Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
1 ~ Changing Behavior in Households and Communities: What Have We Learned? Paul C. Stern hapters 3 to 11 examine the use of what have been called communication and diffusion instruments (Kaufmann-Hayoz et al., 2001) to change envi- ronmentally significant behavior in households and communities. These instruments include information, education, the use of models, other informal social influences, and other interventions that rely primarily on language and visual symbols. Communication and diffusion instruments are used to supple- ment the traditional policy instruments of regulation (command and control), economic influence, and the provision of infrastructure and services to make desired behaviors more feasible. They are the centerpiece of social marketing efforts in environmental policy (McKenzie-Mohr and Smith, 1999. The key policy questions about these instruments are how much they can contribute to environmental protection objectives and how best to use them to achieve this potential. As chapters 3 to 11 indicate, much has been learned about how to design these instruments for greatest effectiveness and about what they can be expected to accomplish, both on their own and combined with other policy instruments. THE POTENTIAL OF COMMUNICATION AND DIFFUSION INSTRUMENTS The potential of any policy instrument depends on its fit with the policy objective. An instrument has the greatest potential when it can provide just what is needed to overcome the barriers to attaining the objective. For example, regulations distinctively provide assurance of fairly equal compliance across tar- get firms. Thus, they have great potential value when the firms would comply 201
202 CHANGING BEHAVIOR IN HOUSEHOLDS AND COMMUNITIES voluntarily, except for the concern that they might be put at a competitive disad- vantage. When there are major barriers to the desired behavior that a policy instrument cannot remove, that instrument has very limited potential. Thus, regulations have limited effect when they call for changes that are technological- ly or economically infeasible. Figure 12-1 identifies causal links between the types of policy instruments identified by Kaufmann-Hayoz et al. (2001) and a range of factors that in turn influence environmentally significant behavior. Although informed observers will disagree on which of these links are most important, available knowledge strongly supports the key point: Each type of policy instrument has particular capabilities and thus can influence only a subset of the many factors that drive behavior. Depending on what is standing in the way of a target behavior, a particular instrument may be highly successful or nearly useless. Communica- tion and diffusion instruments, as shown in Figure 12-1, can influence some aspects of the target individuals and their immediate social contexts, but cannot directly affect the broader social, economic, or technological contexts.2 They cannot make inconvenient behaviors convenient, make expensive behaviors in- expensive, or remove institutional or legal barriers to behavioral change. They often cannot even get people to put environmental actions high enough on their personal to-do lists to get them done, even if they are convinced to act. Environ- ment-related actions must compete with other demands on a person's time and energy. It follows that when such contextual factors stand in the way of a target behavior, communication and diffusion measures by themselves will have little effect. Similarly, when the target behavior is seriously impeded by lack of information, social support, behavioral models, and the like, regulatory and eco- nomic instruments by themselves may have little effect. These points may seem self-evident, but they have not always been reflected in the design of environmental policies and programs. Many documented fail- ures of environmental and energy information programs in the household sector can be attributed in part to a failure to address significant noninformational barriers to behavioral change (see, e.g., National Research Council, 1984; Gard- ner and Stern, 1996; Lutzenhiser, this volume, Chapter 3; Schultz, this volume, Chapter 4~. Similarly, the disappointing performance of many financial incen- tive programs targeting these behaviors can be attributed in part to a failure to diffuse the programs adequately (e.g., Stern et al., 1986~. The most effective interventions tend to combine various types of communication and diffusion instruments with each other and with other policy instruments (Gardner and Stern, 1996; McKenzie-Mohr and Smith, 1999~. The implication for communication and diffusion instruments is that they have their greatest potential under two sets of conditions. In the first, the factors that communication and diffusion can influence (see Figure 12-1) are the only important barriers to the desired behavioral change. Under these conditions, well-designed communication and diffusion programs can bring about important
PAUL C. STERN Policy instruments Communication and diffusion instruments Collaborative agreements - - _ | Command | I And control L ------ _ _ >< Economic I instruments I__ | Service and | infrastructure 203 Drivers of behavior Personal capabilities and constraints Literacy, social status, behavior- specific knowledge and skills ~_= Values, attitudes, beliefs, personal norms Social context -Social norms, persuasion, advertising, personal commitments, informal institutions Institutional, economic, and technological context _ . ( Laws. regulations ~ and rewards ~ Privets contracts - - Available technology Convenience FIGURE 12-1 Paths of influence of five types of environmental policy instruments on five factors that affect environmentally significant behavior.
