Journal of Social Issues, Vol. 56, No. 3, 2000, pp.

407–424

Toward a Coherent Theory of Environmentally

Significant Behavior
Paul C. Stern
National Research Council

This article develops a conceptual framework for advancing theories of environ-

mentally significant individual behavior and reports on the attempts of the
author’s research group and others to develop such a theory. It discusses defini-
tions of environmentally significant behavior; classifies the behaviors and their
causes; assesses theories of environmentalism, focusing especially on
value-belief-norm theory; evaluates the relationship between environmental
concern and behavior; and summarizes evidence on the factors that determine
environmentally significant behaviors and that can effectively alter them. The
article concludes by presenting some major propositions supported by available
research and some principles for guiding future research and informing the
design of behavioral programs for environmental protection.

Recent developments in theory and research give hope for building the under-
standing needed to effectively alter human behaviors that contribute to environ-
mental problems. This article develops a conceptual framework for the theory of
environmentally significant individual behavior, reports on developments toward
such a theory, and addresses five issues critical to building a theory that can inform
efforts to promote proenvironmental behavior.

*This research was supported in part by the U.S. Environmental Protection Agency grant, “The
Social Psychology of Stated Preferences,” and by National Science Foundation grants SES 9211591
and 9224036 to George Mason University. I thank my colleagues Gregory Guagnano, Linda Kalof, and
especially Thomas Dietz and Gerald Gardner for their collaboration, support, and criticism in our
collective effort to theorize about environmental concern and behavior. Correspondence concerning
this article should be addressed to Paul C. Stern, National Research Council, 2101 Constitution Ave.,
N.W. (HA-172), Washington DC 20418 [e-mail: pstern@nas.edu].
407
© 2000 The Society for the Psychological Study of Social Issues
408 Stern

Defining Environmentally Significant Behavior

Environmentally significant behavior can reasonably be defined by its impact:

the extent to which it changes the availability of materials or energy from the
environment or alters the structure and dynamics of ecosystems or the biosphere
itself (see Stern, 1997). Some behavior, such as clearing forest or disposing of
household waste, directly or proximally causes environmental change (Stern,
Young, & Druckman, 1992). Other behavior is environmentally significant indi-
rectly, by shaping the context in which choices are made that directly cause
environmental change (e.g., Rosa & Dietz, 1998; Vayda, 1988). For example,
behaviors that affect international development policies, commodity prices on
world markets, and national environmental and tax policies can have greater envi-
ronmental impact indirectly than behaviors that directly change the environment.
Through human history, environmental impact has largely been a by-product
of human desires for physical comfort, mobility, relief from labor, enjoyment,
power, status, personal security, maintenance of tradition and family, and so forth,
and of the organizations and technologies humanity has created to meet these
desires. Only relatively recently has environmental protection become an impor-
tant consideration in human decision making. This development has given envi-
ronmentally significant behavior a second meaning. It can now be defined from the
actor’s standpoint as behavior that is undertaken with the intention to change
(normally, to benefit) the environment. This intent-oriented definition is not the
same as the impact-oriented one in two important ways: It highlights environmen-
tal intent as an independent cause of behavior, and it highlights the possibility that
environmental intent may fail to result in environmental impact. For example,
many people in the United States believe that avoiding the use of spray cans pro-
tects the ozone layer, even though ozone-destroying substances have been banned
from spray cans for two decades. The possible discrepancy between environmental
intent and environmental impact raises important research questions about the
nature and determinants of people’s beliefs about the environmental significance
of behaviors.
Both definitions of environmentally significant behavior are important for
research but for different purposes. It is necessary to adopt an impact-oriented
definition to identify and target behaviors that can make a large difference to the
environment (Stern & Gardner, 1981a). This focus is critical for making research
useful. It is necessary to adopt an intent-oriented definition that focuses on
people’s beliefs, motives, and so forth in order to understand and change the target
behaviors.
Theory of Environmentally Significant Behavior 409

Types of Environmentally Significant Behavior

Much early research on proenvironmental behavior presumed it to be a uni-

tary, undifferentiated class. More recently it has become clear that there are several
distinct types of environmentally significant behavior and that different combina-
tions of causal factors determine the different types.

Environmental Activism

Committed environmental activism (e.g., active involvement in environmen-

tal organizations and demonstrations) is a major focus of research on social move-
ment participation. This research provides detailed analysis of the “recruitment”
process through which individuals become activists (McAdam, McCarthy, & Zald,
1988).

Nonactivist Behaviors in the Public Sphere

Recently, the social movement literature has pointed to nonactivists’ support

of movement objectives as another important class of behavior (Zald, 1992). Pub-
lic opinion researchers and political scientists sometimes examine such behavior,
but relatively little research has been done to classify the behaviors into coherent
subtypes. It seems reasonable as a first approximation to distinguish between more
active kinds of environmental citizenship (e.g., petitioning on environmental
issues, joining and contributing to environmental organizations) and support or
acceptance of public policies (e.g., stated approval of environmental regulations,
willingness to pay higher taxes for environmental protection). My colleagues and I
have found empirical support for distinguishing these types from each other and
from activism (Dietz, Stern, & Guagnano, 1998; Stern, Dietz, Abel, Guagnano, &
Kalof, 1999). Although these behaviors affect the environment only indirectly, by
influencing public policies, the effects may be large, because public policies can
change the behaviors of many people and organizations at once. An important fea-
ture of public-sphere behaviors, including activism, is that environmental concerns
are within awareness and may therefore be influential.

