Support for Environmental Protection: The Role of Moral Norms
Paul C. Stern National Research Council Thomas Dietz George Mason University I. Stanley Black Illinois Environmental Protection Agency

A theoretical model is presented that traces support for environmental protection to a social-psychological process involving the activation of moral norms against harming innocent people. In a preliminary test of the model's social-psychological hypotheses, judgments about the moral obligations of industry with respect to hazardous chemicals were found to depend both on awareness of harmful consequences to people and on ascription of responsibility for those consequences to industry; government, however, was held to be morally obligated to act even if it was not responsible for the harm. Suggestions are offered for using the model to study the determinants of changing public opinion on the environment, the tactics of advocacy groups in environmental policy conflicts, and the process that mobilizes pressure for political causes in the absence of tangible group interests. The last two decades have brought about substantial changes in the environmental policies of most Western industrial nations. These changes are due in large part to the effective mobilization of public support by the enviThe authors thank Jane Phillips for assistance with data analysis and the Academic Computing Service and Graduate School of George Mason University for research support. Requests for reprints should be sent to Paul C. Stern, National Research Council, 2101 Constitutional Avenue, N.W., Washington, D.C. 20418. 204 Population and Environment, Volume 8, Number 3 & 4, Fall/Winter 1985-86 © 1986 Human Sciences Press

205 PAUL C. STERN ET AL.

ronmental movement. Such mobilization presents an interesting problem to social scientists in that simple models of self-interested behavior cannot easily explain active support for social movements seeking to promote collective goals (Olson, 1965). Because the accomplishments of such movements benefit most members of a society and the benefits accrue to an individual whether or not he or she actively participates in the movement, simple rational actor models imply that few people participate in environmental organizations or in such activities as boycotting polluting products or lobbying for environmental legislation. Nevertheless, active support for environmental groups and their goals is widespread (e.g., Dunlap, 1985; Morrison & Dunlap, 1986), and participants contribute more to these groups than they receive in individual benefits (Mitchell, 1980). The fact of widespread support for environmental protection suggests that motives other than the short-term and the individualistic impel many people to act. This paper suggests that one such motive is provided by a judgment that pollution is, to put it bluntly, morally wrong. This paper presents a theoretical model of support for environmental protection that describes this support in relation to individuals' moral judgments about pollution. It also provides a preliminary test of the social-psychological portion of that model. This section of the paper describes the general model; the next section details the operation of the model at the individual level. We then present a preliminary test of the model. The final section considers some implications of the model and some possible directions for further research. Social scientists have devoted considerable study to the process by which public opinion is formed and by which citizens attempt to influence public policy on environmental and other issues. They have developed knowledge of the socioeconomic and demographic roots of support for and participation in the environmental movement (Morrison & Dunlap, 1986; Van Liere & Dunlap, 1980); the path by which scientific information filters down from technical journals into the media before public concern becomes aroused (Kates, 1978); the ways the framing of political issues can influence public opinion (e.g., Gamson & Modigliani, 1985; Stern & Aronson, 1984: Chapter 2); the conditions under which interest groups get mobilized (e.g., Gamson, 1975; McCarthy & Zald, 1977); and the ways environmental laws and regulations are affected by the process of political agenda-setting and the power relations that affect it (e.g., Crenson, 1971; Melnick, 1984). Such work provides some insight into how scientific information gets citizens' attention; it also provides analysis of the ways citizens' organized demands to codify norms for environmental protection are--or are not --answered in the political system. But the social-psychological part-- the process by which people who have been exposed to information and rheto-

