Journal of Cleaner Production 61 (2013) 130e138

Contents lists available at SciVerse ScienceDirect

Journal of Cleaner Production

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

Environmental knowledge and other variables affecting

pro-environmental behaviour: comparison of university students
from emerging and advanced countries
María Azucena Vicente-Molina*, Ana Fernández-Sáinz, Julen Izagirre-Olaizola
University of the Basque Country (UPV/EHU), Avenida Lehendakari Agirre, 83, E48015 Bilbao, Spain

a r t i c l e i n f o a b s t r a c t

Article history: This paper analyses the influence of environmental knowledge on pro-environmental behaviour among
Received 20 July 2012 university students from countries with different levels of economic development (USA, Spain, Mexico
Received in revised form and Brazil). The explanatory variables include formal and informal education sources, gender, motiva-
27 April 2013
tions, attitudes and perceived effectiveness of pro-environmental behaviour. Differences are found be-
Accepted 2 May 2013
tween students from emerging and developed countries which suggest that external factors (culture,
Available online 20 May 2013
environmental structures and services in each country) might play a relevant role in university students’
behaviour towards the environment. A multinomial ordered logit model is applied to estimate the in-
JEL classifications:
Q57
fluence of the covariates on the environmental performance probability. The results also suggest that
R58 motivation and perceived effectiveness are not only significant variables in both groups but also the most
important ones in explaining pro-environmental behaviour. While knowledge (objective and subjective)
Keywords: influences pro-environmental behaviour, attitude and informal education are not relevant variables.
Environmental knowledge
Ó 2013 Elsevier Ltd. All rights reserved.
University students
Pro-environmental behaviour
Cross-country comparison

1. Introduction the education system (García-Valiñas et al., 2010). Therefore, un-

derstanding how education impacts on environmental knowledge
Education is one of the most important variables in explaining and thus on the development of pro-environmental behaviour is an
high levels of environmental concern and behaviour (Zilahy and important issue for policy makers, marketers, green businesses,
Huisingh, 2009; Zsóka et al., 2012). Researchers suggest that educators and other parties interested in the acceptance and
more highly educated individuals are more concerned about enhancement of pro-environmental behaviour.
environmental quality and are more motivated to engage in envi- Younger generations will be affected by environmental prob-
ronmentally responsible behaviour since they are better aware of lems arising from present actions, so they need to be provided with
the potential damage (Lozano, 2006; Olli et al., 2001). More highly accurate environmental knowledge and skills to develop sustain-
educated individuals seem to possess a higher level of environ- able solutions (Adomssent et al., 2007; Bradley et al., 1999; Oguz
mental knowledge, which is translated into pro-environmental et al., 2010). Thus, higher education has a pivotal role in impelling
behaviour (Schlegelmilch et al., 1996). pro-environmental behaviour and solutions, since it aims to raise
Although education and environmental knowledge seem to be responsible, competent individuals with knowledge, skills and
significantly and directly related, it is not clear what sort of rela- values that will contribute to an environmentally sustainable,
tionship exists between them or how it affects pro-environmental improving world (Adomssent, 2013; Corcoran and Wals, 2004;
behaviour (Zsóka et al., 2012). General knowledge and even specific Lozano et al., 2013).
skills related to environmental issues are often acquired through In recent years, many universities have become engaged in
sustainable development, although the trend has not yet fully
permeated to all disciplines, academics and university leaders
* Corresponding author. Tel.: þ34 946013867.
(Lozano et al., 2013). Considering that universities actually train
E-mail addresses: azucena.vicente@ehu.es (M.A. Vicente-Molina),
ana.fernandez@ehu.es (A. Fernández-Sáinz), julen.izaguirre@ehu.es (J. Izagirre- people to perform important social roles effectively (Frank and
Olaizola). Meyer, 2007), this paper focuses on university students since they

0959-6526/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jclepro.2013.05.015
 M.A. Vicente-Molina et al. / Journal of Cleaner Production 61 (2013) 130e138 131

are the leaders, policy makers, scientists, consumers, researchers, attention is paid to different environmental knowledge measures
and entrepreneurs of the future, and as such the future decision- (objective and subjective). The type of studies/degree (science,
makers1 in different areas (political, social, environmental, eco- engineering or social sciences) is also analysed since previous re-
nomic, etc.) (Lozano, 2006; Waas et al., 2010; Zilahy and Huisingh, sults suggest that there may be differences in pro-environmental
2009). If future talent is able to make decisions that are beneficial to behaviour due to this factor (Talay et al., 2004). This variable and
the environment, society is more likely to make progress along the the number of subjects on environmental topics are considered
path towards sustainability. In this sense, the level of environ- together as an indicator of formal education. Other informal sour-
mental knowledge and the role of environmental education in ces of knowledge (television, newspapers, family, friends, speci-
changing and addressing lifestyles and attitudes could be crucial in alised books and journals, etc.), the reliability of that knowledge
altering individuals’ behaviour and in turning society towards source and its influence on pro-environmental behaviour are also
sustainability (Adomssent, 2013; Bradley et al., 1999; Szerényi et al., taken into account in this work. These variables are included in an
2009). In that change, universities must understand and satisfy the attempt to determine whether they are predictors of pro-
needs of present and future generations, establishing themselves as environmental behaviour for this group.
leaders of change towards sustainability (Lozano et al., 2013). It is The second objective is to provide evidence about different
also essential to find out whether future leaders are ready to direct types of socio-demographic and psychological variables, which
their country towards environmentally sustainable development or according to the relevant literature seem to influence pro-
whether educational and political institutions need to introduce environmental behaviour.
changes to promote higher levels of environmental knowledge and The third objective is to find what similarities and differences
behaviour at universities. exist between the pro-environmental behaviour of students from
Although there are many studies focused on school students and emerging and developed countries.
environmental topics (Asmuni et al., 2012; Bradley et al., 1999), The rest of the paper is structured as follows: Section 2 starts
there are very few that analyse university students as a group per se. with a literature review, presenting a summary on research find-
Some of these studies focus on identifying the influence of a few ings regarding variables such as environmental knowledge, moti-
variables on the behavioural intentions of business students vations, attitudes, perceived effectiveness and gender, among
(Cordano et al., 2010), the environmental beliefs of students others. Section 3 presents the methodology for obtaining data and
majoring in different areas (Ewert and Baker, 2001), the level of explanatory variables, explains the theoretical model and its
general environmental concern of students (Oguz et al., 2010) or econometric formulation. Section 4 presents the multinomial or-
simply use university students to define a general consumer profile dered logit model and discusses the results produced by the model.
(Synodinos, 1990). However, we are unaware of any study focused The main conclusions for supporting recommendations for uni-
on the environmental knowledge of university students and its versity institutions and educational policy makers are set out in
possible links to a set of environmental behaviour variables (recy- Section 5. Finally, the major limitations of the work are explained in
cling of different types of waste, public transport use and green this section.
purchasing). Moreover, much of the research on environmental
behaviour originates from the United States (Cordano et al., 2010; 2. Background and context
Diamantopoulos et al., 2003). Although there has recently been
some expansion across different countries, little is known about There has been a great deal of research into predictors of pro-
pro-environmental behaviour outside the United States (Cordano environmental behaviour. However, despite the plethora of
et al., 2010) and even less about university students and possible studies on pro-environmental behaviour there is still disagreement
differences between those in emerging and advanced countries regarding how pro-environmental behaviour can be predicted from
(International Monetary Fund, 2012). attitudes and other variables (Bamberg and Moser, 2007; Kollmuss
The aim of this paper is to identify the main factors that influ- and Agyeman, 2002; Mobley et al., 2010).
ence the pro-environmental behaviour of university students from This section presents the main results of a literature review on
two emerging countries, Mexico and Brazil, and two advanced variables among individuals (knowledge, attitudes, motivations
countries, the United States and Spain (International Monetary and perceived effectiveness of pro-environmental behaviour).
Fund, 2012) in order to propose a multinomial ordered logit Gender is also studied because it is considered as potentially
model to define students’ pro-environmental behaviour. This affecting both environmental knowledge and behaviour. Other
model will also enable measurements to be taken of how a relative variables such as type of studies/degree, informal education sour-
increase/reduction in a significant variable might affect the pro- ces (television, newspapers, family, etc.), price sensitivity and the
environmental behaviour of university students from these coun- environmental structures and services of each country are also
tries. This main aim can be divided into three objectives: reviewed.
First, given that people with more education are more likely to
develop environmental performance, this paper explores factors 2.1. Environmental knowledge
related to education level (knowledge of the environment) that
may influence pro-environmental behaviour. In this regard, special Environmental knowledge can be defined as one’s ability to
identify a number of symbols, concepts and behaviour patterns
related to environmental protection (Laroche et al., 2001). In the
1
earliest linear regression models, knowledge was defined as a
It is assumed that education is the key to knowledge, especially in the current
source from which environmental attitudes were formed and
knowledge society, and knowledge open the door to elite positions. “Virtually all
elite occupations globally are certified by the university, and nearly all the world’s behaviour manifested (Kollmuss and Agyeman, 2002). It has been
stratification systems are legitimated by university-based knowledge” (Frank and demonstrated that those models were wrong, and it is necessary to
Meyer, 2007). Of course, university students are not the only people who can ac- consider variables explaining pro-environmental behaviour for
cess to top positions and make decisions for a sustainable world. Nevertheless, they future research (Kollmuss and Agyeman, 2002). Although theo-
are more likely to become decision makers and leaders of sustainability, since at
university they acquire the technical and specialised knowledge necessary to make
retical knowledge seems to play a significant role in pro-
important decisions and to carry out legal, social, technological, etc. innovations for environmental behaviour, the empirical evidence is not so clear
a more sustainable world. (Kaiser and Fuhrer, 2003; Laroche et al., 2001; Zsóka et al., 2012).
132 M.A. Vicente-Molina et al. / Journal of Cleaner Production 61 (2013) 130e138