Private-Sphere Environmentalism

Consumer researchers and psychologists have focused mainly on behaviors in

the private sphere: the purchase, use, and disposal of personal and household prod-
ucts that have environmental impact. It is useful to subdivide these according to the
type of decision they involve: the purchase of major household goods and services
that are environmentally significant in their impact (e.g., automobiles, energy for
the home, recreational travel), the use and maintenance of environmentally
410 Stern

important goods (e.g., home heating and cooling systems), household waste dis-
posal, and “green” consumerism (purchasing practices that consider the environ-
mental impact of production processes, for example, purchasing recycled products
and organically grown foods). Making such distinctions has revealed that some
types of choice, such as infrequent decisions to purchase automobiles and major
household appliances, tend to have much greater environmental impact than
others, such as changes in the level of use of the same equipment: the distinction
between efficiency and curtailment behaviors (Stern & Gardner, 1981a, 1981b).
Private-sphere behaviors may also form coherent clusters empirically (e.g., Bratt,
1999a), and different types of private-sphere behavior may have different determi-
nants (e.g., Black, Stern, & Elworth, 1985). Private-sphere behaviors are unlike
public-sphere environmentalism in that they have direct environmental conse-
quences. The environmental impact of any individual’s personal behavior, how-
ever, is small. Such individual behaviors have environmentally significant impact
only in the aggregate, when many people independently do the same things.

Other Environmentally Significant Behaviors

Individuals may significantly affect the environment through other behaviors,

such as influencing the actions of organizations to which they belong. For example,
engineers may design manufactured products in more or less environmentally
benign ways, bankers and developers may use or ignore environmental criteria in
their decisions, and maintenance workers’ actions may reduce or increase the pol-
lution produced by manufacturing plants or commercial buildings. Such behaviors
can have great environmental impact because organizational actions are the largest
direct sources of many environmental problems (Stern & Gardner, 1981a, 1981b;
Stern, 2000). The determinants of individual behavior within organizations are
likely to be different from those of political or household behaviors.

Evidence for Distinguishing Major Behavioral Types

Research my colleagues and I have conducted suggests that this distinction

among behavioral types is not only conceptually coherent but statistically reliable
and psychologically meaningful. For instance, a factor analysis of the behavioral
items in the environment module of the 1993 General Social Survey revealed a
three-factor solution (Dietz et al., 1998). One factor included four private-sector
household behaviors (e.g., buying organic produce, sorting household waste for
recycling); a second included two environmental citizenship behaviors (signing a
petition and belonging to an environmental group); and the third included three
items indicating willingness to make personal financial sacrifices for environmen-
tal goals, which assess policy support. A different pattern of social-psychological
Theory of Environmentally Significant Behavior 411

and socio-demographic predictors was associated with each of the behavioral

types, and even the two citizenship behaviors had quite different sets of predictors.
My colleagues and I had similar results using data from a 1994 national envi-
ronmental survey (Stern et al., 1999). Factor analysis of 17 items measuring
self-reported behaviors and behavioral intentions again revealed three factors:
consumer behaviors (e.g., buying organic produce, avoiding purchases from com-
panies that harm the environment); environmental citizenship (e.g., voting, writing
to government officials); and policy support, expressed as willingness to sacrifice
economically to protect the environment (e.g., by paying much higher taxes or
prices). Self-reported participation in environmental demonstrations and protests,
presumably a measure of committed activism, did not load on any of the above
three factors. Each of these factors was predicted by a different pattern of norms,
beliefs, and values, and activism had yet a different set of predictors.

The Determinants of Environmentalism

Environmentalism may be defined behaviorally as the propensity to take

actions with proenvironmental intent. Some theories treat environmentalism as a
matter of worldview. Perhaps the most prominent example in social psychology is
the idea that it flows from adopting a New Environmental (or Ecological) Para-
digm, within which human activity and a fragile biosphere are seen as inextricably
interconnected (Dunlap, Van Liere, Mertig, & Jones, this issue). Another
worldview theory explains environmentalism in terms of an egalitarian “cultural
bias” or “orienting disposition” (Dake, 1991; Douglas & Wildavsky, 1982; Steg &
Sievers, 2000). Recently, some researchers have begun to explore affective influ-
ences on environmental concern and behavior, including sympathy for others
(Allen & Ferrand, 1999), “emotional affinity” toward nature (Kals, Schumacher, &
Montada, 1999), and empathy with wild animals (Schultz, this issue).
Some theories look to values as the basis of environmentalism. Inglehart
(1990) suggests that it is an expression of postmaterialist values of quality of life
and self-expression that emerge as a result of increasing affluence and security in
the developed countries. Some accounts emphasize religious values, arguing either
that certain Judaeo-Christian beliefs predispose adherents to devalue the environ-
ment (Schultz, Zelezny, & Dalrymple, 2000; White, 1967) or that beliefs that the
environment is sacred enhance environmental concern (e.g., Dietz et al., 1998;
Greeley, 1993; Kempton, Boster, & Hartley, 1995). Others have linked environ-
mental concern and behavior to general theories of values (e.g., Schwartz, 1994)
and have found that values those that focus concern beyond a person’s immediate
social circle (values called self-transcendent or altruistic) are stronger among peo-
ple who engage in proenvironmental activities (e.g., Dietz et al., 1998; Karp, 1996;
Stern & Dietz, 1994; Stern, Dietz, Kalof, & Guagnano, 1995). A related line of
research finds greater evidence of environmental concern among individuals with
412 Stern

