18  Models to explain
environmental behaviour
Linda Steg
University of Groningen,The Netherlands
Annika Nordlund
Umeå University, Sweden
C H A P T E R O U T L I N E
18.1	 INTRODUCTION 186
18.2	 THEORY OF PLANNED BEHAVIOUR 186
18.3	 THE NORM ACTIVATION MODEL 189
18.4	 THE VALUE–BELIEF–NORM THEORY OF
ENVIRONMENTALISM 191
18.5	 GOAL FRAMING THEORY 192
18.6	 SUMMARY 193
	 GLOSSARY 194
	 SUGGESTIONS FOR FURTHER READING 195
	 REVIEW QUESTIONS 195
186  ENVIRONMENTAL PSYCHOLOGY: AN INTRODUCTION
18.1  INTRODUCTION
In this chapter, we discuss models of factors that influence environmental behaviour.
These models mostly assume that people make reasoned choices (see Steg & Vlek,
2009a); the role of automaticity and habits will be discussed in Chapter 19. We first
discuss models that focus on the role of individual costs and benefits, of which the
theory of planned behaviour is the most prominent example. Next, we elaborate on
models that focus on morality, in particular the norm activation model and the valuebelief-norm
theory of environmentalism. Finally, we discuss goal framing theory,
which provides an integrated framework for understanding factors influencing environmental
behaviour.
18.2  THEORY OF PLANNED
BEHAVIOUR
The theory of planned behaviour (TPB; Ajzen, 1985; see Figure 18.1) assumes that
individuals make reasoned choices, and that behaviour results from the intention to
engage in specific behaviour (i.e. whether they plan to do so). The stronger your
intention, the more effort you will make to conduct a particular behaviour, and the
more likely it is that you engage in the behaviour. The intention depends on attitudes
towards the behaviour, subjective norms related to the behaviour and perceived behaviour
control.
Attitudes reflect the extent to which engaging in the behaviour is evaluated positively
or negatively. For example, you may have a favourable attitude towards cycling,
but a negative attitude towards taking a bus. Attitudes are based on beliefs about the
likely costs and benefits of behaviour, weighted with the extent to which these costs
and benefits are considered important. For example, a person may believe that the
car is fast, comfortable, reliable and enjoyable, and consider these aspects as highly
important. On the other hand, this person may think the car is expensive and not
environmentally friendly, but consider these aspects to be less important. This will
result in an overall positive attitude towards car use, as the weighted benefits are
higher than the weighted costs.
Subjective norms reflect the extent to which a person believes that important
others would approve or disapprove of the behaviour (and are thus similar to injunctive
norms; see Chapter 15) and reflect social costs and benefits of behaviour. SubjecEnvironmental
Psychology: An Introduction, First Edition. Edited by Linda Steg, Agnes E. van den Berg,
Judith I. M. de Groot.
© 2013 the British Psychological Society and John Wiley & Sons, Ltd. Published 2013 by the British
Psychological Society and John Wiley & Sons, Ltd.
186
Models to explain environmental behaviour  187
tive norms are based on beliefs about the expectations of relevant reference groups
concerning the behaviour, weighted by one’s motivation to comply with these expectations.
For example, your classmates and friends may expect you to cycle to university
and you may be strongly motivated to comply with their expectations, while
your neighbour, whose opinion you value far less, may approve of your driving to
university. In this case, you will experience a subjective norm in favour of cycling
rather than driving.
Perceived behavioural control refers to the perceived possibility to perform the
behaviour, which depends on beliefs about the presence of factors that may facilitate
or hinder the relevant behaviour. For example, you may believe you cannot afford to
drive to work, resulting in a low perceived behaviour control to drive, or you may
believe that cycling to work is exhausting and that you are not fit enough to cycle,
resulting in a low behaviour control to cycle. Perceived behavioural control can influence
behaviour via intention, as explained in the examples above, but may also influence
behaviour directly. For example, suppose that you intend to take the bus to
work, but then learn that the bus drivers are on strike. In that case, perceived behaviour
control to travel by bus will be low and directly affect your travel behaviour.