206 POPULATION AND ENVIRONMENT

ric come to demand that the state protect the environment--has usually been overlooked. For people to demand environmental legislation and regulation, the idea of environmental hazard must enter their awareness and be transformed into attitudes, judgments, and actions. We propose that support for environmental protection depends in part on a moral judgment, that supporters of demands for environmental protection see environmental problems not only as unfortunate situations but as morally intolerable. This interpretation makes sense of two apparent facts: 1) that the public demands regulation most strongly when faced with a severe and uncontrolled threat to the health, safety, or general welfare of large numbers of people, especially people who previously had no reason to suspect the hazard (see, for example, Slovic, Fischhoff, & Lichtenstein, 1982); and 2) that the clamor sounds loudest when it is also possible to identify the sources or parties directly responsible for the hazard. These two factors, perception of a threat to innocents and identification of responsible parties, figure prominently in social-psychological research on the activation of altruistic norms (Schwartz, 1970; 1977). Although this research recognizes that behavior is determined by a multiplicity of factors, including relevant costs, personal beliefs and attitudes, social pressures, and other constraints, two factors have emerged as particularly important in determining whether normative feelings of right or wrong will be important in guiding behavior. The perceived severity of interpersonal consequences of an action is of primary importance. Awareness of severe and/or widespread harmful consequences (AC) to other people tends to activate the feeling that action should be taken to avert or alleviate the harm. The other important element is the ascription of responsibility (AR) for the remedial action, that is, whether a person judges himself or herself personally responsible for the positive or negative outcome. Someone who denies personaE responsibility is less likely to take action to set things right. This model of norm-activation has been used mainly to study such altruistic behaviors as helping and volunteering, but the theory has been extended to individual beliefs and actions affecting environmental quality (Heberlein, 1972), and that extension has been supported by studies relating to pollutants (e.g., Black & Heberlein, 1976; Van Liere & Dunlap, 1978) and energy use (e.g., Black, 1978; Black, Stern, & Elworth, 1985). The present work further extends the Schwartz model from the realm of personal action to social and political action, particularly in the area of environmental hazards. We hypothesize that beliefs about how environmental hazards ought to be handled are in part moral judgments--but about the people or organizations producing the hazardous substances or regulating their use and disposal rather than about the individual making the judgment. Accordingly, we predict that when a person is aware of serious negative in-

207 PAUL C. STERNETAL.

terpersonal consequences (AC) of activities that are ascribed to particular actors (AR), that person will feel that the responsible party has a moral obligation to act to avert human suffering. For example, when a person can smell the fumes issuing from the grounds of a local chemical plant (AR) and is aware of negative effects on health, tourism, esthetic values, property values, and so forth (AC), the essential elements are present for the activation of normative pressure for a cleanup. This normative pressure may be brought to bear on the chemical company and/or government officials, depending on who is perceived to be responsible to act. We conceive of this extended norm-activation theory in the context of a larger social process by which environmental regulation gains public support. Figure 1 schematically represents this process. We presume in this model that the social-psychological elements that are the focus of the present study are influenced by forces in the society, chiefly those that place individuals within the social structure, those that shape general social ideology, those that offer people information and opinions relevant to environmental hazards, and the widely shared sense of morality that underlies the operation of Schwartz's model of norm-activation. We further presume that the personal norms activated by these social forces and social-psychological processes reenter the social sphere when people take political and social action in accord with their moral judgments. In short, we conceive the social-psychological processes as part of a dialectic in which events and social forces shape individual moral judgment and in which individual action, operating through social groups, helps bring about social change.

MODEL OF SOCIAL-PSYCHOLOGICAL PROCESS
The following paragraphs provide more detail about the operation of the model at the individual level. The rationale for the causal ordering at the individual level is supported by a body of previous research on environmentally relevant behavior (see Stern & Oskamp, 1986, for a review). We use the issue of hazardous chemical wastes to ground the model in a specific environmental issue. Our discussion begins at the lower right of Figure 1, with individual behavior. 1. Overt action. Action relevant to policy on hazardous chemicals may include writing to legislators or regulators about chemical wastes, attending public meetings on the subject, joining and supporting social movement groups and political lobbies, writing letters to newspapers, voting in referenda, and participating in demonstrations or boycotts on the chemical waste issue. We hypothesize that such action will be consistent with an individual's personal norms. Individual action will be directed to government

,£

a - E . -~

~.~

e
° ~

~ m ° ~

¢-

° ~

m e~

0

0 r,~ Z

E
.m

E IJ

~

~J~i

N'-'~

~

ELI

.~

.--

208

209

PAUL C. STERNET AL.