“prosocial” rather than individualistic or competitive social value orientations

(e.g., Joireman, Lasane, Bennett, Richards, & Solaimani, in press; Van Vugt &
Samuelson, 1998).
Theories of altruistic behavior have also been used to explain environmental-
ism. This approach, first articulated by Heberlein (1972), presumes that because
environmental quality is a public good, altruistic motives are a necessary for an
individual to contribute to it in a significant way. The best developed example of
this approach builds on Schwartz’s (1973, 1977) moral norm-activation theory of
altruism. The theory holds that altruistic (including proenvironmental) behavior
occurs in response to personal moral norms that are activated in individuals who
believe that particular conditions pose threats to others (awareness of adverse
consequences, or AC) and that actions they could initiate could avert those conse-
quences (ascription of responsibility to self, or AR). Substantial evidence support-
ing the theory’s applicability to a range of environmental issues has accumulated
over two decades (e.g., Black, 1978; Black et al., 1985; Guagnano, Stern, & Dietz,
1995; Schultz & Zelezny, 1999; Widegren, 1998).
My colleagues and I have developed a value-belief-norm (VBN) theory of
environmentalism that builds on some of the above theoretical accounts and
offers what we believe to be the best explanatory account to date of a variety of
behavioral indicators of nonactivist environmentalism (Stern et al., 1999). The
theory links value theory, norm-activation theory, and the New Environmental
Paradigm (NEP) perspective through a causal chain of five variables leading to
behavior: personal values (especially altruistic values), NEP, AC and AR beliefs
about general conditions in the biophysical environment, and personal norms for
proenvironmental action (see Figure 1). The rationale and empirical support for
this causal ordering is drawn from previous work (Black et al., 1985; Gardner &

Fig. 1. A schematic representation of variables in the VBN theory of environmentalisma

a
Arrows represent postulated direct effects. Direct effects may also be observed on variables more than
one level downstream from a causal variable.
b
Empirically, measures of egoistic values have been negatively correlated with indicators of environ-
mentalism.
Theory of Environmentally Significant Behavior 413

Stern, 1996; Stern, Dietz, & Guagnano, 1995; Stern, Dietz, Kalof, & Guagnano,
1995; Stern & Oskamp, 1987). The causal chain moves from relatively stable,
central elements of personality and belief structure to more focused beliefs about
human-environment relations (NEP), their consequences, and the individual’s
responsibility for taking corrective action. We postulate that each variable in the
chain directly affects the next and may also directly affect variables farther down
the chain. Personal norms to take proenvironmental action are activated by
beliefs that environmental conditions threaten things the individual values (AC)
and that the individual can act to reduce the threat (AR). Such norms create a
general predisposition that influences all kinds of behavior taken with pro-
environmental intent. In addition, behavior-specific personal norms and other
social-psychological factors (e.g., perceived personal costs and benefits of
action, beliefs about the efficacy of particular actions) may affect particular
proenvironmental behaviors, as discussed below.
The VBN theory links value theory to norm-activation theory by generalizing
the latter. It postulates that the consequences that matter in activating personal
norms are adverse consequences to whatever the individual values (AC). Thus,
people who value other species highly will be concerned about environmental
conditions that threaten those valued objects, just as altruists who care about other
people will be concerned about environmental conditions that threaten the other
people’s health or well-being. VBN theory links the NEP to norm-activation
theory with the argument that the NEP is a sort of “folk” ecological theory from
which beliefs about the adverse consequences of environmental changes can be
deduced (for empirical support, see Stern, Dietz, & Guagnano, 1995).
In a recent study (Stern et al., 1999), my colleagues and I used the VBN theory,
as well as measures from three other theories (indicators of four cultural biases,
postmaterialist values, and belief in the sacredness of nature), to account for three
types of nonactivist environmentalism: environmental citizenship, private-sphere
behavior, and policy support (willingness to sacrifice). The VBN cluster of vari-
ables was a far stronger predictor of each behavioral indicator than the other theo-
ries, even when the other theories were taken in combination (see Table 1). None of
the theories, however, was very successful in predicting the sole indicator of activ-
ism (participation in an environmental demonstration), which appears to depend
on other factors in addition to an environmentalist predisposition.
The results provide strong initial support for the VBN theory’s contentions
that personal moral norms are the main basis for individuals’ general predisposi-
tions to proenvironmental action (other studies supporting this conclusion include
Bratt, 1999b, and Widegren, 1998) and that these norms are activated as the theory
specifies. The personal norm variable was the only psychological variable of the 14
in the study that is associated with all three types of nonactivist environmentalism
when the other variables are held constant. Moreover, values, NEP, and AC beliefs
accounted for 56% of the variance in personal norms.
414 Stern

Table 1. Explained variance in Three Indicators of Proenvironmental Behavior

Dependent measures
Source of explanatory Private-sphere Policy Environmental
variables behavior support citizenship

VBN theory .194 .346 .302

Three other theoriesa .094 .199 .187
Added variance from other theoriesb .033 .033 .091
Note. From “A Value-Belief-Norm Theory of Support for Social Movements: The Case of Environmen-
tal Concern,” by P. C. Stern, T. Dietz, T. Abel, G. A. Guagnano, and L. Kalof, 1999, Human Ecology
Review, 6, p. 90. Copyright 1999 by Society for Human Ecology. Reprinted with permission.
aPostmaterialist values, four cultural biases, and beliefs about the sacredness of nature.
bDifference between R2 value for model combining VBN theory variables with the variables from the
other three theories and value for model with VBN theory alone.