Some studies find no significant relationship between environ- environmental behaviour (Kollmuss and Agyeman, 2002; Olli
mental knowledge and pro-environmental behaviour (Bartiaux, et al., 2001). The empirical evidence suggests that attitude alone
2008; Laroche et al., 2001; Maloney and Ward, 1973). Other is a poor predictor of intentional pro-environmental behaviour
studies reveal that a deeper knowledge of environmental issues (Vermeir and Verbeke, 2006). Some relevant studies conclude that
and how to solve them increases the likelihood of individual taking each behaviour pattern has its own cluster of predictors, pointing
actions to protect the environment (Kaiser and Fuhrer, 2003; out that general attitudes are bad predictors for specific environ-
Kollmuss and Agyeman, 2002; Mobley et al., 2010). People who mental behaviour patterns (Balderjahn, 1988; Hines et al., 1986e
have greater knowledge of environmental problems are more 1987). Maineiri et al. (1997) explain this in terms of four reasons:
prone to behave in a pro-environmental way, ceteris paribus (Oguz (1) weak correlations between environmental behaviour patterns
et al., 2010). (not all patterns are interchangeable); (2) different levels of spec-
A shortage of knowledge or the holding of contradictory infor- ificity in the measurement of attitude and behaviour; (3) effects of
mation might limit pro-environmental behaviour. The study con- other variables; and (4) lack of measure of reliability and validity.
ducted by Kennedy et al. (2009) in Canada found that more than
60% of respondents felt that their pro-environmental behaviour 2.3. Motivations
was often constrained by a perceived lack of knowledge. Other
scholars have suggested that a lack of appropriate knowledge or an Motivation can be defined as a strong internal stimulus around
excess of self-perceived knowledge might impel individuals to which behaviour is organised (Wilkie, 1990). Motivation is there-
make environmentally wrong decisions. Thus, if such individuals fore usually understood as a reason for behaviour (Moisander,
are more aware of environment problems and their causes, they 2007). It is shaped by two components: intensity and direction.
will become more motivated to act towards the environment in Both determine the behaviour that is chosen and why. Stern et al.
more responsible ways (Barber et al., 2009). It has recently been (1993) claim that the stronger the egoistic orientation of the indi-
assumed that knowledge is a necessary but not sufficient condition vidual, the stronger the motivation for pro-environmental behav-
for pro-environmental behaviour by an individual (Kaiser and iour is. However, this is contradicted by other studies, which
Fuhrer, 2003; Kollmuss and Agyeman, 2002) since cultural factors suggest that pro-environmental behaviour is positively related to
could act as barriers (Kennedy et al., 2009; Lozano, 2006). unselfish or altruistic values (Hopper and Nielsen, 1991; Thøgersen,
Two main approaches have been used to analyse the environ- 2011).
mental knowledge of individuals: objective and subjective knowl- In this regard, a key motivation for consumers is the importance
edge (Barber et al., 2009). given to price. As a result, price-sensitive people can be prevented
Objective knowledge (actual knowledge) refers to how much a from buying green products (Fotopoulos and Krystallis, 2002; Soyez
person actually knows about a type of product, issue or object. et al., 2011). The Canadian study by Kennedy et al. (2009) reveals
Subjective knowledge (also called perceived knowledge) shows that more than 48% of respondents feel that lack of money is an
how much a person thinks that he/she knows (Dodd et al., 2005). important constraint for green purchasing. Therefore, those in-
Most of the scales for measuring (environmental) knowledge have dividuals that show a high sensitivity to price are expected to be
been developed solely in accordance with subjective (self-rated) less supportive of pro-environmental purchase behaviour, with
knowledge (Amyx et al., 1994; Schlegelmilch et al., 1996) or price acting as a barrier.
objective (actual) knowledge (Bartiaux, 2008; Bradley et al., 1999;
DiEnno and Hilton, 2005; Laroche et al., 2001; Maloney and Ward, 2.4. Perceived consumer effectiveness (PCE)
1973). This might be the source of the contradictory results found
(Laroche et al., 2001; Schlegelmilch et al., 1996). Some individuals may believe that their own efforts will
Martin and Simintiras (1995) measure both objective and sub- significantly mitigate environmental problems, whereas others
jective knowledge of environmental issues. Their study reveals that may think that their efforts will not make any difference.
the ability of consumers to answer questions on environmental PCE was first measured as an element of personality variables to
issues correctly does not correlate with subjective environmental predict environmental concern (Kinnear et al., 1974). Then
knowledge and purchase intention. Additionally, Ellen (1994) finds Straughan and Roberts (1999) measured it as one of the attitudinal
no significant relationship between either objective or subjective variables in predicting behaviour and it was found to be a better
knowledge and recycling-based shopping decisions. Hence, it is predictor than environmental concern. Thus, attitude and PCE are
essential to identify the types of knowledge that encourage envi- currently modelled as two distinct constructs in environmental
ronmental behaviour effectively if the most efficient informational studies (Ellen et al., 1991).
strategies are to be promoted in education (Frick et al., 2004). In general, the findings of empirical research suggest that PCE is
positively related to pro-environmental attitudes (Kim and Choi,
2.2. Attitudes 2003; Laskova, 2007). Some researchers argue that if an individ-
ual believes that an environmental problem can be solved by a
Milfont and Duckitt (2010) define environmental attitude as a specific activity, then that belief should strongly influence the in-
psychological tendency expressed by evaluating the natural envi- dividual’s willingness to engage in that specific activity but not in
ronment with some degree of favour or disfavour. However other another (Ellen et al., 1991). Therefore, PCE might be useful for
authors refer to environmental attitudes as environmental concern predicting general pro-environmental behaviour.
(Dunlap and Jones, 2002).
The empirical findings regarding the relationship between at- 2.5. Gender
titudes and environmental behaviour are contradictory. Some
studies have reported a positive relationship between environ- Previous research has examined gender differences in a variety
mental attitude and pro-environmental behaviour (Kim and of environmental and behavioural variables. In the 90’s most
Choi, 2003; Kollmuss and Agyeman, 2002; Schlegelmilch et al., studies found that women participated more in pro-environmental
1996; Straughan and Roberts, 1999; Tilikidou, 2007), while others behaviour (Davidson and Freudenburg, 1996; Zelezny et al., 2000).
find a negative relationship (Cottrell, 2003). Furthermore, other Zelezny et al. (2000) study gender differences across 14 countries
studies reveal a weak relationship between attitudes and pro- and find significant gender differences in environmental attitudes
 M.A. Vicente-Molina et al. / Journal of Cleaner Production 61 (2013) 130e138 133