204 CHANGING BEHAVIOR IN HOUSEHOLDS AND COMMUNITIES behavioral change without the aid of other policy instruments. In the second set of conditions, the barriers include both factors that communication and diffusion can influence and factors they cannot but other policy instruments are avail- able to remove the other barriers. Under these conditions, combining communi- cation and diffusion instruments with the other policy instruments can bring about important behavioral changes that neither policy type alone could achieve. Communication and diffusion instruments are thus important as adjuncts to or partners with other policy tools. The next section summarizes knowledge about how to design communica- tion and diffusion instruments. Applying this knowledge is essential for the instruments to work well under either of the sets of conditions already described, although under the second set of conditions, communication and diffusion tools are not enough, no matter how well designed. The subsequent section discusses the application of communication and diffusion tools in situations where they are not sufficient for attaining policy objectives. DESIGNING COMMUNICATION AND DIFFUSION FOR GREATEST EFFECT Chapters 3 to 7 summarize current knowledge about how to design commu- nication and diffusion instruments to be as effective as possible. Chapters 3 to 5 cover the most carefully studied applications in environmental policy. Chapters 6 and 7 summarize knowledge from well-studied domains outside environmental policy where there is a long history of research on communication and diffusion instruments. They arrive at conclusions quite consistent with those mentioned in Chapters 3 to 5. The generalizations that receive the most consistent support across domains are described in the following subsections. Design the Intervention from the Behaver's Perspective Environmentally significant behavior is a product of the individual and the situation; more specifically, it is a product of the individual's values and atti- tudes, personal capabilities and constraints, and habits and routines, as well as of contextual factors that provide incentives, possibilities, and constraints (Stern, 2000~. Because these things vary with the individual, successful efforts to change behavior are those that are matched to the individual's needs. This does not mean that effective communications must be individualized. Programs can succeed by targeting types of people whose situations are similar with regard to a target behavior or by being so multifaceted that they can be effective across a variety of people and behavioral contexts. To pursue either approach effective- ly, however, program designers must make explicit efforts to understand the behaver's perspective. This can be done by employing social research tech- niques (e.g., surveys, focus group techniques, ethnographic methods) and by
PAUL C. STERN 205 involving members of the target group or other people who have detailed expe- rience-based understanding of the target audiences in the design of the pro- gram (e.g., Gardner and Stern, 1996; Werner and Adams, 2001~. Build on Interpersonal Communication Impersonal communication efforts, such as mass mailings and mass media advertising, are easy for policy makers to organize at large scales, but have much less influence on any individual than personal communication that comes from people the target individual cares about or trusts. Personal communication is especially important when elements of the message are controversial or when the original source of the message (e.g., a government agency) has limited credibili- ty with portions of the audience (National Research Council, 1984~. Devising ways to gain the benefits that can come from interpersonal com- munication can take ingenuity, especially with large-scale policy objectives. One useful strategy is to induce respected leaders or people central to communication networks to adopt a desired behavior and thus act as models whose behavior may be readily adopted by others (Rogers, 1995; Valente and Schuster, this volume, Chapter 6~. Another strategy is to partner with community groups and voluntary associations that can act as intermediaries who convey messages between policy makers and target individuals. Such groups often can make personal contact and can command a level of attention and trust from their constituencies that mass appeals rarely achieve. These groups are not simply channels for transmitting messages. They are most effective when they adopt the intervention as their own, perhaps adapting the message in the process to make it meaningful to their constituencies. A third strategy is to make existing social norms more visible, as Schultz (this volume, Chapter 4) did in an experimental manipulation with curb- side recycling. This approach is most promising in situations in which, as with curbside recycling, the expectations and opinions of others matter and those others can monitor the relevant behavior. Use Multiple Channels to Communicate the Message As a rule, messages are most influential when they reach audiences in many forms and from many sources (Mileti and Peek, this volume, Chapter 7; Valente and Schuster, this volume, Chapter 6~. This is probably the case because differ- ent people attend to and trust different sources, because different channels may have advantages for conveying different parts of the message, and because mul- tiple channels provide an effective way to repeat and reinforce messages. Apply Psychological Principles for Message Design Messages are most effective when presented in terms, metaphors, and imag-