Data from several studies indicate that the values most strongly implicated in
activating proenvironmental personal norms are, as norm-activation theory pre-
sumes, altruistic or self-transcendent values (Karp, 1996; Stern, Dietz, Kalof, &
Guagnano, 1995; Stern et al., 1999). However, other values are sometimes linked
as well. Self-enhancement or egoistic values and “traditional” values such as
obedience, self-discipline, and family security are negatively associated with
proenvironmental norms and action in some studies. The ways these values affect
behavior are not well understood, but they may be important bases for principled
opposition by some individuals to environmental movement goals. Another poten-
tially important issue, as yet unresolved empirically, is whether a set of biospheric
values is emerging, distinct from altruistic values about other people, that might
provide a distinct basis for people’s support for preserving endangered species and
habitats.
An important element of the VBN theory is that the link from values to envi-
ronmentalism is mediated by particular beliefs, such as beliefs about which kinds
of people or things are affected by environmental conditions (AC) and about
whether there are individual actions that could alleviate threats to valued persons or
things (AR). Thus, environmentalist personal norms and the predisposition to
proenvironmental action can be influenced by information that shapes these
beliefs. This proposition suggests how environmentalism can be affected by the
findings of environmental science (about consequences), publicity and commen-
tary about those findings, and the actual and perceived openness of the political
system to public influence (which may affect perceptions of personal responsibil-
ity). It also suggests an interpretation of environmentalist and antienvironmentalist
rhetoric as efforts to activate or deactivate people’s environmental norms by high-
lighting certain kinds of values or consequences (Stern, Dietz, Kalof, & Guagnano,
1995). The VBN theory offers an account of attitude formation that can deal with
new or changing attitude objects (Stern, Dietz, Kalof, & Guagnano, 1995) and,
Theory of Environmentally Significant Behavior 415

more generally, with how environmental concern and environmental issues are
socially constructed (Dietz, Stern, & Rycroft, 1989). The VBN theory is thus com-
patible with the constructed-preference tendency in cognitive psychology (Dietz &
Stern, 1995; Fischhoff, 1991; Payne, Bettman, & Johnson, 1992).

The Causes of Environmentally Significant Behavior

Because environmental intent and environmental impact are two different

things, theories explaining environmentalism are necessarily insufficient for
understanding how to change environmentally important behaviors. Environmen-
talist intent is only one of the factors affecting behavior, and often, it is not one of
the most important. Many environmentally significant behaviors are matters of
personal habit or household routine (e.g., the setting of thermostats or the brand of
paper towels purchased) and are rarely considered at all. Others are highly con-
strained by income or infrastructure (e.g., reinsulating homes, using public trans-
port). For others, environmental factors are only minor influences on major actions
(e.g., choosing an engine size option in a new automobile, deciding whether to
centrally air condition a home), or the environmental effects are unknown to the
consumer (e.g., choosing between products that have different environmental
impacts from their manufacturing processes). Sometimes, as with spray cans, peo-
ple may act in ways that are proenvironmental in intent but that in fact have little or
no positive environmental impact. Environmentally beneficial actions may also
follow from nonenvironmental concerns, such as a desire to save money, confirm a
sense of personal competence, or preserve time for social relationships (De Young,
this issue). And environmental concerns may fail to lead to proenvironmental
action for various reasons (Gardner & Stern, 1996; Kempton, 1993). To under-
stand any specific environmentally significant behavior requires empirical analy-
sis. The evidence suggests that the role of environmentalist predispositions can
vary greatly with the behavior, the actor, and the context.

ABC Theory

A first step toward understanding the complexities is to elaborate on the truism

that behavior is a function of the organism and its environment. In one formulation
(Guagnano et al., 1995), behavior (B) is an interactive product of personal-sphere
attitudinal variables (A) and contextual factors (C). The attitude-behavior associa-
tion is strongest when contextual factors are neutral and approaches zero when
contextual forces are strongly positive or negative, effectively compelling or
prohibiting the behavior in question (an inverted U-shaped function). We found
supportive evidence for this formulation in a study of curbside recycling
(Guagnano et al., 1995).
416 Stern

This “ABC theory” formulation implies that for personal behaviors that are
not strongly favored by context (e.g., by being required or tangibly rewarded), the
more difficult, time-consuming, or expensive the behavior, the weaker its depen-
dence on attitudinal factors. Supporting evidence for this implication exists in stud-
ies that have used the same attitudinal variables to account for different
proenvironmental behaviors. For example, in a study of household energy conser-
vation, the relative explanatory power of social-psychological variables declined
as effort or cost increased, from 59% of the explainable variance in self-reported
home thermostat settings to 50% for minor curtailments such as shutting off heat in
unused rooms, 44% for low-cost energy efficiency improvements such as caulking
and weather-stripping, and 25% for major investments such as adding insulation or
storm windows (Black et al., 1985). There are similar findings for public-sphere
behaviors. The social-psychological variables of the VBN theory accounted for
35% of the variance in expressed policy support for environmentalism and 30% of
the variance in environmental citizenship behaviors but only 4% of the variance in
committed activism (Stern et al., 1999). These findings suggest a provocative
hypothesis that is worthy of further exploration, namely that the more important a
behavior is in terms of its environmental impact, the less it depends on attitudinal
variables, including environmental concern.

Four Types of Causal Variables

It is useful to refine the personal-contextual or organism-environment distinc-

tion and to group the causal variables into four major types. Attitudinal factors,
including norms, beliefs, and values, are one. The VBN theory provides a good
theoretical account of one such factor, the general predisposition to act with
proenvironmental intent, which can influence all behaviors an individual considers
to be environmentally important. Other attitudinal variables affect only certain
environmentally relevant behaviors. These include behavior-specific predisposi-
tions (e.g., specific personal moral norms in the terms of norm-activation theory,
attitudes toward acts in the terms of the theory of planned behavior) and behavior-
specific beliefs (e.g., about the difficulty of taking certain actions or about their
consequences for self, others, or the environment). Several social-psychological
theories, including cognitive dissonance theory, norm-activation theory, and the
theory of planned behavior, have been shown to explain variance in specific
proenvironmental behaviors. This research has demonstrated that pro-
environmental behaviors can be affected by personal commitment and the per-
ceived personal costs and benefits of particular actions (e.g., Katzev & Johnson,
1987) as well as by behavior-specific beliefs and personal norms (e.g., Black et al.,
1985). As already noted, environmentally significant behavior can also be affected
by nonenvironmental attitudes, such as those about attributes of consumer prod-
ucts that are correlated with environmental impact (e.g., speed, power, and luggage
Theory of Environmentally Significant Behavior 417