and behaviour patterns within countries, with women being cultural influences might explain differences in the characteristics
consistently more pro-environmental than men. However, of a particular country’s pro-environmental consumers. According
Davidson and Freudenburg (1996) claim that gender differences in to Cordano et al. (2010) pro-environmental behaviour seems to be
environmentalism are not universal. Most authors conclude that influenced by environmental beliefs, which differ from one culture
men have more knowledge about environmental issues than to another. Environmental problems are not perceived in the same
women (Diamantopoulos et al., 2003). These differences may be way in all countries. In advanced economies the perceived impor-
due to the different socialisation patterns of boys and girls (Schahn tance of environmental issues has increased in the past few de-
and Holzer, 1990). cades. In emerging economies the situation is different. Countries
such as Brazil and Mexico are still undergoing rapid industrialisa-
2.6. Academic discipline and number of environment-related tion and environmental standards (and people’s attitudes) are not
subjects studied able to cope with the pressure that is generated by rapid growth
(Wang et al., 2011).
Sometimes the type of discipline or the environment-related As mentioned above, country specific factors are included in this
subjects studied might affect an individual’s knowledge of the study by comparing emerging countries to developed ones.
environment and affect the relationship between attitudes, beliefs
and behaviour (Guagnano et al., 1995). A study of university stu- 3. Methodology: data collection and econometric model
dents from 14 different faculties in Ankara (Turkey) concluded that choice
most of those who were studying environment-related subjects
were from the social sciences (Talay et al., 2004). Therefore, it Data were collected using a structured questionnaire. The
makes sense to analyse the possible influence of social science measuring procedures established in the relevant literature
disciplines on individual pro-environmental behaviour. The num- (Tilikidou, 2007) were followed in developing the scales. From
ber of environment-related subjects is also analysed, since little November 2009 to May 2011 the questionnaire was given out at
attention has been paid to this issue in the relevant literature four publicly-run universities: the University of the Basque Country
(Ewert and Baker, 2001). (Spain), the University of Reno-Nevada (USA), UNAM (Mexico) and
Salvador do Bahia University (Brazil). A sample of 2226 students
was obtained.
2.7. Informal education
The questionnaire was divided into blocks. The first part
measured objective/subjective knowledge. Objective knowledge
People can learn more about environmental issues through the
was tested via ten questions on current environmental issues and
media (watching television or reading newspapers), the Internet or
behaviour patterns with three possible answers (true/false/no
social interaction (family, friends).2 The literature on the matter
answer). Subjective environmental knowledge was measured using
widely recognises the importance of non-formal education chan-
two single self-report items (before and after answering the
nels in generating environmental responsibility (Chan, 1998). The
objective knowledge test). The second part of the questionnaire
media can thus influence environmental performance (Thamwipat
comprised several items intended to measure attitudes, motiva-
et al., 2012), while the effect of socialisation factors depends on the
tions and perceived effectiveness in regard to a variety of topics
type of environmental behaviour that is measured (Chan, 1998).
related to the environment. Respondents were asked to specify on a
Moreover social norms, which are generated within individuals
five-point Likert scale whether they agreed or disagreed with each
through informal education and the media, seem to influence pro-
one. They were then asked to indicate their two main sources of
environmental behaviour through attitudes and behavioural in-
informal environmental information (television, radio, newspa-
tentions (Bamberg and Moser, 2007), so informal environmental
pers, specialised journals and books, family, membership of groups
education seems to be essential for individuals to acquire envi-
and others) and their level of credibility.
ronmentally favourable attitudes (Ballantyne et al., 1998).
In another section, respondents were asked to answer how often
they recycled, used public transport for environmental protection
2.8. Country-specific factors: access to environmental structures and purchased green products. Variables such as environmental
and cultural differences courses attended, type of discipline/degree, price sensitivity, etc.,
were also included in other parts of the questionnaire. The final
The quality and availability of community environmental ser- block included socio-demographic variables such as gender, age,
vices play a significant role in determining individuals’ participa- income and employment status, although only gender is analysed
tion in pro-environmental behaviour (Kennedy et al., 2009). Thus, here.
in areas where there are accessible structures, pro-environmental Ordered logit models are a kind of discrete choice models which
behaviour will arise much more easily than in areas without have been used in various economic applications including choice
structural services (Derksen and Gartrell, 1993). For example, of vehicle type (Choo and Mokhtarian, 2004), consumer brand
people may be motivated to buy green products but if those (Paap and Franses, 2000) and recycling preferences (Nixon et al.,
products are not offered for sale in an accessible location they will 2009).
not actually buy them. Moreover, the lack of easy access to recycling These logit models are used to measure the effect of each
bins might prevent people from recycling and the absence of public explanatory variable on the probability of specific behaviour choice.
transport may require them to use more polluting transportation. Additionally, they can indicate the sensitivity with which the
Diamantopoulos et al. (2003) argue that country-specific factors dependent variable responds to changes in the explanatory
such as availability of green products, environmental legislation or variables.
In these models a set of individuals (i ¼ 1.N) are faced with a
range of mutually exclusive alternatives, i.e. they have to make a
2
It must be stressed that in this paper informal education is considered as ed- choice (j) from a given set of options (A). In the ordered logit model
ucation that does not lead directly to a degree. However, the increasingly important
role that informal sources of knowledge play within formal education cannot be
the ordinal response variable, Yi, is the discrete realisation of an
ignored. For further discussion about different knowledge types, see Adomssent underlying, unobservable (latent) continuous random variable, Yi* .
(2013). The categories are envisaged as contiguous intervals on the
134 M.A. Vicente-Molina et al. / Journal of Cleaner Production 61 (2013) 130e138