206 CHANGING BEHAVIOR IN HOUSEHOLDS AND COMMUNITIES es the audience understands and finds attractive. When they involve calls for action, they can be made more effective by emphasizing the costs or dangers of inaction but only when they also provide clear advice on what to do to avoid those hazards, thus giving audience members a sense that they can control their fates rather than creating fear and anxiety (Rogers, 1983; Weinstein and Sand- man, 1992; Gardner and Stern, 1996~. Useful summaries of the research on message design can be found in Chapters 6 and 7 of this volume and elsewhere (e.g., McKenzie-Mohr and Smith, 1999; Morgan et al., 2002~. 3 As already noted, effective message design depends on understanding the target audiences and how they perceive the target behavior. Maintain a Program's Momentum Experience with communication and diffusion efforts indicates that pro- grams maintained over long periods can be much more effective than one-shot or short-term programs (disaster preparedness and public health provide good ex- amples, as noted in Chapters 6 and 7~. Repetition helps messages sink in and increases the likelihood that a message will be received when the recipient is receptive, such as during a crisis or near the time of a noncrisis decision. Set Realistic Expectations Communication and diffusion instruments take time to be effective. A mes- sage must get into awareness and penetrate into a decision process in order to bring about behavioral change (Th0gerson, this volume, Chapter 5; the Knowledge- Attitude-Practice curves described by Valente and Schuster, this volume, Chap- ter 6~. In addition, action may be delayed even when someone is psychologically prepared to change. For example, a household will acquire a more energy- efficient motor vehicle only when the time comes to change vehicles. Someone may not change an old habit until the right occasion arises (e.g., reconsidering the use of mass transit when one's work location changes). Because of such predictable delays, communication and diffusion instruments should be evaluated against a behaviorally defensible timetable for progress. Expectations should also consider contextual factors that may limit the effect of understanding on behavior change. This point is discussed further in the next section. Continually Evaluate and Modify Programs Policy interventions should not be expected to be at their best the first time they are tried, nor to maintain a constant level of performance in a changing environment. They need to be evaluated and adjusted if they are to achieve their full potential.
PAUL C. STERN 207 USING COMMUNICATION AND DIFFUSION WITH OTHER TOOLS As already noted, communication and diffusion instruments can have little direct effect on changing the institutional, economic, or technological contexts of environmentally significant behavior. Where these contexts are unfavorable, the best use of communication and diffusion is in conjunction with other policy tools that address the relevant contextual issues. Therefore, it is important to understand the context in order to find the best use of communication and diffu- sion tools. One important kind of contextual influence is discussed by Press and Balch (this volume, Chapter 11~. They argue that the effectiveness of all locally imple- mented environmental policy instruments, including communication and diffu- sion instruments, is contingent on the policy capacity of local institutions. Put more provocatively, their argument implies that no matter how well designed a community-based communication program may be, it will only be effective in certain kinds of communities. If a community is lacking in local finances, ad- ministrative expertise, civic involvement, and some other qualities, Press and Balch argue, implementation likely will fail. Communication and diffusion in- struments need to be supplemented with, or to follow after, efforts at community capacity building. In some contexts, communication and diffusion can be combined with other policy instruments for synergistic effect. A good historical example was the financial incentives used to promote energy efficiency in homes in the aftermath of the 1970s energy crises. Several U.S. electric utility companies offered such incentives, but the rate of acceptance was fairly low apparently due in part to inadequate communication and diffusion efforts. Some programs, however, were 10 or more times as effective as others that offered identical incentives, but marketed them in different ways (Stern et al., 1986~. Apparently, communica- tion and incentives had complementary functions: communication drew atten- tion to the programs (as indicated by requests for energy audits), and once con- sumers noticed, larger incentives increased acceptance of the financial incentives (see Figure 12-2~. When incentives were large enough, communication and diffusion had a very large practical effect by getting consumers to consider the incentives (Stern, 1999~. Communication and diffusion may have similar synergistic effects with ser- vice and infrastructure instruments such as the provision of new public transit lines or curbside recycling services: These new services may not be well used unless they are well marketed. Available evidence suggests that the standard marketing strategy simple information dissemination is usually not enough. What is needed is to combine new services and infrastructure with communica- tion programs designed according to the principles described in the previous section.