capacity in motor vehicles), or about frugality, luxury, waste, or the importance of

spending time with family.
A second major type of causal variable is external or contextual forces. These
include interpersonal influences (e.g., persuasion, modeling); community expecta-
tions; advertising; government regulations; other legal and institutional factors
(e.g., contract restrictions on occupants of rental housing); monetary incentives
and costs; the physical difficulty of specific actions; capabilities and constraints
provided by technology and the built environment (e.g., building design, availabil-
ity of bicycle paths, solar energy technology); the availability of public policies to
support behavior (e.g., curbside recycling programs); and various features of the
broad social, economic, and political context (e.g., the price of oil, the sensitivity
of government to public and interest group pressures, interest rates in financial
markets). It is worth nothing that a contextual factor may have different meanings
to people with different attitudes or beliefs. For example, the higher price of
“organic” produce may be an economic barrier to purchase for some people,
whereas for others it is a marker of a superior product.
Personal capabilities are a third type of causal variable. These include the
knowledge and skills required for particular actions (e.g., the skills of a movement
organizer for activism, mechanical knowledge for energy-conserving home
repairs), the availability of time to act, and general capabilities and resources such
as literacy, money, and social status and power. Sociodemographic variables such
as age, educational attainment, race, and income may be indicators or proxies for
personal capabilities. Although these variables have very limited explanatory
power for many environmentally significant behaviors (e.g., Dietz et al., 1998),
they may be important for behaviors that depend strongly on particular capabili-
ties. For instance, in a recent study (Stern et al., 1999), sociodemographic variables
were found to be unrelated to consumer behavior and policy support when
social-psychological variables were held constant, but environmental citizenship
was found to be positively associated with income and with White race. The find-
ings reflect the fact that the efficacy of environmental citizenship depends on an
individual’s social and economic resources. Also, environmental activism, for
which attitudinal variables had very little explanatory power, was significantly
associated (negatively) with age and income.
Finally, habit or routine is a distinct type of causal variable. Behavior change
often requires breaking old habits and becomes established by creating new ones
(Dahlstrand & Biel, 1997). Habit, in the form of standard operating procedure, is
also a key factor in environmentally significant organizational behavior.
The evidence suggests that different types of causal variables are important,
depending on the particular behavior (Gardner & Stern, 1996; Stern, 2000). Expen-
sive behaviors such as reinsulating homes are likely to be strongly influenced by
monetary factors; difficult behaviors such as reducing automobile use in the sub-
urbs are likely to be strongly influenced by public policy supports (e.g., for
418 Stern

alternative transport modes); behaviors that require specialized skills are likely to
be strongly influenced by whether or not one possesses those capabilities; and so
forth. Such hypotheses, though fairly obvious, do not go without saying. They offer
a good starting point for efforts to understand particular environmentally signifi-
cant behaviors.
Different causal variables also appear to work different ways in influencing
behavior. For example, certain attitudinal factors create a general predisposition to
act, which may be shaped into specific action largely by personal capabilities and
contextual forces. A new context may make old habits untenable and lead someone
to consider his or her attitudes and values explicitly in developing new ones
(Dahlstrand & Biel, 1997). Or financial incentives may favor behaviors that never-
theless do not occur unless information makes individuals aware that the incentive
is available (Stern, 1999).
The insight of the ABC formulation, that the different types of causal factors
may interact, implies that interpretations based only on main effects can be seri-
ously misleading. Studies that examine only attitudinal factors are likely to find
effects only inconsistently, because the effects are contingent on capabilities and
context. Similarly, studies that examine only contextual variables, such as material
incentives, social norms, or the introduction of new technology, may find effects
but fail to reveal their dependence on individuals’ attitudes or beliefs. Single-
variable studies may demonstrate that a particular theoretical framework has
explanatory power but may not contribute much to the comprehensive understand-
ing of particular environmentally significant behaviors that is needed to change
them. I return to this point later.

Toward a Synthesis

The field now needs synthetic theories or models that incorporate variables
from more than one of the above broad classes, postulate relationships among
them, and use them to explain one or more types of environmentally significant
behavior. Researchers are beginning to propose such models (e.g., Dahlstrand &
Biel, 1997; Fransson & Gärling, 1999; Gardner & Stern, 1996; Hines, Hungerford,
& Tomera, 1987; Ölander & Thøgerson, 1995; Stern & Oskamp, 1987; Vlek,
2000). Some of the models expand on familiar theories of altruistic behavior (e.g.,
Schwartz, 1977) or planned behavior (e.g., Ajzen, 1991), which emphasize attitu-
dinal factors almost exclusively. Because the new models also take into account
personal capabilities, context, and habits, they are more suitable for explaining
behaviors that have significant environmental impacts, which are often strongly
influenced by such nonattitudinal factors.
A dialogue among such models is needed to move the field toward synthesis. It
is also likely to build links to other psychological theories. For example, the
distinction between attitudes and habits as causes of behavior closely parallels the
Theory of Environmentally Significant Behavior 419

distinction in a variety of “dual-process” models (Smith & DeCoster, 2000)

between conscious and effortful behaviors and automatic or associative ones.
Dual-process models may therefore have something to say about pro-
environmental behavior.