continuous scale. The unobservable Yi* is assumed to satisfy a linear Consequently, in the marginal effects neither the sign nor the
0
regression model, i.e. Yi* ¼ b xi þ εi , where xi is a covariate vector, b magnitude of the coefficient is informative about the result above,
a vector of regression coefficients and εi the error term. The so direct interpretation of the coefficients is fundamentally
observable categorical variable, Yi, is assumed to arise from Yi* as ambiguous (Greene and Hensher, 2010). The implication of these
follows: findings is that the effect of a change in one of the variables in the
8 model depends on all model parameters, on data and on the
>
< 0; if Yi*  0 probability (cell) which is of interest.
Yi ¼ 1; if 0 < Yi*  m1 This paper analyses the highest probability of pro-
>
: 2; environmental behaviour.
if m1 < Yi*
In this type of model, the only certainties in the signs of the
partial effects are the following:
where the m1 is an unknown cut-off point (category boundary) in
the distribution of Yi* . In this paper environmental performance, Yi,
 If there is a considered variable with a positive coefficient, in-
is an ordinal variable with three categories: low level (0), medium
creases in that variable will raise the probability in the highest
level (1) and high level (2) behaviour. In principle, the individuals
cell (pro-environmental behaviour) and decrease the proba-
can respond to the questionnaire with their own Yi* if asked to do
bility in the lowest cell (non-environmental behaviour), and
so. Given three possible answers, they can choose the option that
vice versa.
most closely represents their own feelings on the question.
 The sum of all the changes will be zero, since the new proba-
Therefore, the different probabilities of the categories are:
bilities must still add up to one.
 0 
PðYi ¼ 0Þ ¼ F b xi ;
4. Results and discussions
  0 
0 
PðYi ¼ 1Þ ¼ F m1  b xi  F b xi Table 1 shows the mean and standard deviation of the variables
considered for the four countries that make up the total sample,
 0 
PðYi ¼ 2Þ ¼ 1  F m1  b xi and briefly explains how they are measured.
An exploratory factor analysis was conducted with the items
where F is the cumulative distribution function of the logistic related to perception of the need for more information, motiva-
distribution. tions, attitudes, consumer perceived effectiveness and price
Interpretation of the coefficients in ordered logit models is more sensitivity to evaluate the ability of the items to measure each
complicated than in the ordinary regression setting (Daykin and construct. Validity and reliability tests were conducted on each
Moffatt, 2002), since there is no natural conditional mean func- factor with satisfactory results, Cronbach alphas varied from 0.4 to
tion in the model. The outcome variable, Yi, is merely a label for the 0.7 (Cronbach and Shavelson, 2004; Schmitt, 1996).
ordered, non-quantitative outcomes. The results of the exploratory factor analysis reveal that Bra-
As usual, the marginal effects of the regressors, xi, on the zilian students show higher levels of altruistic motivation, Spanish
probabilities are not equal to the coefficients. For the probabilistic students have the most favourable attitudes towards the environ-
model used, the marginal effects of changes in the regressors are: ment and students from these two countries have the highest levels
of PCE. Finally, American college students allocate most importance
vPðYi ¼ 0Þ  0  to price in their consumer activities.
¼ f b xi bm The results show similar numbers of men and women in the
vxim
whole sample (with more women in the case of Spain).
In general, the results reveal that there is little knowledge
vPðYi ¼ 1Þ   0   0 
¼ f m1  b xi  f b xi bm of environmental topics, as suggested in previous research
vxim (Schlegelmilch et al., 1996; Kollmus and Agyeman, 2002). Brazilian
students obtain the highest score (5.1 out of 10) in objective
vPðYi ¼ 2Þ  0  knowledge, while Mexicans get the lowest (3.65). In all cases,
¼ f m1  b xi bm
vxim pretest and posttest subjective knowledge is higher than objective
knowledge. Brazil also stands out for the number of subjects related
where f is the density function of the logistic distribution. to environmental issues studied by university students.

Table 1
Statistics of the explanatory variables: Mean and standard deviation in brackets.

Variable Description All Spain Brazil Mexico USA

N Number of observations 2226 640 347 642 597

Gender Dummy, 1 if male 0.4995 (0.500) 0.4203 (0.494) 0.5216 (0.500) 0.5561 (0.497) 0.5109 (0.500)
Objective knowledge In the test, sum of correct answers 4.1801 (1.693) 4.2515 (1.740) 5.1008 (1.380) 3.6511 (1.546) 4.1373 (1.726)
Subjective knowledge pre-test Pre-test subjective knowledge 5.9615 (1.779) 5.6151 (1.673) 6.8912 (1.272) 6.2121 (1.727) 5.5273 (1.946)
Subjective knowledge post-test Post-test subjective knowledge 6.1790 (2.157) 5.898 (2.100) 6.7446 (1.829) 6.1403 (2.008) 6.1692 (2.476)
Type of degree/formal education Dummy, 1 if science and/or 0.3800 (0.485) 0.4625 (0.499) 0.6080 (0.489) 0.2212 (0.415) 0.3299 (0.471)
engineering
Number of subjects Number of subjects 0.6488 (1.358) 0.3744 (0.908) 1.629 (2.276) 0.2898 (0.781) 0.7441 (1.224)
Perception of lack of information Factor by factor analysis 3.475 (1.093) 3.1604 (1.183) 4.1220 (1.087) 3.8122 (1.070) 3.0705 (0.643)
Motivation (altruism) Factor by factor analysis 3.1733 (0.754) 2.9826 (0.687) 3.5512 (0.760) 3.1494 (0.733) 3.1833 (0.761)
Attitudes Factor by factor analysis 2.2669 (0.812) 2.5149 (0.732) 1.9903 (0.837) 2.2983 (0.818) 2.1274 (0.794)
Perceived consumer effectiveness Factor by factor analysis 4.005 (0.719) 3.7791 (0.686) 4.1981 (0.678) 4.2456 (0.679) 3.8809 (0.712)
(PCE)
Price sensitivity Factor by factor analysis 3.5604 (1.147) 3.4742 (1.136) 3.1072 (1.197) 3.4319 (1.100) 4.0519 (1.006)
 M.A. Vicente-Molina et al. / Journal of Cleaner Production 61 (2013) 130e138 135