or industry according to which party a person holds responsible, and an individual will support or oppose a particular action consistent with that individual's normative judgment. In the preliminary study we assess behavioral commitments, which we interpret as a resultant of normative judgments and as an influence on actual behavior. 2. Personal norms about chemical waste disposal. These are beliefs about what ought to be done about various aspects of the chemical waste problem. They involve judgments about what actions by chemical companies, regulators, or others ought to be prohibited, permitted, preferred, or prescribed. Several issues are involved and could be measured separately. It is possible to assess normative judgments about what chemical companies ought to do with their wastes, what they ought to do with past wastes, and what they ought to do about damage that has been done by waste disposal. It is also possible to assess normative judgments about what local, state, and national governments should do to regulate future waste disposal, to clean up past dump sites, and to compensate those who have been hurt. Such normative judgments are central to the theory: the antecedent variables in the model are all believed to be explanatory of the variation in these judgments and these normative judgments are believed to be a critical determinant of politically meaningful action. The extended normactivation theory implies that overt action depends not only on a normative judgment about the moral obligations of polluters or regulators, but also on a norm for personal action that depends on a belief that one's action can make a difference (AC) and that one is personally responsible (AR) for putting pressure on industry or government to do what is right. In the preliminary study, we do not try to assess norms for personal action, but construct a pair of scales to assess normative beliefs about action by industry and by government, the norms most significant to our extension of the norm-activation model.

3. Ascription of responsibility for consequences of waste disposal (AR). The theory predicts that normative judgments will be made about the behavior of whatever actor is seen as responsible for the consequences that have harmed people. It is possible to measure people's beliefs about the responsibility of government, industry, and others for the past, present, and future health effects of disposal of chemical wastes. In the present study, we construct two general scales to assess ascription of responsibility to industry and to government. AR may be affected by antecedent psychological variables in the model; by information received from acquaintances, reference groups, news stories, advertising, and other sources, about responsibility for toxic waste disposal; and by direct experience coming from physical proximity to places where chemical wastes have been found in the environment or have become a political issue.

210 POPULATION A N D ENVIRONMENT

4. Awareness of consequences of chemical waste disposal (AC). This is theoretically prior to the AR variable because only when someone is aware of harmful consequences does responsibility for those consequences become a moral issue. It is possible to assess people's beliefs about the health effects of particular chemicals that have been involved in waste disposal incidents and to assess beliefs about the seriousness of the effects of chemical wastes in general. Because concern about technological risk depends on qualitative properties of hazards (e.g., Slovic, Fischhoff, & Lichtenstein, 1982), it may be important to assess perceptions of the hazardous waste problem along a few qualitative dimensions that have proved in past studies to be significant determinants of risk perception and that might explain variance in overall awareness of consequences. These qualitative variables might include beliefs about the catastrophic potential of risk from toxic chemicals, beliefs about the concentration or diffusion of the hazards, and bel.iefs about reversibility of the effects. T For this preliminary study, we constructed one overall scale for AC. Like AR, AC can be influenced by information and opinions in the social system as well as by individual psychological processes. 5. General attitudes, underlying values, and personality dispositions. Broad personal attitudes and values affect a person's attention to, memory for, and belief in information. Values concerning technology, progress, and environmental purity may affect beliefs and attitudes about chemical wastes in this way. Some research shows that values (e.g., Dunlap, Grieneeks, & Rokeach, 1983) and general measures of environmental concern (e.g., Black, 1978; Black, Stern, & EIworth, 1985; Dunlap & Van Liere, 1984) affect more specific attitudes and personal norms concerning environmental issues and, indirectly, behaviors consistent with those attitudes and norms. Personality dimensions such as internal-external locus of control may also be relevant, especially to personal beliefs about the locus of responsibility for action. These broad psychological influences are shaped by social ideology, acculturation, early experience, and other forces far beyond the scope of this paper to discuss. In the present study, we assess one such variable, general environmental concern, using a measure adapted from Dunlap and Van Liere's (1980) New Environmental Paradigm scale. 6. Position in social structure. Age/cohort, race, sex, education, income, and occupation are traditional social structural measures, several of which correlate with support for environmental regulation and concern with environmental problems (Van Liere & Dunlap, 1980, provide a review). However, the theoretical meaning of these relationships is unclear. We expect that social structural variables may directly influence both general environmental concern and specific norms and beliefs about toxic chemical issues. We also expect that the strength of various links in the social psycho-

211 PAUL C. STERN ET AL.

logical model will differ in different subgroups of the population defined by position in the social structure. However, we do not explore these hypotheses in the preliminary study.