Changing Environmentally Significant Behavior

Many approaches toward changing individuals’ environmentally significant

behavior have been tried. Gardner and Stern (1996) reviewed the evidence on four
major types of intervention: religious and moral approaches that appeal to values
and aim to change broad worldviews and beliefs; education to change attitudes and
provide information; efforts to change the material incentive structure of behavior
by providing monetary and other types of rewards or penalties; and community
management, involving the establishment of shared rules and expectations. They
found that each of these intervention types, if carefully executed, can change
behavior. However, moral and educational approaches have generally disappoint-
ing track records, and even incentive- and community-based approaches rarely
produce much change on their own. By far, the most effective behavior change
programs involve combinations of intervention types.
These findings underline the limits of single-variable explanations for inform-
ing efforts at behavior change. The behavior is determined by multiple variables,
sometimes in interaction. There is strong evidence, for example, that incentives
and information interact, with the combination sometimes being much more effec-
tive than the sum of the two interventions (Stern, 1999). In one evaluation study,
increased financial incentives for major investments in home energy conservation
were necessary but far from sufficient for programs to be successful. Even when
electric utility companies offered to subsidize 93% of the cost of home insulation,
consumer response varied from 1% to almost 20% adoption per year, apparently
depending on how the subsidy was made known to householders (Stern et al.,
1986).
Often the nature of the interaction can be well described in terms of barriers or
limiting conditions to behavior change (Gardner & Stern, 1996). Interventions do
little or nothing until one of them removes an important barrier to change. To pro-
mote investments in home insulation, for example, it is necessary to reduce the
financial barriers, provide accurate information on which actions would be effec-
tive, and reduce the difficulty of getting the information and finding a reliable
contractor. Programs that did all these things were vastly more successful than pro-
grams that did only one or two (Stern et al., 1986). Since different individuals face
different impediments to behavior change and the impediments are often multiple,
little happens until the right combination of intervention types is found. The con-
cept of limiting conditions also implies that particular kinds of interventions have
diminishing returns after they have fulfilled their major function. For example,
420 Stern

once financial incentives are large enough to demonstrate a clear personal benefit,
increasing the incentive may be far less effective in producing behavior change
than providing information through marketing (see Stern, 1999).
Theory has progressed to the point at which it is possible to identify useful and
practical principles for intervention (see Table 2; for a guide to the application of
these principles, see McKenzie-Mohr & Smith, 1999). Space does not permit elab-
oration of all the principles here. The admonitions to combine multiple interven-
tion types, to understand the situation from the actor’s perspective, to continually
monitor and adjust programs, and to use participatory methods all suggest ways to
make practical progress with incomplete theory.
For researchers who would like to advance the understanding necessary to
make behavioral approaches to environmental protection more successful, a related
set of principles applies (see Gardner & Stern, 1996, chap. 10). First, identify target
behaviors that are environmentally significant in terms of impact. Then analyze the
behaviors to identify the responsible actors and actions. Then consider the full range
of causal variables and explore their possible relevance to the target behavior from
the actor’s standpoint. By exploring the possibilities directly with representatives of
the population whose behavior is to be changed, it is possible to find promising
strategies for intervention without trying them all out experimentally.
This research strategy offers the best approach to developing useful theory
about specific behavioral types that have important environmental impacts. In
addition to its practical value, such small-scale theory provides the essential build-
ing blocks for broader, inductively developed theory about environmentally signif-
icant behavior.
Table 2. Principles for Intervening to Change Environmentally Destructive Behavior
A. Use multiple intervention types to address the factors limiting behavior change
1. Limiting factors are numerous (e.g., technology, attitudes, knowledge, money,
convenience, trust)
2. Limiting factors vary with actor and situation, and over time
3. Limiting factors affect each other
B. Understand the situation from the actor’s perspective
C. When limiting factors are psychological, apply understanding of human choice processes
1. Get the actors’ attention; make limited cognitive demands
2. Apply principles of community management (credibility, commitment, face-to-face
communication, etc.)
D. Address conditions beyond the individual that constrain proenvironmental choice
E. Set realistic expectations about outcomes
F. Continually monitor responses and adjust programs accordingly
G. Stay within the bounds of actors’ tolerance for intervention
H. Use participatory methods of decision making
Note. From Environmental Problems and Human Behavior (p. 159), by G. T. Gardner and P. C. Stern,
1996, Boston: Allyn and Bacon. Copyright 1996 by Allyn and Bacon. Reprinted with permission.
Theory of Environmentally Significant Behavior 421

Conclusions

Environmentally significant behavior is dauntingly complex, both in its vari-

ety and in the causal influences on it. Although a general theory lies far in the dis-
tance, enough is known to present a framework that can increase theoretical
coherence. This framework includes typologies of environmentally significant
behaviors and of their causes (see Table 3) and a growing set of empirical proposi-
tions about these variables. For example:
• The VBN approach offers a good account of the causes of the general
predisposition toward proenvironmental behavior.
• Environmentally significant behavior depends on a broad range of
causal factors, both general and behavior-specific. A general theory of
environmentalism may therefore not be very useful for changing
specific behaviors.
• Different kinds of environmentally significant behavior have different
causes. Because the important causal factors may vary greatly across
behaviors and individuals, each target behavior should be theorized
separately.

Table 3. Major Types of Environmentally Significant Behaviors and Causal Variables

Influencing These Behaviors
Causal variables Environmentally significant behaviors

Attitudinal Environmental activism

General environmentalist predispositiona Nonactivist public-sphere behaviors
Behavior-specific norms and beliefsb Environmental citizenship
Nonenvironmental attitudes (e.g., petitioning, joining groups)
(e.g., about product attributes) Policy support
Perceived costs and benefits of action
Personal capabilities Private-sphere environmentalism
Literacy Consumer purchase behaviors
Social status Maintenance of household equipment
Financial resources Changes in equipment use, lifestyle (curtailment)
Behavior-specific knowledge and skills Waste disposal behaviors
“Green consumerism”
Contextual factors Other
Material costs and rewards Behaviors affecting organizational decisions
Laws and regulations
Available technology
Social norms and expectations
Supportive policies
Advertising
Habit and routine
aThe VBN theory incorporates various attitudinal variables believed to create this predisposition.
bThese norms and beliefs figure prominently in applications of norm-activation theory and the theory of
planned behavior to specific proenvironmental behaviors.
422 Stern