A predominance of social science students can also be observed, 1) In the case of the United States and Spain, a relative increase in
except in the case of Brazil. objective knowledge (correct answers in the test) increases the
Table 2 shows the distribution of the dependent variable (IND), probability3 of high environmental performance. The same
the components that comprise (I1, I2, I3) and the indicators in the applies to pretest subjective knowledge, perceived lack of in-
different sub-samples. The dependent variable e environmental formation, motivation and PCE. On the other hand, price
performance e is constructed using three components: recycling, sensitivity is a major barrier to pro-environmental behaviour.
public transport use and green purchasing. All individuals are
placed in one of 3 classes: low (0), medium (1) or high (2) level for Surprisingly, an increase in the number of study subjects con-
each component. Depending on these values, individuals are cerned with environmental issues slightly reduces the probability
considered to have a low (0), medium (1) or high (2) level of pro- of high environmental performance. Science and engineering stu-
environmental behaviour. dents are more likely to behave pro-environmentally than social
Students in the Spanish subsample recycle most, while Mexi- science students. The results also reveal that being male decreases
cans are the most likely to use public transport as a way to protect the likelihood of high environmental behaviour.
the environment. USA students top the list in green purchasing
behaviour. In general, Spanish university students have the most 2) In the Brazil and Mexico subsample, the direction of the effects
advanced environmental performance, followed by Mexican stu- of pretest subjective knowledge, motivation and PCE are also
dents, with US and Brazilians showing lower levels. This suggests positive. However, perceived lack of information, type of de-
that structural and cultural differences across countries might be gree and number of subjects studied are no longer relevant,
influencing pro-environmental behaviour. It is clear that the lack of while the effect of objective knowledge, attitudes, gender and
easy access to recycling bins might prevent people from recycling sensitivity to price become negative. This last finding might be
and the absence of public transport may require them to use more due to the fact that the subsample results from joining two
polluting transportation. The behaviour patterns analysed (recy- emerging countries, and thus covers very different objective
cling, public transport use and green purchasing) seem to be knowledge levels. By contrast, post-test subjective knowledge
influenced by other factors (cultural and/or infrastructural) which has a significant positive effect, so an increase in the responses
are different in each considered country. This suggests that struc- that an individual believes that he/she has got right increases
tural and cultural differences across countries might be influencing the likelihood of high environmental behaviour.
pro-environmental behaviour, which are different in each country
(Diamantopoulos et al., 2003; Cordano et al., 2010). Environmental On the other hand, the model estimate reveals that PCE and
problems are not perceived in the same way in all countries. motivation show the highest partial effects (0.05432/0.03541 and
0.09818/0.07948 respectively) for the most pro-environmental
4.1. Pro-environmental behaviour ordered logit model behaviour in both subsamples. This suggests that the most impor-
tant effects in all the countries considered are motivation and PCE.
Table 3 shows the results of the model estimates (showing
positive or negative direction and the effect of each explanatory 5. Conclusions, recommendations and limitations
variable on the dependent variable) and the marginal effects. The
four countries are grouped in pairs in line with the IMF classifica- While it would be desirable to be able to demonstrate a clear,
tion (2012) and the objectives of this study, taking into account coherent causal link between education and pro-environmental
advanced economies (USA and Spain) and emerging countries behaviour, such a link, as previous literature suggest (Zsóka et al.,
(Brazil and Mexico). 2012), is difficult to be established reliably.
Although informal education was one of the variables to be Formal education and knowledge of environmental issues
analysed in the logit model, it seemed to provide little relevant clearly influence pro-environmental behaviour, but they do so in a
information and is therefore finally not included. complex way. Objective and subjective knowledge influence the
The results show that most of the variables included in the environmental performance of students from both advanced and
model have a significant effect on the dependent variable, and the emerging countries, while the type of degree impacts only in the
effects are generally consistent with findings reported in previous case of advanced countries.
research. Nevertheless, there are some significant differences be- Pretest subjective knowledge is the most relevant of all the
tween the two subsamples: knowledge-related factors analysed, i.e. what college students
think they know is more relevant than what they actually know.
Table 2
Distribution and composition of dependent variables. This implies that students might be making wrong decisions about
their environmental performance. The prevalence of this variable
Variable Description All Spain Brazil Mexico USA
might partially explain the contradictory results found in objective
Dependent variable: Low 0.1181 0.0469 0.2017 0.0841 0.1826 knowledge and environment-related subjects studied. In this sense,
IND Medium 0.6986 0.6859 0.6513 0.7352 0.7002
it may be desirable for students to become familiar with current
High 0.1833 0.2672 0.1470 0.1807 0.1173
Mean 1.0651 1.2203 0.9452 1.0966 0.9347
environmental issues, rather than developing technical environ-
I1: Recycling Low 0.2421 0.0594 0.4986 0.3178 0.2077 mental knowledge.
Medium 0.4003 0.1688 0.3775 0.5218 0.5310 The results also reveal that psychological and gender variables
High 0.3576 0.7719 0.1239 0.1604 0.2613 are important in explaining pro-environmental behaviour. Thus,
Mean 1.1154 1.7125 0.6253 0.8427 1.0536
motivation and the PCE are the factors that best explain the prob-
I2: Public transport Low 0.2727 0.1234 0.2738 0.1044 0.6131
use Medium 0.2659 0.3219 0.1787 0.1978 0.3300 ability of high environmental performance in all countries. Pro-
High 0.4614 0.5547 0.5476 0.6978 0.0570 environmental behaviour is basically linked to altruistic motiva-
Mean 1.1887 1.4312 1.2738 1.5935 0.4439 tions. Overall, it can be said that individuals who perceive lower
I3: Green purchasing Low 0.3935 0.5891 0.3919 0.4050 0.1725
Medium 0.2947 0.2188 0.3084 0.3660 0.2915
High 0.3118 0.1922 0.2997 0.2290 0.5360
3
Mean 0.9182 0.6031 0.9078 0.8239 1.3635 In this case the probability of medium and low environmental performance
decreases since the sum is constant.
136 M.A. Vicente-Molina et al. / Journal of Cleaner Production 61 (2013) 130e138

Table 3
Ordered logit model for pro-environmental behaviour.

Estimations Marginal effects

USAeSpain BrazileMexico USAeSpain BrazileMexico

Coefficients e Coefficients e Effect Elasticity Effect Elasticity

index function index function (pr ¼ 2)a (pr ¼ 2) (pr ¼ 2) (pr ¼ 2)

Constant 0.94699**b 0.93899**

Objective knowledge (test result) 0.06219** 0.06210** 0.01520 0.40528 -0.01450 0.40489
Pretest subjective knowledge 0.09331*** 0.05660** 0.02280 0.79013 0.01322 0.56789
Postest subjective knowledge 0.00649 0.04056* 0.00159 -0.05936 0.00947 0.40033
Gender (male) 0.15886** 0.15224* 0.03864 0.23992 0.03591 0.23877
Type of Studies (science and engineering) 0.19581** 0.04987 0.04865 0.30206 0.01157 0.07691
Number of subjects 0.07478** 0.03096 0.01827 0.06680 0.00723 0.03613
Perceived lack of information 0.08256* 0.06578 0.02017 0.39130 0.01536 0.40113
Motivation (altruistic) 0.40179*** 0.34028*** 0.09818 1.87812 0.07948 1.73030
Attitude 0.02804 0.04532 0.00685 0.09860 0.01059 0.15304
Perceived effectiveness 0.22230*** 0.15161** 0.05432 1.28998 0.03541 0.99494
Price sensitivity 0.19825*** 0.07310* -0.04845 1.13576 -0.01707 0.37754

