Although our research fo-
they acquire environmental knowledge, a more positive environmental cuses on the relationship among environmental knowledge,
attitude is promoted. environmental attitude, environmental behavior intention and pro-
environmental behaviors, we can integrate environmental knowledge
6. Conclusion and directions for future research into a complete TPB model for further research in the future. Fourth,
this study measures general environmental knowledge. Polonsky et al.
The objectives of this study were to examine how general environ- (2012) pointed out that knowledge could be general or more specific.
mental knowledge (EK) shapes environmental attitudes (EA), environ- General environmental knowledge and specific environmental knowl-
mental behavioral intention (EBI), and more environmentally friendly edge may lead to various types of behaviors. Barber et al. (2009) claimed
behaviors. This study used the SEM to evaluate a causal chain from EA that only particular product ecological knowledge influenced con-
to pro-environmental behavior (PEB). We estimated the model using sumers' purchase of environmentally friendly products, while general
a nationally representative multistage stratified sample of Chinese peo- environmental knowledge was not important at all. Further research
ple. The results showed that EK significantly positively affects EA, EA sig- on specific environmental knowledge will supplement this research.
nificantly positively affect EBI and PEB, and EBI significantly positively Clearly, emotional responses to environmental attitudes and intentions
affect PEB. EK did not have a significant direct effect on PEB. However, rely on many situational variables. Therefore, there is a reason to ex-
the indirect effects of EK on PEB were significant, and the influence of plore what leads to environmental emotions, especially considering
EK on PEB was completely mediated by EA and EBI. The results further their main contribution to environmentally friendly behaviors, which
show the moderating effect of external context on the relationship be- is our further research direction.
tween environmental behavioral intentions and pro-environmental be-
haviors and different demographic variables (e.g., gender, urban vs. CRediT authorship contribution statement
rural residence, education level, and region) have different effects on
the model in this study. We believe that our research has made an out- Pihui Liu: Conceptualization, Formal analysis, Writing - original
standing contribution to theory on transforming environmental knowl- draft. Minmin Teng: Data curation, Writing - review & editing, Method-
edge into behavior, and it also provides a new method and basis for ology. Chuanfeng Han: Conceptualization, Methodology, Project ad-
implementing environmental education. Human behaviors are the ministration, Funding acquisition.
root cause of environmental problems. The greatest challenge to envi-
ronmental issues is to find a way to promote sustainable lifestyles or en- Declaration of competing interest
vironmental behaviors. Many people believe that with the
improvement of environmental knowledge, individuals will be encour- We declare that we have no financial and personal relationships
aged to engage in more ecological behaviors (Brosdahl and Carpenter, with other people or organizations that can inappropriately influence
2010; Ogbeide et al., 2015). However, we decided that knowledge per our work, there is no professional or other personal interest of any na-
se does not work; protective behaviors are more probable to occur ture or kind in any product, service and/or company that could be con-
only when the emotional system is stimulated. We suggest that envi- strued as influencing the position presented in, or the review of, the
ronmental education activities be conducted more on the basis of emo- manuscript entitled, “How Does Environmental Knowledge Translate
tion than on perception. Environmental attitudes and emotions can be into Pro- Environmental Behaviors?: The Mediating Role of Environ-
motivated through nonformal educations, such as watching environmental Attitudes and Behavioral Intentions”.
mentally friendly movies, visiting nature reserves. Decision makers
should use effective means or propaganda to promote residents' per- Acknowledgements
ception and awareness of environmental pollution, enhance the correla-
tion between individuals and their environment, so as to make the This study was supported by grants from the National Natural
individual's environmental behavioral intentions transform into envi- Science Foundation of China (Grants no. 71972127, 71874123,
ronmental protection behaviors. Other strategies include concretizing 71974122, 71503185), the Ministry of Education of China (Grants
environmental impacts as more vivid, urgent, personalized or using no. 18JZDW02), the Shanghai Science and Technology Committee
self-auditing to link environmental impacts to individual daily lives. (Grants no. 18DZ1206800, 19DZ1209202). We also appreciate
This study has some limitations as follows. First, we use cross- Prof. Chao Zhang of Tongji University for his valuable suggestions
sectional data, which may have endogenous issue, which is a common which help us a lot in improving the logic and presentation of
problem in empirical research based on single survey data. Therefore, the paper.
Appendix A
Table A1
Questionnaire items.
Constructs Items content
General daily environmental knowledge 1 Carbon dioxide contributes to the creation of the greenhouse effect
2 Overusing fertilizer and pesticide will damage the environment
Table A1 (continued)
Constructs Items content
3 Using detergent powder containing phosphorus will not cause water pollution
4 Chlorofluorocarbons (CFC) emission from refrigerators is one of the causes to ozone depletion
5 Acid rain has nothing to do with coal burning
6 Due to the interdependency among species, one species vanishing will cause chain reactions
7 The increasing carbon dioxide in the atmosphere is one of the factors causing a warming climate
General professional environmental 1 A single-species forest is more likely to lead to pests and diseases
knowledge
2 In a report of air quality, level-three air quality is better than level-one air quality
3 In a report on water pollution, V (Arcury et al., 1987) water quality is better than I (Ajzen, 1991) water quality
New Environmental Paradigm (NEP) NEP The balance of nature is strong enough to cope with the impacts of modern industrial nations
1
NEP The balance of nature is very delicate and easily upset
2
NEP If things continue on their present course, we will soon experience a major ecological catastrophe
3
Human Exceptionalism Paradigm (HEP) NEP Humans are severely abusing the environment
4
NEP Humans are born to be the master, to rule over the rest of nature.
5
NEP Human beings are the most important, in order to meet the our own needs we can change the natural environment.
6
Environmental Behavior Intention (EBI) EBI 1 I would give part of my income if I were certain that the money would be used to prevent environmental pollution.
EBI 2 I would agree to an increase in taxes if the extra money were used to prevent environmental pollution.
EBI 3 The Government should reduce environmental pollution, but it should not cost me any money.
Pro-environmental Behaviors (PEB) PEB 1 Do you often specifically classify glass, aluminum cans, plastics or newspapers for easy recycling?
PEB 2 Will you reduce the consumption of energy or fuel such as oil, gas, electricity, etc. in your home to protect the
environment?
PEB 3 Do you often save or reuse water for environmental protection?
PEB 4 Do you often avoid buying certain products for the sake of environmental protection?
Appendix B
Table B1
Invariance tests across gender.
Model comparison Df diff CMIN diff P-value
Unconstrained vs Measurement weights 10 8.226 0.607