The TPB assumes that all other factors, such as socio-demographics, general
beliefs and values, influence behaviour indirectly, via attitudes, subjective norms and
perceived behavioural control. For example, strong biospheric values (see Chapter 14)
may result in positive attitudes towards cycling and a negative attitude towards
driving, given their different environmental impacts. Also, owing to poor public
transport services, people living in the countryside may have a lower perceived
behavioural control to take the bus than urbanites.
The TPB has been successful in explaining various types of environmental behaviour.
Bamberg and Schmidt (2003) found that students were more likely to travel by
car to university when they intended to do so. The intention to drive was higher
when they had a positive attitude towards driving, when subjective norms were in
favour of car use, and, to a lesser extent, when they thought they were able to drive
to university. Harland, Staats and Wilk (1999) found that the TPB variables, and
Figure 18.1  A schematic representation of the theory of planned behaviour.
BehaviourIntention
Perceived behaviour
control
Attitude
Subjective norm
188  ENVIRONMENTAL PSYCHOLOGY: AN INTRODUCTION
particularly attitudes and perceived behavioural control, predicted a wide range of
environmental behaviours, including the intention to use transport forms other than
the car, the use of unbleached paper, reductions in meat consumption and the use
of energy-saving light bulbs. They did not include intentions as a predictor of behaviour,
since they measured intentions and past behaviour at the same moment of time;
indeed, it makes theoretically no sense to predict past behaviour from future
intention.
The predictive power of the TPB increases when other motivational predictors
are included in the model (e.g. habits; see Chapter 19). Harland and colleagues
(1999) found that personal norms, that is, feelings of moral obligation to engage in
particular pro-environmental actions, predicted different pro-environmental intentions
and behaviours over and above the TPB variables (see Box 18.1 for details).
Similar results were reported by Bamberg and Schmidt (2003): personal norms predicted
students’ car use and bus use, respectively, over and above the TPB
variables.
Personal norms are a key factor in two other prominent theories in environmental
psychology that we will discuss below: the norm activation model and the value–belief–
norm theory of environmentalism. These models assume that morality plays a key role
in pro-environmental behaviour because pro-environmental actions often involve
higher costs and effort for individuals: in many cases, people will only act proenvironmentally
because they feel it is the right thing to do.
BOX 18.1  EXTENDING THE TPB WITH  
PERSONAL NORMS
Harland and colleagues (1999) asked respondents
to indicate how often they had used
unbleached paper during the last 6 months.
Respondents indicated (1) how they evaluated
the use of unbleached paper (attitudes), (2) to
what extent important others expect them to
use unbleached paper (subjective norms), (3)
whether they could in most instances use
unbleached paper when they wanted to do so
(perceived behaviour control) and (4) whether
they felt personally obliged to use unbleached
paper (personal norms). The TPB variables
explained 28 per cent of the variance in the
behaviour: positive attitudes towards using
unbleached paper and a higher perceived
behavioural control resulted in a higher use of
unbleached paper.Subjective norms did not significantly
contribute to this regression model.
When adding personal norms to this regression
model, 34 per cent of the variance in behaviour
was explained. Personal norms appeared to be
the strongest predictor: participants used
unbleached paper more often when they felt
morally obliged to do so. Attitudes and perceived
behaviour control were still significant
predictors as well. So, adding personal norms
significantly improved the predictive power of
the TPB. Similar results were found for other
types of pro-environmental intentions and
behaviour included in this study.
Models to explain environmental behaviour  189
18.3  THE NORM ACTIVATION MODEL
The norm activation model (NAM; Schwartz, 1977; Schwartz & Howard, 1981; Steg
& De Groot, 2010) proposes that pro-environmental actions follow from the activation
of personal norms, reflecting feelings of moral obligation to perform or refrain
from specific actions. Personal norms are activated by four key variables:
•	 Problem awareness (or awareness of need): level of awareness of the adverse
consequences of not acting pro-environmentally.
•	 Ascription of responsibility: feeling responsible for the negative consequences
of not acting pro-environmentally.
•	 Outcome efficacy: the identification of actions to reduce environmental
problems.