A PRELIMINARY TEST OF THE MODEL
This section reports the results of a preliminary test the social-psychological portion of our model. Although the model we propose gives emphasis to the links between social structure and individual attitudes and norms, the ways changes in social structural factors and public events produce attitude changes, and the ways in which morally committed individuals bring about social change, to examine all aspects of the model simultaneously would be too ambitious a task for an initial study. Here, we limit ourselves to a first step, an examination of the structure of norms and behavioral commitment that extends Schwartz' model from its usual use to analyze moral norms for personal action to a new application to moral norms for the behavior of social institutions. We examine the model synchronically, looking at individual attitudes, norms, and beliefs at a single point in time rather than tracking the postulated relationships over time. Sample Our preliminary test is based on a convenience sample of undergraduate students enrolled in lower division sociology and anthropology courses at George Mason University. This sample is not viewed as representative of any larger population, but is useful for a preliminary examination of the model under the assumption that the attitudinal and normative structure of this group will not differ radically from that in populations of more general interest. Questionnaires were distributed in introductory sociology, introductory anthropology, and social problems courses, with students offered extra credit for returning the survey. In all, 74 questionnaires were returned. The sample was 47% male, with a median age of 20 and an age range from 15 to 47. About two-thirds of respondents were first- or second-year students, the rest being third- or fourth-year students. About one-fourth were business majors, one-third social science majors, and one-sixth engineering or computer science majors. About 10% of the sample was nonwhite. Attitude Scales Industry and government are the two entities that can reasonably be ascribed responsibility for chemical waste problems. Since attitudes toward

212 POPULATION AND ENVIRONMENT

government and industry are likely to be rather distinct, we have constructed separate scales to measure ascription of responsibility to each, as well as separate scales to measure norms about government and industry action, and commitment to act to change government or industry behavior. Thus, we posit two separate models, one for norms and action regarding industry, the other for norms and action regarding government. In each model, causation is presumed to flow from general environmental concern to awareness of consequences of chemicals in the environment, to ascription of responsibility to industry or government, and then to normative judgments about the behaviors of the appropriate institutions and a personal commitment to act accordingly. While we have separate measures of variables specifically related to government and industry, we have developed single measures of environmental concern and of awareness of consequences. Table 1 displays the items used to construct the attitude scales, the alpha reliability of the scales, and their means and standard deviations. Students were asked to indicate where they fell on a 1-5 "strongly disagree" to "strongly agree" continuum. Scales were constructed by summing item scores after reversing the scoring of the items indicated. Items that reduced the alpha reliability of a scale were eliminated before constructing the final scales that appear in Table 1. The alpha reliabi[ities for the scales are low, which is to be expected for brief scales such as these. With the exception of ascription of responsibility to industry, reliabilities are at acceptable levels for preliminary work such as this. It is interesting to note that, without exception, the reliability of government scales is higher than the reliability of industry scales. This may indicate that students have more clearly developed beliefs about the role of government in society than about the role of industry, and thus it is harder to develop highly reliable scales of attitudes about industry roles.

Results
Table 2 presents estimates of an ordinary least squares regression model of personal norms and commitment to action concerning industry. Note that while we report tests of significance associated with coefficients, these tests assume either a simple random sample or normally distributed measurement error, and so may not hold exactly in our convenience sample with measurement error of an unknown character. As predicted by the theoretical model, norms about industry behavior have a substantial direct effect on behavioral commitment. Two variables that appear earlier in the causal model, awareness of consequences and environmental concern, also have direct effects on behavioral commitment and those effects are as great as the effect of norms.

213 PAUL C. STERN ET AL.

TABLE 1 Items and Statistical Data for Attitude Scales

Environmental Concern
Alpha = 0.49 Mean = 16.5 Standard deviation = 2.5 Humanity is severely abusing the environment. The balance of nature is delicate and easily upset. The earth is like a spaceship with only limited room and resources. *Humanity was created to rule over the rest of nature. *Humans need not adapt to their natural environment because they can remake it to suit their needs.

Awareness of Consequences of Chemicals in the Environment
Alpha = 0.42 Mean = 7.3 Standard deviation = 1.4 If we continue to dispose of chemicals the way we do now, there will be a large increase in cancer deaths in the U.S. *Science will find a way to clean our water before anyone is hurt by chemical waste in it. *We have always had chemicals in our food and water; people who worry about danger from chemicals are making a big fuss over nothing.

Ascription of Responsibility to Government
Alpha = 0.51 Mean = 8.9 Standard deviation = 1.9 Ineffective governmental regulation is primarily responsible for environmental pollution from hazardous wastes. We have a problem with hazardous chemicals because Congress doesn't have the courage to pass strong laws and fund their enforcement. *The government is doing all it can to protect people from toxic chemicals in the environment.