Measurement weights vs Structural weights 7 21.072 0.004**
Structural weights vs Structural covariances 1 0.006 0.938
P: *p b 0.05, **p b0.01, ***p b 0.001.
Table B2
Comparison of differences in path coefficients between different gender groups.
Hypothesis Path Male Female Z
H1 PEB b ---EK 0.120 −0.044 −1.165

H2 EA b ---EK 0.696*** 0.757*** 2.613
H3 EBI b ---EK 0.056 −0.030 −0.69
H4 PEB b ---EA 0.142 0.435*** 1.52
H5 EBI b ---EA 0.137 0.218** 0.104
H6 PEB b ---EBI 0.341*** 0.218*** −1.367
Table B3
Invariance tests across urban vs. rural residence.

Measurement weights vs Structural weights 7 6.15 0.522
P: *p b 0.05, **p b 0.01, ***p b 0.001.
Table B4
Invariance tests across different region groups.
Eastern China vs Central China vs Western China
(continued on next page)

Table B4 (continued)

Measurement weights vs Structural weights 14 42.806 0.000***
Eastern China vs Central China

Central China vs Western China

Structural weights vs Structural covariances 1 4.462 0.035**
Eastern China vs Western China

P: *p b 0.05, **p b 0.01, ***p b 0.001.
Table B5
Comparison of differences in path coefficients between different region groups.
Hypothesis Path E and C W Z
H1 PEB b ---EK 0.085 0.248 0.796

H2 EA b ---EK 0.705*** 0.829*** 3.964
H3 EBI b ---EK 0.003 0.238 1.137
H4 PEB b ---EA 0.207** 0.091 −1.254
H5 EBI b ---EA 0.17** 0.09 −0.866
H6 PEB b ---EBI 0.29*** 0.311*** −0.195
Table B6
Invariance tests across education Levels
.Model comparison Df diff CMIN diff P-value