•	 Ability or self-efficacy: recognition of own ability to provide relief to
environmental threats.
Schwartz and Howard (1981) called these situational variables, because the
strength of these variables differs across situations; they are not stable over time. The
basic idea is that personal norms are activated when someone is aware of the environmental
problems caused by his or her behaviour, and when he or she feels personally
responsible for these problems and does not attribute these problems to the
actions of others, industry or the government. Moreover, people should believe that
their actions may help to reduce the relevant problems (i.e. outcome efficacy). This
is particularly important in case of environmental problems that can be solved only
when lots of people cooperate, such as to reduce harmful emissions or resource use.
In these cases, outcome efficacy strongly depends on the expectation that others will
engage in pro-environmental actions too. Finally, one should feel to be able to engage
in the actions needed to reduce the relevant environmental problems (i.e. selfefficacy);
this is comparable to perceived behavioural control in the TPB.
The NAM appeared to be successful in explaining various types of proenvironmental
intentions and behaviours, such as energy conservation (Black, Stern,
& Elworth, 1985), willingness to pay for environmental protection (Guagnano, Dietz,
& Stern, 1994), intention or willingness to reduce car use (Abrahamse, Steg, Gifford,
& Vlek, 2009; Eriksson, Garvill, & Nordlund, 2006), using the car for short distances
and turning off the tap while brushing teeth (Harland, Staats, & Wilke, 1999), political
behaviour (Gärling, Fujii, Gärling, & Jakobsson, 2003) and general pro-environmental
behaviour (Nordlund & Garvill, 2002). However, studies typically did not include all
four situational variables listed above. Notably, most studies did not include selfefficacy
and included either ascription of responsibility or outcome efficacy.
Two issues make it difficult to compare results across studies (Steg & De Groot,
2010). First, the main constructs of the NAM have been conceptualised differently
across studies. Some studies measured the situational variables listed above on a
general level such as general awareness of environmental problems (e.g. Stern, Dietz,
190  ENVIRONMENTAL PSYCHOLOGY: AN INTRODUCTION
Abel, Guagnano, & Kalof, 1999), while other studies included behaviour-specific variables,
such as awareness of problems caused by energy use (e.g. Nordlund & Garvill,
2003; Steg, Dreijerink, & Abrahamse, 2005). Behaviour-specific beliefs are generally
more strongly related to intentions and behaviours than are general beliefs (Ajzen,
1985). This suggests that the NAM, like the TPB, will be better able to explain variance
in environmental intentions and behaviour when the predictor variables are
tuned towards the intention or behaviour explained (see also Schwartz, 1977).
Second, the causal relationships between the model variables have been interpreted
differently. Some scholars argued that situational factors moderate the relationship
between personal norms and intention or behaviour (e.g. Schultz & Zelezny,
1998), which implies that personal norms are particularly predictive of intentions and
behaviour when problem awareness, responsibility feelings, outcome efficacy and
self-efficacy are strong. Other scholars argued that the situational factors predict
personal norms (e.g. Bamberg & Schmidt, 2003; Nordlund & Garvill, 2002, 2003),
and that the strength of personal norms thus depend on the situational variables.
Steg and De Groot (2010; De Groot & Steg, 2009) compared the predictive power of
both model interpretations, and found most support for the latter mediator model.
Apparently, people first should be aware of the problems caused by their behaviour
before they consider their own responsibility for these problems, and before considering
whether they can effectively help to reduce these problems and feeling morally
obliged to do so (see Box 18.2). The mediator model is also theoretically more plausible,
because it is not likely that people will think about their obligation to protect
BOX 18.2  TESTING CAUSAL RELATIONSHIPS  
IN THE NAM
Steg and De Groot (2010) tested causal relationships
in the NAM in an experimental study.They
manipulated problem awareness by presenting
half the respondents with a text in which the
problems of particulate matters were stressed,
whereas the other respondents read a text in
which these problems were trivialised. The
manipulation was successful: problem awareness
was higher in the first group than in the
second group. Next, they examined to what
extent this manipulation influenced feelings of
responsibility to take action to reduce emissions
of particulate matters (e.g. I believe that I am
co-responsible for the reduction of particulate
matters in the city), personal norms towards
doing so (e.g. I feel morally obliged to demonstrate
against particulate matters) and intention
to participate in such actions (e.g.to what extent
are you prepared to collect signatures to reduce
the emissions of particulate matters?). As
expected,higher problem awareness resulted in
stronger ascription of responsibility, personal
norms and intention to participate in actions to
reduce the emission of particulate matters. As
this was an experimental study, in which one
variable was manipulated while all other variables
were kept constant, we can conclude that
higher problem awareness causes stronger
responsibility feelings, personal norms and
increased pro-environmental intentions.