• The causal factors may interact. Attitudinal causes have the greatest
predictive value for behaviors that are not strongly constrained by con-
text or personal capabilities. For behaviors that are expensive or diffi-
cult, contextual factors and personal capabilities are likely to account
for more of the variance.
In addition to such empirical principles, past research has yielded important
insights for research and action on environmental protection, as described above
and in Table 2. One cannot overemphasize to behavioral scientists the importance
of identifying target behaviors from an environmental perspective (in terms of their
impact), even though understanding them requires an actor-oriented approach that
focuses on their causes. It is also critical to underscore the need to draw on insights
from across the behavioral and social sciences, because the important causal vari-
ables lie in the domains of various disciplines and because the variables interact.
Thus, interdisciplinary research is necessary for full understanding.
By following these insights and elaborating on the above principles, behav-
ioral researchers can further advance understanding of environmentally significant
individual behavior and can provide useful input to practical programs for environ-
mental protection. They are also likely to make contributions to the broader project
of behavioral science.

References

Ajzen, I. (1991). The theory of planned behavior. Organizational Decision and Human Decision Pro-
cess, 50, 179–211.
Allen, J. B., & Ferrand, J. L. (1999). Environmental locus of control, sympathy, and proenvironmental
behavior: A test of Geller’s actively caring hypothesis. Environment and Behavior, 31, 338–353.
Black, J. S. (1978). Attitudinal, normative, and economic factors in early response to an energy-use
field experiment (Doctoral dissertation, University of Wisconsin, 1978). Dissertation Abstracts
International, 39, 436B.
Black, J. S., Stern, P. C., & Elworth, J. T. (1985). Personal and contextual influences on household
energy adaptations. Journal of Applied Psychology, 70, 3–21.
Bratt, C. (1999a). Consumers’ environmental behavior: Generalized, sector-based, or compensatory?
Environment and Behavior, 31, 28–44.
Bratt, C. (1999b). The impact of norms and assumed consequences on recycling behavior. Environment
and Behavior, 31, 630–656.
Dahlstrand, U., & Biel, A. (1997). Pro-environmental habits: Propensity levels in behavioral change.
Journal of Applied Social Psychology, 27, 588–601.
Dake, K. (1991). Orienting dispositions in the perception of risk: An analysis of contemporary
worldviews and cultural biases. Journal of Cross-Cultural Psychology, 22, 61–82.
Dietz, T., & Stern, P. C. (1995). Toward a theory of choice: Socially embedded preference construction.
Journal of Socio-Economics, 24, 261–279.
Dietz, T., Stern, P. C., & Guagnano, G. A. (1998). Social structural and social psychological bases of
environmental concern. Environment and Behavior, 30, 450–471.
Dietz, T., Stern, P. C., & Rycroft, R. W. (1989). Definitions of conflict and the legitimation of
resources: The case of environmental risk. Sociological Forum, 4, 47–70.
Douglas, M., & Wildavsky, A. (1982). Risk and culture: An essay on the selection of technological and
environmental dangers. Berkeley and Los Angeles: University of California Press.
Theory of Environmentally Significant Behavior 423

Fischhoff, B. (1991). Preference elicitation: Is there anything in there? American Psychologist, 46,
835–847.
Fransson, N., & Gärling, T. (1999). Environmental concern: Conceptual definitions, measurement
methods, and research findings. Journal of Environmental Psychology, 19, 369–382.
Gardner, G. T., & Stern, P. C. (1996). Environmental problems and human behavior. Boston: Allyn and
Bacon.
Greeley, A. (1993). Religion and attitudes toward the environment. Journal for the Scientific Study of
Religion, 32, 19–28.
Guagnano, G. A., Stern, P. C., & Dietz, T. (1995). Influences on attitude-behavior relationships: A natu-
ral experiment with curbside recycling. Environment and Behavior, 27, 699–718.
Heberlein, T. A. (1972). The land ethic realized: Some social psychological explanations for changing
environmental attitudes. Journal of Social Issues, 28(4), 79–87.
Hines, J. M., Hungerford, H. R., & Tomera, A. N. (1987). Analysis and synthesis of research on respon-
sible environmental behavior: A meta-analysis. Journal of Environmental Education, 18, 1–18.
Inglehart, R. (1990). Culture shift in advanced industrial society. Princeton, NJ: Princeton University
Press.
Joireman, J. A., Lasane, T. P., Bennett, J., Richards, D., & Solaimani, S. (in press). Integrating social
value orientation and the consideration of future consequences within the extended norm activa-
tion model of proenvironmental behavior. British Journal of Social Psychology.
Kals, E., Schumacher, D., & Montada, L. (1999). Emotional affinity toward nature as a motivational
basis to protect nature. Environment and Behavior, 31, 178–202.
Karp, D. G. (1996). Values and their effects on pro-environmental behavior. Environment and Behav-
ior, 28, 111–133.
Katzev, R. D., & Johnson, T. R. (1987). Promoting energy conservation: An analysis of behavioral
techniques. Boulder, CO: Westview Press.
Kempton, W. (1993). Will public environmental concern lead to action on global warming? Annual
Review of Energy and Environment, 18, 217–245.
Kempton, W., Boster, J. S., & Hartley, J. A. (1995). Environmental values in American culture. Cam-
bridge, MA: MIT Press.
McAdam, D., McCarthy, J. D., & Zald, M. N. (1988). Social movements. In N. J. Smelser (Ed.), Hand-
book of sociology (pp. 695–738). Newbury Park, CA: Sage.
McKenzie-Mohr, D., & Smith, W. (1999). Fostering sustainable behavior: An introduction to commu-
nity-based social marketing. Gabriola Island, British Columbia, Canada: New Society Publishers.
Ölander, F., & Thøgerson, J. (1995). Understanding consumer behavior as a prerequisite for environ-
mental protection. Journal of Consumer Policy, 18, 345–385.
Payne, J. W., Bettman, J. R., & Johnson, E. J. (1992). Behavioral decision research: A constructive pro-
cessing perspective. Annual Review of Psychology, 43, 87–131.
Rosa, E. A., & Dietz, T. (1998). Climate change and society: Speculation, construction and scientific
investigation. International Sociology, 13, 421–425.
Schultz, P. W., & Zelezny, L. C. (1999). Values as predictors of environmental attitudes: Evidence for
consistency across cultures. Journal of Environmental Psychology, 19, 255–265.
Schultz, P. W., Zelezny, L. C., & Dalrymple, N. J. (2000). A multinational perspective on the relation
between Judeo-Christian religious beliefs and attitudes of environmental concern. Environment
and Behavior, 32, 576–591.
Schwartz, S. H. (1973). Normative explanations of helping behavior: A critique, proposal, and empiri-
cal test. Journal of Experimental Social Psychology, 9, 349–364.
Schwartz, S. H. (1977). Normative influences on altruism. In L. Berkowitz (ed.), Advances in experi-
mental social psychology (Vol. 10, pp. 221–279). New York: Academic Press.
Schwartz, S. H. (1994). Are there universal aspects in the structure and contents of human values? Jour-
nal of Social Issues, 50(4), 19–46.
Smith, E. R., & DeCoster, J. (2000). Dual-process models in social and cognitive psychology: Concep-
tual integration and links to underlying memory systems. Personality and Social Psychology
Review, 4, 108–131.
Steg, L., & Sievers, I. (2000). Cultural theory and individual perceptions of environmental risks. Envi-
ronment and Behavior, 332, 250–269.
424 Stern