Structural weights vs Structural covariances 2 31.822 0.07**
P: *p b 0.05, **p b 0.01, ***p b 0.001.
References Biasutti, M., Frate, S., 2017. A validity and reliability study of the attitudes toward sustain-
able development scale. Environ. Educ. Res. 23 (2), 214–230.
Ajzen, I., 1991. The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 50 (2), Black, J.S., Stern, P.C., Elworth, J.T., 1985. Personal and contextual influences on
179–211. househould energy adaptations. J. Appl. Psychol. 70 (1), 3.
Ajzen, I., Madden, T.J., 1986. Prediction of goal-directed behavior: attitudes, intentions, Blok, V., Wesselink, R., Studynka, O., Kemp, R., 2015. Encouraging sustainability in the
and perceived behavioral control. J. Exp. Soc. Psychol. 22 (5), 453–474. workplace: a survey on the pro-environmental behavior of university employees.
Aman, A.L., Harun, A., Hussein, Z., 2012. The influence of environmental knowledge and J. Clean. Prod. 106, 55–67.
concern on green purchase intention the role of attitude as a mediating variable. Brit- Bock, G.W., Zmud, R.W., Kim, Y.G., Lee, J.N., 2005. Behavioral intention formation in
ish Journal of Arts and Social Sciences 7 (2), 145–167. knowledge sharing: examining the roles of extrinsic motivators, social-
Anderson, J.C., Gerbing, D.W., 1988. Structural equation modeling in practice: a review psychological factors, and organizational climate. MIS Q. 29 (1), 87–111.
and recommended two-step approach. Psychol. Bull. 103 (3), 411–423. Boerschig, S., De Young, R., 1993. Evaluation of selected recycling curricula: educating the
Arcury, T.A., Scollay, S.J., Johnson, T.P., 1987. Sex differences in environmental concern and green citizen. J. Environ. Educ. 24 (3), 17–22.
knowledge: the case of acid rain. Sex Roles 16 (9–10), 463–472. Bollen, K.A., Stine, R., 1990. Direct and indirect effects: classical and bootstrap estimates of
Arcury, T.A., Christianson, E.H., 1993. Rural-urban differences in environmental knowl- variability. Sociol. Methodol. 115–140.
edge and actions. J. Environ. Educ. 25 (1), 19–25. Boomsma, A., 2000. Reporting analyses of covariance structures. Struct. Equ. Model. 7 (3),
Auhagen, A.E., Neuberger, K., 1994. Verantwortung gegenüber der Umwelt: Eine Studie 461–483.
über umweltbewußtes Handeln. Gruppendynamik. 26 pp. 319–332. Brosdahl, D.J., Carpenter, J.M., 2010. Consumer knowledge of the environmental impacts
Bamberg, S., Ajzen, I., Schmidt, P., 2003. Choice of travel mode in the theory of planned of textile and apparel production, concern for the environment, and environmentally
behavior: the roles of past behavior, habit, and reasoned action. Basic Appl. Soc. friendly consumption behavior. J. Text. Appar. Technol. Manag. 6 (4), 1–9.
Psychol. 25 (3), 175–187. Byrne, B.M., 2010. Structural Equation Modeling With AMOS: Basic Concepts, Applica-
Barber, N., Taylor, C., Strick, S., 2009. Wine consumers’ environmental knowledge and at- tions, and Programming (Multivariate Applications Series). 396. Taylor & Francis
titudes: influence on willingness to purchase. International Journal of Wine Research Group, New York, pp. 73–84.
1 (1), 59–72. Byrne, B.M., Shavelson, R.J., Muthén, B., 1989. Testing for the equivalence of factor covari-
Baron, R.M., Kenny, D.A., 1986. The moderator–mediator variable distinction in social psy- ance and mean structures: the issue of partial measurement invariance. Psychol. Bull.
chological research: Conceptual, strategic, and statistical considerations. J. Pers. Soc. 105 (3), 456–467.
Psychol. 51 (6), 1173–1185. Carmi, N., Arnon, S., Orion, N., 2015. Transforming environmental knowledge into behav-
Barr, S., Shaw, G., Gilg, A.W., 2011. The policy and practice of ‘sustainable lifestyles’. ior: the mediating role of environmental emotions. J. Environ. Educ. 46 (3), 183–201.
J. Environ. Plan. Manag. 54 (10), 1331–1350. Casaló, L.V., Escario, J.J., 2018. Heterogeneity in the association between environmental at-
Bartiaux, F., 2008. Does environmental information overcome practice titudes and pro-environmental behavior: a multilevel regression approach. J. Clean.
compartmentalisation and change consumers’ behaviors? J. Clean. Prod. 16 (11), Prod. 175, 155–163.
1170–1180. Casaló, L.V., Escario, J.J., Rodriguez-Sanchez, C., 2019. Analyzing differences between dif-