Models to explain environmental behaviour  191
the environment and to act accordingly when they are not aware of any environmental
problems caused by their behaviour, and when they do think they can not
engage in effective actions to reduce these problems.
18.4  THE VALUE–BELIEF–NORM
THEORY OF ENVIRONMENTALISM
The value–belief–norm theory of environmentalism (VBN theory; Stern, 2000; Stern
et al., 1999) is an extension of the NAM. The VBN theory proposes that the situational
factors, notably problem awareness, depend on values (i.e. general goals that
serve as guiding principles in your life) and ecological worldviews (i.e. beliefs on relationships
between humans and the environment; see also Chapter 14). The VBN
theory proposes that egoistic values are negatively related and altruistic and biospheric
values are positively related to ecological worldviews. Ecological worldviews predict
problem awareness, which in turn influences one’s beliefs on whether one can act to
reduce the environmental threat, personal norms and subsequently behaviour (see
Figure 18.2). Each variable in the causal chain is related to the next variable, but may
also be directly related to variables further down the chain, although these relationships
should be weaker. This is in line with the NAM as a mediator model discussed
above. Personal norms may influence all kinds of behaviours taken with proenvironmental
intent, including environmental activism (e.g. active involvement in
Figure 18.2  A schematic representation of the value–belief–norm theory of environmentalism.
Adapted from Stern (2000).
Biospheric
Altruistic
Egoistic
Values Beliefs
Ecological
worldview
(NEP)
Awareness of
consequences
Ascription of
responsibility
Sense of
obligation to
take pro-
environmental
action
Activism
Nonactivist
behaviour in
the public
sphere (e.g.
acceptability
policies)
Private-sphere
behaviours
Organisational
behaviours
pro-environmental
personal norms Behaviour
192  ENVIRONMENTAL PSYCHOLOGY: AN INTRODUCTION
environmental organisations or demonstrations), non-activist behaviours in the
public sphere (e.g. environmental citizenship, acceptability of public policies), privatesphere
environmentalism (i.e. the purchase, use and disposal of products that have
environmental impact) and organisational actions (e.g. the design of environmentally
benign products; see Figure 18.2).
The VBN appears to be successful in explaining environmental citizenship (Stern
et al., 1999), household energy use (Abrahamse & Steg, 2011) and policy acceptability
(Eriksson, Garvill, & Nordlund, 2006; 2008; Steg et al., 2005). Steg and colleagues
(2005) found support for the causal structure as proposed in the VBN theory: all
variables were significantly related to the next variable in the causal chain, and in
most cases, the explanatory power of the model hardly increased when other predictor
variables further up the causal chain were entered into the regression model as
well. Only biospheric values were also significantly related to feelings of moral obligation
when intermediate variables were controlled for, which suggests that
biospheric values can directly activate personal norms.
Overall, studies revealed that the NAM and VBN theory are particularly successful
in explaining low-cost pro-environmental behaviours and ‘good intentions’ such as
willingness to change behaviour, political behaviour, environmental citizenship and
policy acceptability (as explained above), but have less explanatory power in situations
characterised by high behavioural costs or strong constraints on behaviour, such
as reducing car use (e.g. Abrahamse & Steg, 2009; Bamberg & Schmidt, 2003). In
such settings, the TPB appears to be more powerful in explaining environmental
behaviour (Bamberg & Schmidt, 2003), probably because the TPB considers a wider
range of factors, notably non-environmental motivations. However, it is not clear yet
which theoretical model is most useful in which situation because systematic research
on the range of application of each theory is lacking.