Notes:
a
pr ¼ 2 is the probability of the highest pro-environmental behaviour.
b
*** e Significant at P ¼ 0.01, ** e Significant at P ¼ 0.05, * e Significant at P ¼ 0.10.

costs (selfish point of view) and higher profits (the perception that behaviour. Firms intending to develop green marketing activities
positive results can be achieved for society) are more likely to should strive to get their potential consumers to perceive specific
develop high environmental performance. benefits arising from this kind of behaviour. Additionally, including
In addition, the results confirm that gender influences envi- or increasing the supply of green products in conventional retailing
ronmental behaviour. Women are more likely to carry out envi- (hypermarkets and supermarkets) to facilitate environmentally-
ronmentally friendly activities in both advanced and emerging friendly purchases might be an interesting option in those coun-
countries. tries where such purchases are lower, e.g. Spain and Brazil.
Attitude is not a significant variable in explaining pro- Data show that students with greener behaviour have altruistic
environmental behaviour. This is not surprising, since empirical motivations in going green. Nevertheless, as Rex and Baumann
evidence states that attitude is not a good predictor of pro- (2007) argue, to achieve greener production and consumption
environmental behaviour (Vermeir and Verbeke, 2006). Never- patterns it is necessary to address a wider range of consumers, so
theless, the contradictory results found in the relationship between arguments from conventional marketing must not be forgotten.
attitudes and pro-environmental behaviour suggest that deeper In regard to educational institutions, the main challenge is to
research is needed. In fact, the way in which attitude (“environ- increase both objective and subjective environmental knowledge
mental concern”) and motivation (“internal stimulus”) have been levels. As noted above, the subsequent relationship between
defined in earlier research and indeed here, could be partly knowledge and behaviour is complex, so the integration of sus-
responsible for those contradictory results, since these factors seem tainable development into higher education must consider and
not to be clearly defined. In this respect, Kollmuss and Agyeman address that complexity. Nevertheless, while increasing objective
(2002) argue “that this difficulty in delimiting the different fac- knowledge might be a way for higher education institutions to
tors is due to the fact that most are broadly and vaguely defined, improve pro-environmental behaviour in advanced countries, this
interrelated, and often do not have clear boundaries”. measure does not seem to be worth applying in emerging coun-
The estimated ordered logit models for both emerging and tries. Additionally, educational institutions and governments
advanced countries show differences but also similarities. The should attempt to reduce the gap between objective and subjective
factors with the biggest weight in terms of environmental perfor- knowledge of environmental issues so that their future leaders
mance are psychological variables such as motivation, PCE and the become aware of their actual cognitive capacity to tackle environ-
importance of price as a motivational factor for engaging in green mental issues effectively.
purchasing. Knowledge-related factors are also relevant, and Finally, researchers should consider in their investigations that
gender must also be taken into account in explaining pro- different patterns of environmental behaviour might be explained
environmental behaviour. by structural and cultural differences across countries. Therefore,
The main conclusion in regard to differences between emerging the scales constructed previously to define the influence of envi-
and developed countries is that this economic-development- ronmental factors in pro-environmental behaviour are likely to
related classification cannot be applied simplistically to environ- have to be adapted to the research objectives, and also to the spe-
mental performance. Moreover, pro-environmental behaviour is cific cultural and geographical context of the research.
analysed as the sum of three components and the results reveal There are several limitations that should be considered before
that different patterns occur in each context. Thus, Spanish stu- these results are generalised. First, the subsamples were taken at a
dents stand out for their recycling behaviour, Mexican and Spanish single university in each of the four countries studied. This implies
students show a greater tendency to use public transport and USA that they might not reflect different situations that may exist in
students show far higher buying behaviour than any of the others each country, so the results cannot be generalised to these coun-
but are the least likely to use public transport. These patterns are tries as a whole.
probably due to the different cultural habits and structures that Second, although the sample comes from four different coun-
exist in each country. tries, generalisation to other cultures may be problematic and more
Recommendations for companies must be linked to improving future research needs to be done so that the conclusions can be
the perception of the cost-benefit link for pro-environmental applied to other countries with different socio-economic
 M.A. Vicente-Molina et al. / Journal of Cleaner Production 61 (2013) 130e138 137