Bentler, P.M., 1980. Multivariate analysis with latent variables: Causal modeling. Annu. ferent types of pro-environmental behaviors: do attitude intensity and type of
Rev. Psychol. 31 (1), 419–456. knowledge matter? Resour. Conserv. Recycl. 149, 56–64.
Chng, S., White, M.P., Abraham, C., Skippon, S., 2019. Consideration of environmental fac- Laroche, M., Bergeron, J., Barbaro-Forleo, G., 2001. Targeting consumers who are willing to
tors in reflections on car purchases: attitudinal, behavioral and sociodemographic pay more for environmentally friendly products. J. Consum. Mark. 18 (6), 503–520.
predictors among a large UK sample. J. Clean. Prod. 230, 927–936. Laroche, M., Tomiuk, M.A., Bergeron, J., Barbaro-Forleo, G., 2002. Cultural differences in
Dayong, H., Chenyang, X., 2007. Sociological analysis on gender difference of environmen- environmental knowledge, attitudes, and behaviours of Canadian consumers. Can.
tal concern [J]. Soc. Stud. 2, 111–135. J. Adm. Sci. 19 (3), 267–282.
Dayong, H., Ye Chao, F., 2016. Public environmental knowledge measurement: the pre- Lee, T.H., Jan, F.H., Huang, G.W., 2015. The influence of recreation experiences on environ-
sentation and testing of a local scale. Journal of renmin university of china 30 (4), mentally responsible behavior: the case of Liuqiu Island, Taiwan. J. Sustain. Tour. 23
110–121 (Chinese). (6), 947–967.
de Miranda Coelho, J.A.P., Gouveia, V.V., de Souza, G.H.S., Milfont, T.L., Barros, B.N.R., 2016. Levine, D.S., Strube, M.J., 2012. Environmental attitudes, knowledge, intentions and be-
Emotions toward water consumption: conservation and wastage. Revista haviors among college students. J. Soc. Psychol. 152 (3), 308–326.
Latinoamericana de Psicología 48 (2), 117–126. Li, D., Zhao, L., Ma, S., Shao, S., Zhang, L., 2019. What influences an individual’s pro-
Denys, P., Holmes, M., 1998. Animal Magnetism: At Home With Celebrities & Their Animal environmental behavior? A literature review. Resour. Conserv. Recycl. 146, 28–34.
Companions. Smithmark Pub. Luchs, M.G., Mooradian, T.A., 2012. Sex, personality, and sustainable consumer behavior:
Diamantopoulos, A., Schlegelmilch, B.B., Sinkovics, R.R., Bohlen, G.M., 2003. Can socio- elucidating the gender effect. J. Consum. Policy 35 (1), 127–144.
demographics still play a role in profiling green consumers? A review of the evidence MacKinnon, D.P., Fairchild, A.J., Fritz, M.S., 2007. Mediation analysis. Annu. Rev. Psychol.
and an empirical investigation. J. Bus. Res. 56 (6), 465–480. 58, 593–614.
Diekmann, A., Franzen, A., 1995. Umwelthandeln zwischen Moral und Okonomie Ecolog- Mainieri, T., Barnett, E.G., Valdero, T.R., Unipan, J.B., Oskamp, S., 1997. Green buying: the
ical behavior among moral and economy. University of Bern, Bern, Switzerland. influence of environmental concern on consumer behavior. J. Soc. Psychol. 137 (2),
Duarte, R., Escario, J.J., Sanagustín, M.V., 2017. The influence of the family, the school, and 189–204.
the group on the environmental attitudes of European students. Environ. Educ. Res. Malik, C., Singhal, N., 2017. Consumer environmental attitude and willingness to purchase
23 (1), 23–42. environmentally friendly products: an SEM approach. Vision 21 (2), 152–161.
Dunlap, R.E., Van Liere, K.D., Mertig, A.G., Jones, R.E., 2000. New trends in measuring en- Maloney, M.P., Ward, M.P., 1973. Ecology: let’s hear from the people: an objective scale
vironmental attitudes: measuring endorsement of the new ecological paradigm: a re- for the measurement of ecological attitudes and knowledge. Am. Psychol. 28 (7),
vised NEP scale. J. Soc. Issues 56 (3), 425–442. 583–594.
Eagly, A.H., Chaiken, S., 1993. The Psychology of Attitudes. Harcourt Brace Jovanovich Col- Manzanal, R.F., Tudela, M.B., 2015. Las ciencias de la naturaleza en la Educación Infantil: el
lege Publishers. ensayo, la sorpresa y los experimentos se asoman a las aulas Pirámide.
Fielding, K.S., Head, B.W., 2012. Determinants of young Australians’ environmental ac- McDonald, R.P., Ho, M.H.R., 2002. Principles and practice in reporting structural equation
tions: the role of responsibility attributions, locus of control, knowledge and atti- analyses. Psychol. Methods 7 (1), 64–83.
tudes. Environ. Educ. Res. 18 (2), 171–186. Miller, D., Merrilees, B., Coghlan, A., 2015. Sustainable urban tourism: understanding and