When acting pro-environmentally is costly, people are probably tempted to reduce
feelings of moral obligation via self-serving denial (Lindenberg & Steg, 2007; Schwartz
& Howard, 1981). They can do so by downplaying the four situational factors
described earlier. For example, people can deny the seriousness of environmental
problems, reject their liability for these problems or identify others such as authorities
or industry as responsible for environmental problems, indicate that individual proenvironmental
actions are not effective in reducing environmental problems or deny
their personal ability to perform the necessary actions.
18.5  GOAL FRAMING THEORY
Various scholars have integrated concepts and variables from different theories, such
as including variables from the NAM in the TPB, as described above (e.g. Abrahamse
& Steg, 2009; Harland et al., 1999). This suggests that behaviour result from cost–
benefit considerations (as proposed by the TPB) as well as normative considerations.
Goal framing theory (Lindenberg & Steg, 2007; see also Chapter 12) proposes that
Models to explain environmental behaviour  193
goals govern or ‘frame’ the way people process information and act upon it. When
a goal is activated (that is, when it is ‘focal’), it influences what a person thinks of at
the moment, what information he or she is sensitive to, what alternatives he or she
perceives, and how he or she will act. As described in Chapter 12, goal framing theory
distinguishes three general goals that influence behaviour: hedonic goals ‘to feel
better right now’, gain goals ‘to guard and improve one’s resources’ and normative
goals ‘to act appropriately’. Hedonic goals are a priori strongest, while especially
normative goals need external support to become focal.
According to goal framing theory, one goal is focal and influences information
processing the most, that is, it is the goal frame, while other goals are in the background
and increase or decrease the strength of the focal goal. Thus, multiple goals
may be active at any given time. When background goals are compatible with the
goal frame, they strengthen it. For example, when normative goals are focal, you are
probably more motivated to act pro-environmentally, and when this choice is also
most beneficial (e.g. cycling is cheaper than driving) or comfortable (e.g. insulating
your house also increases comfort), gain or hedonic goals will strengthen your normative
goal. But when the goal frame and background goals conflict, the latter
weaken the strength of the goal frame. For example, when your normative goal is
focal and then you realise that organic food is very expensive (so it conflicts with your
gain goal), the normative goal will be weakened, and one may decide to choose
products that are somewhat less environmentally friendly yet not as expensive. Normative
goals provide the most stable basis for pro-environmental actions, because
acting pro-environmentally is always the most appropriate way to act. However, if
people act pro-environmentally based on gain or hedonic goals, they will do so only
as long as it is profitable and comfortable to do so, which may not always be the case
(De Groot & Steg, 2009). For example, when hedonic goals are focal, people may
engage in pro-environmental actions when they are in a good mood but not when
they are in a bad mood.
The three goal frames coincide with three theoretical frameworks commonly
used in environmental psychology (Lindenberg & Steg, 2007): theories and models
on affect (see Chapter 16) focus on hedonic goals, the TPB focuses on gain goals,
while the NAM and VBN theory focus on normative goals. Thus, goal framing theory
offers an integrative framework for understanding environmental behaviour.
18.6  SUMMARY
This chapter discussed a number of prominent models explaining environmental
behaviour. The theory of planned behaviour (TPB) assumes that individuals make
reasoned choices, and that behaviour results from the intention to engage in specific
behaviour. Environmental intentions and behaviours are more likely when people
have a positive attitude towards the relevant behaviour, when subjective norms
support this behaviour, and when one feels in control over the behaviour. The norm
194  ENVIRONMENTAL PSYCHOLOGY: AN INTRODUCTION
activation model (NAM) and the value–belief–norm theory of environmentalism
(VBN) focus on the relation between morality and environmental behaviour. The
NAM proposes that pro-environmental actions follow from the activation of personal
norms, reflecting feelings of moral obligation to perform or refrain from specific
actions. Personal norms are activated when someone is aware of the environmental
problems caused by his or her behaviour, feels personally responsible for these problems,
has the feeling that her or his individual actions may help to reduce the relevant
problems, and perceives her- or himself to have the abilities to engage in actions
needed to reduce the relevant environmental problems. The VBN theory is an extension
of the NAM, and proposes that the situational factors, notably problem awareness,
depend on ecological worldviews and on value orientations. The NAM and
VBN theory have successfully explained low-cost environmental behaviour and ‘good
intentions’, while the TPB shows more explanatory power in situations characterised
by high behavioural costs or strong constraints on behaviour.