Stern, P. C. (1997). Toward a working definition of consumption for environmental research and policy.
In P. C. Stern, T. Dietz, V. R. Ruttan, R. H. Socolow, & J. L. Sweeney (Eds.), Environmentally
significant consumption: Research directions (pp. 12–35). Washington, DC: National Academy
Press, 1997.
Stern, P. C. (1999). Information, incentives, and proenvironmental consumer behavior. Journal of
Consumer Policy, 22, 461–478.
Stern, P. C. (2000). Psychology, sustainability, and the science of human-environment interactions.
American Psychologist, 55, 523–530.
Stern, P. C., Aronson, E., Darley, J. M., Hill, D. H., Hirst, E., Kempton, W., & Wilbanks, T. J. (1986).
The effectiveness of incentives for residential energy conservation. Evaluation Review, 10(2),
147–176.
Stern, P. C., & Dietz, T. (1994). The value basis of environmental concern. Journal of Social Issues,
50(3), 65–84.
Stern, P. C., Dietz, T., Abel, T., Guagnano, G. A., & Kalof, L. (1999). A value-belief-norm theory of sup-
port for social movements: The case of environmental concern. Human Ecology Review, 6, 81–97.
Stern, P. C., Dietz, T., & Guagnano, G. A. (1995). The new environmental paradigm in social psycho-
logical perspective. Environment and Behavior, 27, 723–745.
Stern, P. C., Dietz, T., Kalof, L., & Guagnano, G. A. (1995). Values, beliefs and proenvironmental
action: Attitude formation toward emergent attitude objects. Journal of Applied Social Psychol-
ogy, 25, 1611–1636.
Stern, P. C., & Gardner, G. T. (1981a). Psychological research and energy policy. American Psycholo-
gist 36, 329–342.
Stern, P. C., & Gardner, G. T. (1981b). The place of behavior change in managing environmental prob-
lems. Zeitschrift für Umweltpolitik, 2, 213–239.
Stern, P. C., & Oskamp, S. (1987). Managing scarce environmental resources. In D. Stokols &
I. Altman (Eds.), Handbook of environmental psychology (pp. 1043–1088). New York: Wiley.
Stern, P. C., Young, O. R., & Druckman, D. (Eds.). (1992). Global environmental change: Under-
standing the human dimensions. Washington, DC: National Academy Press.
Van Vugt, M., & Samuelson, C. D. (1998). The impact of personal metering in the management of a
natural resource crisis: A social dilemma analysis. Personality and Social Psychology Bulletin,
25, 731–745.
Vayda, A. P. (1988). Actions and consequences as objects of explanation in human ecology. In R. J.
Borden, J. Jacobs, & G. L. Young (Eds.), Human ecology: Research and applications (pp.
9–18). College Park, MD: Society for Human Ecology.
Vlek, C. (2000). Essential psychology for environmental policy making. International Journal of
Psychology, 35, 153–167.
White, L., Jr. (1967). The historical roots of our ecological crisis. Science, 155, 1203–1207.
Widegren, Ö. (1998). The new environmental paradigm and personal norms. Environment and Behav-
ior, 30, 75–100.
Zald, M. (1992). Looking backward to look forward: Reflections on the past and future of the resource
mobilization research program. In A. D. Morris & C. M. Mueller (Eds.), Frontiers in social
movement theory (pp. 326–348). New Haven, CT: Yale University Press.

PAUL C. STERN is Study Director of the Committee on the Human Dimensions

of Global Change at the U.S. National Research Council. He is also a Research
Professor of Sociology at George Mason University and President of the Social and
Environmental Research Institute. His current research interests include the study
of environmental values, beliefs, and behavior and the development of deliberative
approaches to environmental decision making. Recent publications include the
coedited volumes Understanding Risk: Informing Decisions in a Democratic
Society (1996), Environmentally Significant Consumption: Research Directions
(1997), Making Climate Forecasts Matter (1999), and International Conflict Res-
olution After the Cold War (2000), all published by National Academy Press.