Flamm, B., 2009. The impacts of environmental knowledge and attitudes on vehicle own- developing visitor pro-environmental behaviors. J. Sustain. Tour. 23 (1), 26–46.
ership and use. Transp. Res. Part D: Transp. Environ. 14 (4), 272–279. Mobley, C., Vagias, W.M., DeWard, S.L., 2010. Exploring additional determinants of envi-
Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with unobservable ronmentally responsible behavior: the influence of environmental literature and en-
variables and measurement error. J. Mark. Res. 18, 39–50. vironmental attitudes. Environ. Behav. 42 (4), 420–447.
Frick, J., Kaiser, F.G., Wilson, M., 2004. Environmental knowledge and conservation behav- Moisander, J., 2007. Motivational complexity of green consumerism. Int. J. Consum. Stud.
ior: exploring prevalence and structure in a representative sample. Personal. Individ. 31 (4), 404–409.
Differ. 37 (8), 1597–1613. Morren, M., Grinstein, A., 2016. Explaining environmental behavior across borders: A
Fryxell, G.E., Lo, C.W., 2003. The influence of environmental knowledge and values on meta-analysis. J. Environ. Psychol. 47, 91–106.
managerial behaviours on behalf of the environment: an empirical examination of Mostafa, M.M., 2007. Gender differences in Egyptian consumers’ green purchase behav-
managers in China. J. Bus. Ethics 46 (1), 45–69. iour: the effects of environmental knowledge, concern and attitude. Int. J. Consum.
Gambro, J.S., Switzky, H.N., 1999. Variables associated with American high school stu- Stud. 31 (3), 220–229.
dents’ knowledge of environmental issues related to energy and pollution. Mostafa, M.M., 2009. Shades of green: a psychographic segmentation of the green
J. Environ. Educ. 30 (2), 15–22. consumer in Kuwait using self-organizing maps. Expert Syst. Appl. 36 (8),
Gifford, R., Nilsson, A., 2014. Personal and social factors that influence pro-environmental 11030–11038.
concern and behaviour: a review. Int. J. Psychol. 49 (3), 141–157. Ogbeide, O.A., Ford, C., Stringer, R., 2015. The environmental benefits of organic wine: ex-
Goldman, D., Yavetz, B., Pe’er, S., 2006. Environmental literacy in teacher training in Israel: ploring consumer willingness-to-pay premiums? J. Food Prod. Mark. 21 (5), 482–502.
environmental behavior of new students. J. Environ. Educ. 38 (1), 3–22. Olli, E., Grendstad, G., Wollebaek, D., 2001. Correlates of environmental behaviors: bring-
Guagnano, G.A., Stern, P.C., Dietz, T., 1995. Influences on attitude-behavior relationships: a ing back social context. Environ. Behav. 33 (2), 181–208.
natural experiment with curbside recycling. Environ. Behav. 27 (5), 699–718. Oreg, S., Katz-Gerro, T., 2006. Predicting proenvironmental behavior cross-nationally:
Halkos, G., Matsiori, S., 2017. Environmental attitude, motivations and values for marine Values, the theory of planned behavior, and value-belief-norm theory. Environ.
biodiversity protection. Journal of Behavioral and experimental Economics 69, 61–70. Behav. 38 (4), 462–483.
Hinds, J., Sparks, P., 2008. Engaging with the natural environment: the role of affective Paço, A., Lavrador, T., 2017. Environmental knowledge and attitudes and behaviors to-
connection and identity. J. Environ. Psychol. 28 (2), 109–120. wards energy consumption. J. Environ. Manag. 197, 384–392.
Hines, J.M., Hungerford, H.R., Tomera, A.N., 1987. Analysis and synthesis of research on re- Paço, A.D., Ferreira, J., Raposo, M., Rodrigues, R., Dinis, A., 2010. Universities’ Entrepre-
sponsible environmental behavior: a meta-analysis. J. Environ. Educ. 18 (2), 1–8. neurship Education and Regional Development: A stakeholders’ Approach.
Hoyle, R.H., Panter, A.T., 1995. Writing about structural equation models in structural Recuperado em. 5 pp. 63–75.
equation modeling. Structural Equation Modeling: Concepts, Issues, and Applications, Park, E., 2019. Social acceptance of green electricity: evidence from the structural equa-
pp. 158–176. tion modeling method. J. Clean. Prod. 215, 796–805.
Huang, X., Ge, J., 2019. Electric vehicle development in Beijing: an analysis of consumer Pickett-Baker, J., Ozaki, R., 2008. Pro-environmental products: marketing influence on
purchase intention. J. Clean. Prod. 216, 361–372. consumer purchase decision. J. Consum. Mark.