Goal framing theory provides an integrated framework for understanding factors
influencing environmental behaviour, with an emphasis on multiple goals being
active at any given time. Goal framing theory distinguishes three general goals that
influence behaviour (hedonic, gain and normative goals), which coincide with theoretical
frameworks in environmental psychology: theories and models on affect
(hedonic goals), the TPB (gain goals), and the NAM and VBN theory (normative
goals).
GLOSSARY
ascription of responsibility  Feeling responsible for negative consequences of not acting
pro-environmentally.
attitudes  Mental dispositions to evaluate an attitude object (i.e. a person, place, thing, or event)
with some degree of favour or disfavour.
biospheric values  A value type reflecting the concern with the quality of nature and the environment
for its own sake.
ecological worldviews  Beliefs regarding humanity’s ability to upset the balance of nature, the
existence of limits to growth and rejecting humanity’s right to rule over the rest of nature.
gain goal  The goal to maintain and improve one’s resources.
goal frame  The focal goal in a particular situation.
goal framing theory  Integrated framework for understanding factors influencing environmental
behaviour, with an emphasis on the relative strength of hedonic, gain, and normative goals.
hedonic goal  The goals to feel good right now.
intention  A person’s specific purpose to engage in a particular action.
mediator model  A model that seeks to identify the mechanism that underlies an observed
relationship between an independent variable and a dependent variable via a third explanatory
variable, known as a mediator variable.
moderate  A moderator variable increases or decreases the strength of an effect.
norm activation model  A model proposing that pro-environmental action follows from the
activation of personal norms.
Models to explain environmental behaviour  195
REVIEW QUESTIONS
1.  Describe the theory of planned behaviour.
2.  What is the main difference between the norm activation model and the value–belief–norm
model of environmentalism?
3.  Which four situational factors influence the strength of personal norms?
4.  Which goals steer behaviour according to the goal framing theory?
normative goal  The goal to behave appropriately, conforming to social norms and legitimate
rules.
outcome efficacy  The extent to which a person identifies actions to reduce environmental
problems.
perceived behaviour control  The perceived possibility to perform behaviour in light of present
facilitating or hindering factors.
personal norm  Feelings of moral obligation to perform or refrain from specific actions.
problem awareness  The extent to which one is aware of the adverse consequences of not
acting pro-environmentally.
self-efficacy  The extent to which one recognises own ability to provide relief to environmental
threats.
subjective norm  Perceived social pressure to engage in behaviour.
theory of planned behaviour  A model assuming that individuals make reasoned choices and
that behaviour results from the intention to engage in specific behaviour.
value–belief–norm theory of environmentalism  An extension of the norm activation model,
proposing that problem awareness depends on ecological worldviews and value orientations.
values  Desirable transsituational goals varying in importance, which serve as guiding principles
in the life of a person or other social entity.
SUGGESTIONS FOR FURTHER READING
Bamberg, S., & Möser, G. (2007). Twenty years after Hines, Hungerford, and Tomera: A new
meta-analysis of psycho-social determinants of pro-environmental behaviour. Journal of Environmental
Psychology, 27, 14–25.
Steg, L., & Vlek, C. (2009). Encouraging pro-environmental behaviour: An integrative review and
research agenda. Journal of Environmental Psychology, 29, 309–317.
Vining, J., & Ebreo, A. (2002). Emerging theoretical and methodological perspectives on conservation
behaviour. In R. B. Bechtel & A. Churchman, Handbook of environmental psychology
(pp. 551–558). New York: Wiley.