Ascription of Responsibility to Industry
Alpha = 0.35 Mean = 11.9 Standard deviation = 1.8 We have a toxic chemical problem because most companies don't care enough about public safety. Industries have often chosen the cheapest method for disposal of hazardous substances, even when they knew they were dangerous.

214

POPULATION AND ENVIRONMENT

TABLE 1 (continued)
*Industry can't be blamed for hazardous chemicals in the environment because the chemicals were around for years before we knew of the danger. *Only a few companies in the United States that produce hazardous waste have allowed them to harm the environment.

Personal Norm for Government Actions
Alpha = 0.60 Mean = 9.9 Standard deviation = 1.8 Protecting people from cancer-causing chemicals should be a top priority of federal and state government. Justice and morality demand that we have the strongest laws possible to protect people from toxic chemicals. *Government is overstepping its bounds in environmental regulation.

Personal Norm for Industry Action
Alpha = 0.44 Mean = 12.8 Standard deviation = 1.4 The chemical industry has a moral obligation to keep its products from harming people and the environment. As part of this society, industries are ethically responsible for keeping their waste products from threatening people's health. *The major job of industry is to earn money and provide jobs, not to protect the environment. *Industry is doing all that it ethically should to protect the environment.

Commitment to Change Government Actions
Alpha = 0.64 Mean = 12.4 Standard deviation = 2.1 I would sign a petition in support of tougher environmental protection laws. I would contribute money to a drive to strengthen the laws against dumping hazardous chemicals. I would campaign for a candidate who would support tougher environmental protection laws.

Commitment to Change Industry Actions
Alpha = 0.60 Mean = 9.2 Standard deviation = 2.2 I wouldn't take a job with a company I knew produced hazardous wastes. *If I were investing money in stocks, I wouldn't worry about the environmental record of the companies I invested in.

215 PAUL C. STERNET AL.

I would take part in a boycott of the products of a chemical company if knew it was evading the environmental protection laws. *The scoring of this item was reversed before coding (strongly agree = 1 strongly disagree = 5, etc.).

The factors influencing norms for industry action are very much as predicted by the extended norm-activation model. Ascription of responsibility to industry is a strong predictor of norms about industry action, and awareness of harmful consequences of hazardous chemicals has a direct effect on norms for industry action independent of the effect of ascription of responsibility. Ascription of responsibility to industry depends on both of the antecedent variables in the model that we have measured. Finally, awareness of consequences of hazardous chemicals is predicted by general environmental concern, paralleling results reported in other studies (Black, 1978; Black, Stern, & E[worth, 1985). Thus, it appears that the extended Schwartz normactivation model works quite well when applied to judgments and intended actions regarding the moral obligation of industry to resolve problems with hazardous chemicals. Table 3 presents results of estimating a similar model of personal norms and commitment to action regarding government. Behavioral commitment is predicted by exactly the same kinds of variables that influence behavioral commitment toward industry: normative judgments, awareness of consequences, and environmental concern. But when the antecedents of the personal norms for government action are analyzed, some predictions of the norm-activation model do not hold. Ascription of responsibility to government does not directly affect norms about government behavior, holding other factors constant, although awareness of consequences does have an effect. Ascription of responsibility to government is influenced by awareness of harmful consequences, as the model predicts, although it is not influenced directly by general environmental concern. In sum, the Schwartz norm-activation model applies well to moral norms about the conduct of industry but not of government. The difference concerns the relation of judgments about responsibility to beliefs about the moral obligation to act to correct environmental problems: These cognitions are closely connected when people think about industry but not when they think about government. This difference may be understandable in terms of the different roles people see for government and industry in advanced capitalist countries. People's views about corporate behavior seem consistent with the legal fiction of the corporation as an individual. The belief that corporations should act to solve environmental problems is stronger when cor-

~

o~

e--

> c,m 00 r'q

,.-0 L¢3 C'q ,..0 c'q

v I'-.,. ' ~ cm c'q c'4

z

©

~--~

c~
I

i.

E
0 Lm v Lt'b C'4 C'q

$e ,q,..

05
(3", Lt~, i"-.

~

~

,4,-

c;

c;

c;

Q

-X" uu ml

*

~

m

.o .~ =
._~-© ~

,-. v cq

,_4
0"',

<
I--'

-

I

I

>~ tL ~