Hungerford, H.R., Volk, T.L., 1990. Changing learner behavior through environmental ed- Polonsky, M.J., Vocino, A., Grau, S.L., Garma, R., Ferdous, A.S., 2012. The impact of general
ucation. J. Environ. Educ. 21 (3), 8–21. and carbon-related environmental knowledge on attitudes and behavior of US con-
Hynes, N., Wilson, J., 2016. I do it, but don’t tell anyone! Personal values, personal and so- sumers. J. Mark. Manag. 28 (3–4), 238–263.
cial norms: can social media play a role in changing pro-environmental behaviors? Preacher, K.J., Hayes, A.F., 2008. Asymptotic and resampling strategies for assessing and
Technol. Forecast. Soc. Chang. 111, 349–359. comparing indirect effects in multiple mediator models. Behav. Res. Methods 40
Iozzi, L.A., 1989. What research says to the educator: part two: environmental education (3), 879–891.
and the affective domain. J. Environ. Educ. 20 (4), 6–13. Rajapaksa, D., Gifford, R., Torgler, B., Garcia-Valiñas, M., Athukorala, W., Managi, S.,
Jensen, B.B., 2002. Knowledge, action and pro-environmental behavior. Environ. Educ. Wilson, C., 2019. Do monetary and non-monetary incentives influence environmen-
Res. 8 (3), 325–334. tal attitudes and behavior? Evidence from an experimental analysis. Resour. Conserv.
Juvan, E., Dolnicar, S., 2017. Drivers of pro-environmental tourist behaviors are not uni- Recycl. 149, 168–176.
versal. J. Clean. Prod. 166, 879–890. Rauwald, K.S., Moore, C.F., 2002. Environmental attitudes as predictors of policy support
Kaiser, F.G., Gutscher, H., 2003. The proposition of a general version of the theory of across three countries. Environ. Behav. 34 (6), 709–739.
planned behavior: predicting ecological behavior 1. J. Appl. Soc. Psychol. 33 (3), Roberts, D.W., 1996. Landscape vegetation modelling with vital attributes and fuzzy sys-
586–603. tems theory. Ecol. Model. 90 (2), 175–184.
Kaiser, L., Briones, M.S., Hayden, F.G., 1999. Performance of virus isolation and Saratale, R.G., Saratale, G.D., Shin, H.S., Jacob, J.M., Pugazhendhi, A., Bhaisare, M., Kumar, G.,
Directigen® Flu A to detect influenza A virus in experimental human infection. 2018. New insights on the green synthesis of metallic nanoparticles using plant and
J. Clin. Virol. 14 (3), 191–197. waste biomaterials: current knowledge, their agricultural and environmental appli-
Kaiser, F.G., Roczen, N., Bogner, F.X., 2008. Competence formation in environmental educations. Environ. Sci. Pollut. Res. 25 (11), 10164–10183.
cation: advancing ecology-specific rather than general abilities. Umweltpsychologie Saripah, A.L., Yeop Hussin, B., Zainudin, A., 2013. Towards the realization of green cities:
12 (2), 56–70. The moderating role of the residents’ education level. ASEAN Conference on
Kollmuss, A., Agyeman, J., 2002. Mind the gap: why do people act environmentally and Environmental-Behavior Studies, Hanoi, Vietnam.
what are the barriers to pro-environmental behavior? Environ. Educ. Res. 8 (3), Scannell, L., Gifford, R., 2013. Personally relevant climate change: the role of place attach-
239–260. ment and local versus global message framing in engagement. Environ. Behav. 45 (1),
Kong, H.M., Ko, E., Chae, H., Mattila, P., 2016. Understanding fashion consumers’ attitude 60–85.
and behavioral intention toward sustainable fashion products: focus on sustainable Schahn, J., Holzer, E., 1990. Studies of individual environmental concern: the role of
knowledge sources and knowledge types. J. Glob. Fash. Market. 7 (2), 103–119. knowledge, gender, and background variables. Environ. Behav. 22 (6), 767–786.
Schreiber, J.B., Nora, A., Stage, F.K., Barlow, E.A., King, J., 2006. Reporting structural equa- Wan, C., Cheung, R., Qiping Shen, G., 2012. Recycling attitude and behavior in university
tion modeling and confirmatory factor analysis results: A review. J. Educ. Res. 99 campus: a case study in Hong Kong. Facilities 30 (13/14), 630–646.
(6), 323–338. Wang, B.Z., Cheng, Z., 2017. Environmental perceptions, happiness and pro-
Schultz, P.W., 2010. New environmental theories: empathizing with nature: the effects of environmental actions in China. Soc. Indic. Res. 132 (1), 357–375.
perspective taking on concern for environmental issues. J. Soc. Issues 56 (3), Wang, P., Liu, Q., Qi, Y., 2014. Factors influencing sustainable consumption behaviors: a
391–406. survey of the rural residents in China. J. Clean. Prod. 63, 152–165.