characteristics. However, this analysis may be sufficient for the Barber, N., Taylor, D.C., Strick, S., 2009. Environmental knowledge and attitudes:
influencing the purchase decisions of wine consumers. In: Int. CHRIE
purpose of looking for differences between university students in
Conference-Refereed Track. Event 16 [2009]. http://scholarworks.umass.edu/
countries with different levels of economic and social development. refereed/Sessions/Wednesday/16.
Moreover, the dependent variable e environmental perfor- Bartiaux, F., 2008. Does environmental information overcome practice compart-
mance e is measured through three different components: recy- mentalisation and change consumers’ behaviours? J. Clean. Prod. 16 (11), 1170e
1180.
cling, public transport use and green purchasing. Several studies Bradley, J.C., Waliczek, T.M., Zajicek, J.M., 1999. Relationship between environmental
reported in this paper indicate that the effect of explanatory vari- knowledge and environmental attitude of high school students. J. Environ.
ables might vary depending on the issue analysed. Thus, it might be Educ. 30, 17e21.
Chan, K., 1998. Mass communication and proenvironmental behaviour: waste
interesting to supplement this research by including other aspects recycling in Hong Kong. J. Environ. Manag. 52, 317e325.
of pro-environmental behaviour. Choo, S., Mokhtarian, L., 2004. What type of vehicle do people drive? the role of
In the model drawn up, the behaviour patterns analysed (recy- attitude and lifestyle in influencing vehicle type choice. Transp. Res. Part. A:
Policy Pract. 38 (3), 201e222.
cling, public transport use and green purchasing) seem to be Corcoran, P.B., Wals, A.E.J., 2004. Higher Education and the Challenge of Sustain-
influenced by cultural and infrastructural factors, which are ability: Problematics, Promise and Practice. Kluwer Academic Publishers,
different in each country. Nevertheless, our results are, in general, Netherlands.
Cordano, M., Welcomer, S., Scherer, R., Pradenas, L., Parada, V., 2010. Understanding
consistent with previous research that highlights the significance of cultural differences in the antecedents of pro-environmental behaviour: a
knowledge and formal education, motivations, PCE and gender in comparative analysis of business students in the United States and Chile.
pro-environmental behaviour. J. Environ. Educ. 41 (4), 224e238.
Cottrell, S.P., 2003. Influence of sociodemographic and environmental attitudes of
In the model, the assumption that students will deliver their
general responsible environmental behaviour among recreational boaters. En-
proenvironmental awareness and performance in their future job- viron. Behav. 35, 347e375.
posts is underlying. Though it is highly wishful, this is a rather Cronbach, L.J., Shavelson, R.J., 2004. My current thoughts on coefficient alpha and
successor procedures. Educ. Psychol. Meas. 64, 391e418.
dubious position considering the limitations (mostly of a financial
Davidson, D., Freudenburg, W., 1996. Gender and environmental risk concerns: a
nature) that executives must consider in their decision making review and analysis of available research. Environ. Behav. 28, 302e339.
process. But it should be kept in mind that university students are Daykin, A., Moffatt, P., 2002. Analyzing ordered responses: a review of the ordered
the consumers, researchers and entrepreneurs of the future, and if probit model. Underst. Stat. 1 (3), 157e166.
Derksen, L., Gartrell, J., 1993. The social context of recycling. Am. Sociol. Rev. 58 (3),
future talent is able to make decisions that are beneficial to the 434e442.
environment, society is more likely to make progress along the Diamantopoulos, A., Schlegelmilch, B.B., Sinkovics, R.R., Bohlen, G.M., 2003. Can
path towards sustainability. In this sense, the level of environ- socio-demographics still play a role in profiling green consumers? A review of
the evidence and an empirical investigation. J. Bus. Res. 56, 465e480.
mental knowledge and the role of environmental education in DiEnno, C.M., Hilton, S.C., 2005. High school students’ knowledge, attitudes, and
changing and addressing lifestyles and attitudes could be crucial in levels of enjoyment of an environmental education unit on nonnative plants.
altering individuals’ behaviour and in turning society towards J. Environ. Educ. 37 (1), 13e25.
Dodd, T.H., Laverie, D.A., Wilcox, J.F., Duhan, D.F., 2005. Differential effects of
sustainability. experience, subjective knowledge, and objective knowledge on sources of in-
Moreover, it should be taken into account that the dynamic formation used in consumer wine purchasing. J. Hosp. Tour. Res. 29 (1), 3e19.
nature of the inter-relationships among explanatory variables may Dunlap, R.E., Jones, R.E., 2002. Environmental concern: conceptual and measure-
ment issues. In: Dunlap, R.E., Michelson, W. (Eds.), Handbook of Environmental
produce a spiral with mutual impacts resulting from changes to any
Sociology. Greenwood Press, Westport, CT, pp. 482e524.
of the variables. Since this is a cross-sectional study, this fact has Ellen, P.S., 1994. Do we know what we need to know? Objective and subjective
not been considered. Time evolution dimension may be a future knowledge effects on pro-ecological behaviors. J. Bus. Res. 10 (1), 43e52.
research line using longitudinal data, as has been recommended by Ellen, P.M., Wiener, J.L., Cobb-Walgren, C., 1991. The role of perceived consumer
effectiveness in motivating environmentally conscious behaviours. J. Public
Vicente-Molina et al. (2012). Policy Mark 10, 102e117.
Ewert, A., Baker, D., 2001. Standing for where you sit: and exploratory analysis of
the relationship between academic major and environment beliefs. Environ.
Acknowledgements Behav. 33, 687e707.
Fotopoulos, C., Krystallis, A., 2002. Organic product avoidance. Reasons for rejection
The authors acknowledge financial support from Cátedra and potential buyers’ identification in a countrywide survey. Br. Food J. 104 (3,
4, 5), 233e260.
UNESCO (Basque Government, grant 08/25), the FESIDE Foundation
Frank, D.J., Meyer, J.W., 2007. University expansion and the knowledge society.
(2009, 2010 and 2011) and the Econometrics Research Group Theor. Soc. 36, 287e311.
(Basque Government grant IT-IT642-13). Frick, J., Kaiser, F.G., Wilson, M., 2004. Environmental knowledge and conservation
behaviour: exploring prevalence and structure in a representative sample. Pers.
Individ. Differ. 37, 1597e1613.
References García-Valiñas, M.A., Muñiz-Pérez, M.A., Cordero-Ferrera, J.M., 2010. The role of
schools in providing environmental knowledge in science. In: Mancebón-
Adomssent, M., 2013. Exploring universities’ transformative potential for Torrubia, M.J., Xímenez-de-Embún, D.P., Gómez-Sancho, J.M., Giménez-
sustainability-bound learning in changing landscapes of knowledge commu- Esteban, G. (Eds.), 2010. Investigaciones de Economía de la Educación, vol. 5 (5).
nication. J. Clean. Prod. 49, 11e24. Asoc. de Econ. de la Educ, pp. 87e100.
Adomssent, M., Godemann, J., Michelsen, G., 2007. Transferability of approaches to Greene, W.H., Hensher, D.A., 2010. Modelling Ordered Choices. Cambridge Univer-
sustainable development at universities as a challenge. Int. J. Sust. High. Educ. 8 sity Press, Cambridge.
(4), 385e402. Guagnano, G., Stern, P., Dietz, T., 1995. Influences on attitude-behaviour relation-
Amyx, D.A., DeJong, P.F., Lin, X., Chakraborty, G., Wiener, J.L., 1994. Influences of ships: a natural experiment with curbside recycling programs. Environ. Behav.
purchase intentions for ecologically safe products: an exploratory study. In: 27, 699e718.
Park, C.W., Smith, D.C. (Eds.), Marketing Theory and Applications, Proceedings Hines, J.M., Hungerford, H.R., Tomera, A.N.,1986e1987. Analysis and synthesis of research
of the 1994 American Marketing Association Winter Educators Conference. on responsible environmental behaviour: a meta-analysis. J. Environ. Educ. 18, 1e8.
AMA, Chicago, pp. 341e347. Hopper, J.R., Nielsen, J.M., 1991. Recycling as altruistic behavior. Normative and
Asmuni, S., Khalili, J.M., Zain, Z.M., 2012. Sustainable consumption practices of behavioral strategies to expand participation in a community recycling pro-
students in an urban setting: a case in Selangor. Procedia e Soc. Behav. Sci. 36, gram. Environ. Behav. 23 (2), 195e220.
716e722. International Monetary Fund, 2012. World Economic Outlook April 2012. Series:
Balderjahn, I., 1988. Personality variables and environmental attitudes as predictors World Economic and Financial Surveys. IMF Publication Services, Washington.
of ecologically responsible consumption patterns. J. Bus. Res. 17 (1), 51e56. Kaiser, F.G., Fuhrer, U., 2003. Ecological behaviour’s dependency on different forms
Ballantyne, R., Connell, S., Fien, J., 1998. Students as catalysts of environmental of knowledge. Appl. Psychol. Int. Rev. 52 (4), 598e613.
change: a framework for researching intergenerational influence through Kennedy, E.H., Beckley, T.M., McFarlane, B.L., Nadeau, S., 2009. Why we don’t walk
environmental education. Environ. Educ. Res. 4, 285e298. the talk: understanding the environmental values/behaviour gap in Canada.
Bamberg, S., Moser, G., 2007. Twenty years after Hines, Hungerford and Tomera: a Hum. Ecol. Rev. 16 (2), 151e160.
new meta-analysis of psycho-social determinants of pro-environmental Kim, Y., Choi, S., 2003. Antecedents of pro-environmental behaviours: an exami-
behaviour. J. Environ. Psychol. 27, 14e25. nation of cultural values, self-efficacy, and environmental attitudes. In: Paper
138 M.A. Vicente-Molina et al. / Journal of Cleaner Production 61 (2013) 130e138