Shen, J., Saijo, T., 2008. Reexamining the relations between socio-demographic character- Weber, E.U., 2006. Experience-based and description-based perceptions of long-
istics and individual environmental concern: evidence from Shanghai data. J. Environ. term risk: why global warming does not scare us (yet). Clim. Chang. 77 (1–2),
Psychol. 28 (1), 42–50. 103–120.
Steg, L., Vlek, C., 2009. Encouraging pro-environmental behavior: an integrative review Wiek, A., Withycombe, L., Redman, C.L., 2011. Key competencies in sustainability: a
and research agenda. J. Environ. Psychol. 29 (3), 309–317. reference framework for academic program development. Sustain. Sci. 6 (2),
Stone, C.A., Sobel, M.E., 1990. The robustness of estimates of total indirect effects in co- 203–218.
variance structure models estimated by maximum. Psychometrika 55 (2), 337–352. Williams, K.J., Cary, J., 2002. Landscape preferences, ecological quality, and biodiversity
Tabernero, C., Hernández, B., 2011. Self-efficacy and intrinsic motivation guiding environ- protection. Environ. Behav. 34 (2), 257–274.
mental behavior. Environ. Behav. 43 (5), 658–675. Williams, J., MacKinnon, D.P., 2008. Resampling and distribution of the product methods
Taufique, K.M.R., Vaithianathan, S., 2018. A fresh look at understanding Green consumer for testing indirect effects in complex models. Struct. Equ. Model. Multidiscip. J. 15
behavior among young urban Indian consumers through the lens of Theory of (1), 23–51.
Planned Behavior. J. Clean. Prod. 183, 46–55. Wyles, K.J., Pahl, S., White, M., Morris, S., Cracknell, D., Thompson, R.C., 2013. Towards a
Taylor, A.B., MacKinnon, D.P., Tein, J.Y., 2008. Tests of the three-path mediated effect. marine mindset: visiting an aquarium can improve attitudes and intentions regard-
Organ. Res. Methods 11 (2), 241–269. ing marine sustainability. Visitor Studies 16 (1), 95–110.
Thompson, B., 2007. Factor analysis. The Blackwell Encyclopedia of Sociology. Xiao, C., Hong, D., 2010. Gender differences in environmental behaviors in China. Popul.
Trotta, G., 2018. Factors affecting energy-saving behaviors and energy efficiency invest- Environ. 32 (1), 88–104.
ments in British households[J]. Energy Policy 114, 529–539. Xiao, C., Hong, D., 2017. Gender differences in concerns for the environment among the
Turaga, R.M.R., Howarth, R.B., Borsuk, M.E., 2010. Pro-environmental behavior. Ann. N. Y. Chinese public: an update. Soc. Nat. Resour. 30 (6), 782–788.
Acad. Sci. 1185 (1), 211–224. Yadav, R., Pathak, G.S., 2016a. Intention to purchase organic food among young con-
Untaru, E.N., Epuran, G., Ispas, A., 2014. A conceptual framework of consumers’ pro- sumers: evidences from a developing nation. Appetite 96, 122–128.
environmental attitudes and behaviours in the tourism context. Bulletin of the Tran- Yadav, R., Pathak, G.S., 2016b. Young consumers' intention towards buying green prod-
silvania University of Brasov. Economic Sciences. Series V 7(2), pp. 85–94. ucts in a developing nation: Extending the theory of planned behavior. J. Clean.
Varela-Candamio, L., Novo-Corti, I., García-Álvarez, M.T., 2018. The importance of envi- Prod. 135, 732–739.
ronmental education in the determinants of green behavior: A meta-analysis ap- Zhang, J., Cheng, M., Wei, X., Gong, X., Zhang, S., 2019. Internet use and the satisfaction
proach. J. Clean. Prod. 170, 1565–1578. with governmental environmental protection: evidence from China. J. Clean. Prod.
Vicente-Molina, M.A., Fernández-Sáinz, A., Izagirre-Olaizola, J., 2013. Environmental 212, 1025–1035.
knowledge and other variables affecting pro-environmental behavior: comparison Zhou, H., Yin, H., Yuan, F., Wang, F., 2018. Social relationships, public media, and pro-
of university students from emerging and advanced countries. J. Clean. Prod. 61, environmental behaviors. Empir. Econ. 1–20.
130–138. Zsóka, Á., Szerényi, Z.M., Széchy, A., Kocsis, T., 2013. Greening due to environmental edu-
Vicente-Molina, M.A., Fernández-Sainz, A., Izagirre-Olaizola, J., 2018. Does gender make a cation? Environmental knowledge, attitudes, consumer behavior and everyday pro-
difference in pro-environmental behavior? The case of the Basque Country university environmental activities of Hungarian high school and university students. J. Clean.
students. J. Clean. Prod. 176, 89–98. Prod. 48, 126–138.

Liu Et Al (2020) How Does Enviormental Knowledge Translate Into Pro-Envionmental Behaviors

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Liu Et Al (2020) How Does Enviormental Knowledge Translate Into Pro-Envionmental Behaviors

Uploaded by

Copyright:

Available Formats

Science of the Total Environment 728 (2020) 138126

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

How does environmental knowledge translate into pro-environmental

• Assessing a causal chain from environ-

Fig. 1. Research model.

3.2.3. Environmental behavioral intentions 3.2.6. External context

Variable Parameter Estimation Convergent Validity Reliability

EA 0.38 0.62 Norm* b5 b0.05 N0.90 N0.95 N0.95 N0.95 N0.95

Fig. 2. Path diagram of the structural equation model.

SE Z Lower Upper Lower Upper

Note: 5000 bootstrap samples.

Item EK EA EBI PEB

Mean SD p Mean SD p Mean SD p Mean SD p

P: *p b 0.05, **p b 0.01, ***p b 0.001.

with past research, environmental attitudes were found to be a strong

Constructs Items content

Unconstrained vs Measurement weights 10 8.226 0.607

P: *p b 0.05, **p b0.01, ***p b 0.001.

Hypothesis Path Male Female Z

H1 PEB b ---EK 0.120 −0.044 −1.165

Model comparison Df diff CMIN diff P-value

Unconstrained vs Measurement weights 10 15.74 0.107

P: *p b 0.05, **p b 0.01, ***p b 0.001.

Model comparison Df diff CMIN diff P-value

Eastern China vs Central China vs Western China

(continued on next page)

Model comparison Df diff CMIN diff P-value

Unconstrained vs Measurement weights 20 51.616 0.134

Eastern China vs Central China

Central China vs Western China

Eastern China vs Western China

P: *p b 0.05, **p b 0.01, ***p b 0.001.

Hypothesis Path E and C W Z

H1 PEB b ---EK 0.085 0.248 0.796

.Model comparison Df diff CMIN diff P-value

Unconstrained vs Measurement weights 20 35.370 0.18

P: *p b 0.05, **p b 0.01, ***p b 0.001.

You might also like

P: p b 0.05, p b 0.01, p b 0.001.

P: p b 0.05, p b0.01, p b 0.001.

P: p b 0.05, p b 0.01, p b 0.001.

P: p b 0.05, p b 0.01, p b 0.001.

P: p b 0.05, p b 0.01, p b 0.001.