Presented at the Annual Meeting of the International Communication Associ- Schlegelmilch, B.B., Bohlen, G.M., Diamantopoulos, A., 1996. The link between green
ation. online from. http://citation.allacademic.com/meta/p_mla_apa_research_ purchasing decisions and measures of environmental consciousness. Eur. J.
citation/1/1/1/5/2/p111527_index.html. Mark 30, 35e55.
Kinnear, T.C., Taylor, J.R., Ahmed, S.A., 1974. Ecologically concerned consumers: who Schmitt, N., 1996. Uses and abuses of coefficient alpha. Psychol. Assess. 8 (4), 350e
are they? J. Mark 11, 20e24. 353.
Kollmuss, A., Agyeman, J., 2002. Mind the gap: why do people act environmentally Soyez, K., Francis, J., Smirnova, M., 2011. How individual, product and situational
and what are the barriers to pro-environmental behaviour? Environ. Educ. Res. determinants affect the intention to buy and organic food buying behavior: a
8 (3), 239e260. cross-national comparison in five nations. Der Mark 51 (1), 27e35.
Laroche, M., Bergeron, J., Barbaro-Forleo, G., 2001. Targeting consumers who are Stern, P.S., Dietz, T., Karlof, L., 1993. Values orientation, gender, and environmental
willing to pay more for environmentally friendly products. J. Consum. Mark 18 concern. Environ. Behav. 25 (3), 322e348.
(6), 503e520. Straughan, R.D., Roberts, J.A., 1999. Environmental segmentation alternatives: a look
Laskova, A., 2007. Perceived consumer effectiveness and environmental concerns. at green consumer behaviour in the new millennium. J. Consum. Mark 16 (6),
In: Proceedings of the 13th Asia Pacific Management Conference, Melbourne, 558e575.
Australia, pp. 206e209. Synodinos, N.E., 1990. Environmental attitudes and knowledge: a comparison of
Lozano, R., 2006. Incorporation and institutionalisation of SD into universities: marketing and business students with other groups. J. Bus. Res. 20 (2), 161e170.
breaking through barriers to change. J. Clean. Prod. 14 (9e11), 787e796. Szerényi, Z.M., Zsóka, Á., Széchy, A., 8e10 June 2009. Environmental Education and
Lozano, R., Lukman, R., Lozano, F.J., Huisingh, D., Lambrechts, W., 2013. Declarations Pro-environmental Consumer Behaviour-Results of a University Survey. Joint
for sustainability in higher education: becoming better leaders, through Actions on Climate Change, Denmark.
addressing the university system. J. Clean. Prod. 48, 10e19. Talay, I., Gündüz, S., Akpinar, N., 2004. On the status of environmental education
Mainieri, T., Barnett, E., Valdero, T., Unipan, J., Oskamp, S., 1997. Green buying: the and awareness of undergraduate students at Ankara University, Turkey. Inter. J.
influence of environmental concern on consumer behaviour. J. Soc. Psychol. 137 Environ. Pollut. 21 (3), 293e308.
(2), 189e204. Thamwipat, K., Maneewan, S., Pumiaroen, T., 2012. Behavior of media exposure and
Maloney, M.P., Ward, M.P., 1973. Ecology: let’s hear from the people. An objective participation in environmental activities of king Mongkut’s University Technol-
scale for the measurement of ecological attitudes and knowledge. Am. Psychol. ogy Thonburi dormitory students. World Acad. Sci. Eng. Technol. 61, 247e251.
28, 583e586. Thøgersen, J., 2011. Green shopping for selfish reasons or the common good? Am.
Martin, B., Simintiras, A.C., 1995. The impact of green product lines on the envi- Behav. Sci. 55 (8), 1052e1076.
ronment: does what they know affect how they feel? Mark. Intell. Plan. 13 (4), Tilikidou, I., 2007. The effects of knowledge and attitudes upon Greeks’ pro-
16e23. environmental purchasing behaviour. Corp. Soc. Responsib. Environ. Manag.
Milfont, T.L., Duckitt, J., 2010. The environmental attitudes inventory: a valid and 14, 121e134.
reliable measure to assess the structure of environmental attitudes. J. Environ. Vermeir, I., Verbeke, W., 2006. Sustainable food consumption: exploring the con-
Psychol. 30 (1), 80e94. sumer “attitude e behavioral intention” gap. J. Agric. Environ. Ethics 19 (2),
Mobley, C., Vagias, W.M., DeWard, S.L., 2010. Exploring additional determinants of 169e194.
environmentally responsible behaviour: the influence of environmental liter- Vicente-Molina, M.A., Tamayo-Orbegozo, U., Izagirre-Olaizola, J., 2012. Revisión de
ature and environmental attitudes. Environ. Behav. 42 (4), 420e447. la metodología empleada y resultados alcanzados en la investigación sobre
Moisander, J., 2007. Motivational complexity of green consumerism. Int. J. Consum. actuación medioambiental y rendimiento económico. J. Quant. Method. Econ.
Stud. 31 (4), 404e409. Bus. Adm. 14 (5), 5e35.
Nixon, H., Saphores, J.D.M., Ogunseitan, O.A., Shapiro, A.A., 2009. Understanding Waas, T., Verbruggen, A., Wright, T., 2010. University research for sustainable devel-
preferences for recycling electronic waste in California: the influence of envi- opment: definition and characteristics explored. J. Clean. Prod. 18 (7), 629e636.
ronmental attitudes and beliefs on willingness to pay. Environ. Behav. 41, 101e Wang, Y., Liu, J., Hansson, L., Zhang, K., Wang, R., 2011. Implementing stricter
124. environmental regulation to enhance eco-efficiency and sustainability: a case
Oguz, D., Çakci, I., Kavas, S., 2010. Environmental awareness of university students study of Shandong Province’s pulp and paper industry. J. Clean. Prod. 19 (4),
in Ankara, Turkey. Afr. J. Agric. Res. 5, 2629e2636. 303e310.
Olli, E., Grendstad, G., Wollebaek, D., 2001. Correlates of environmental behaviours: Wilkie, W.L., 1990. Consumer Behaviour, second ed. John Wiley and Sons, New York.
bringing back social context. Environ. Behav. 33 (2), 181e208. Zelezny, L.C., Chua, P.P., Aldrich, C., 2000. Elaborating on gender differences in
Paap, R., Franses, P.H., 2000. A dynamic multinomial probit model for brand choice environmentalism. J. Soc. Issue. 56 (3), 443e457.
with different long-run and short-run effects of marketing mix variables. Zilahy, G., Huisingh, D., 2009. The roles of academia in regional sustainability ini-
J. Appl. Econom. 15, 717e744. tiatives. J. Clean. Prod. 17 (2), 1057e1066.
Rex, E., Baumann, H., 2007. Beyond ecolabels: what green marketing can learn from Zsóka, A., Szerényi, Z., Széchy, A., Kocsis, T., 2012. Greening due to environmental
conventional marketing. J. Clean. Prod. 15 (6), 567e576. education? Environmental knowledge, attitudes, consumer behaviour and
Schahn, J., Holzer, E., 1990. Studies of individual environmental concern: the role of everyday pro-environmental activities of Hungarian high school and university
knowledge, gender, and background variables. Environ. Behav. 22 (6), 767e786. students. J. Clean. Prod.. http://dx.doi.org/10.1016/j.jclepro.2012.11.030.