208 100 - 80 - ~n s In O 60- a ~ 40- c' 20 - CHANGING BEHAVIOR IN HOUSEHOLDS AND COMMUNITIES Energy audits per eligible household per year 88.5 (bar represents range across programs) Loans or grants accepted per audit (bar represents range across programs) 28.1 2.6 1.0 23.2 4.0 23.2 1 ~ 90.9 o 1 1 1_1 1 .. _ 1 Reduced-rate loan program (9 utilities) Zero-interest loan program (11 utilities) 93% average rebate program (7 utilities) FIGURE 12-2 Households requesting energy audits (white bars) and accepting incen- tives once they have received audits (shaded bars) in three home energy conservation incentive programs. Source: Stern et al. (1986~. Reprinted with permission of Allyn and Bacon. THE ROLE OF EDUCATION Although education relies primarily on communication, it is different in its objectives from the kinds of social marketing efforts already described. When communication is used to achieve an environmental policy goal (i.e., for social marketing), its objective is to change the prevalence or frequency of a target behavior that directly affects environmental quality. In environmental education as defined by Ramsey and Hungerford (this volume, Chapter 9), communication is used to improve understanding of environmental and related phenomena and to enable and encourage environmental citizenship, but not normally to change specific behaviors that directly alter environmental conditions.4 If students from a single environmental education class all became active in environmental lob- bying, but took opposite sides on an issue, the class might be properly counted an educational success. The effects of environmental education, defined in this way, on environ- mental quality are hard to assess. For one thing, they are mainly indirect, operat- ing through public policy. For another, the effects on policy may not all be in the same direction. What good environmental education does for environmental
PAUL C. STERN 209 policy is to raise public discussion to a higher level. Disagreements based on misinformation give way to those based on alternative interpretations of correct but ambiguous information or different judgments about what to do under envi- ronmental uncertainty. The policies that result from better informed debates are not predictable, but if democracy works well, they tend toward results that citi- zens want including environmental results. When environmental education is the self-education of adults, it can target a wider range of behaviors, as Andrews and colleagues point out (Chapter 10~. A community that devises its own adult environmental educational program may begin with consensus on environmental objectives. For example, it may decide that water conservation is imperative and devise a program aimed at explaining local water supply conditions, showing why water conservation is necessary, and teaching people how to conserve water. Such a program may include elements of both education and social marketing and, if well designed, may greatly influ- ence environmental outcomes. However, combining education and social mar- keting in this way can be highly controversial when it is proposed as a public policy strategy because of objections to government attempting proactively to influence the publics it is supposed to represent. This objection can be overcome if a legitimate public decision is made to adopt a social influence policy, as has been done in the United States for combating the use of illegal drugs and for driving under the influence of alcohol. A useful guideline for when it is appro- priate to use education for social influence has been stated in another context by the National Research Council (1989:90~: It is justifiable "only to the extent that some legitimate public process has culminated in a decision that using [educa- tion] to influence behavior serves an important public purpose." Community- based social marketing has the potential to meet this test and thus achieve wide acceptance. CONCLUSION Research has shown that communication and diffusion instruments can, un- der certain conditions, make significant contributions to meeting environmental policy objectives. It has identified a number of robust principles for designing these instruments to reach their potential. It has shown that these principles must be implemented in ways that are sensitive to the situation and that systematic evaluation is needed to achieve and maintain the fit of programs to settings. Research also has helped distinguish three kinds of situations: those in which communication and diffusion instruments can yield significant environmental effects on their own, those in which they have potential only when supplemented by other instruments, and those in which they are unlikely to be successful even when combined with other policies. Thus, it has helped to clarify the functions of communication and diffusion in the environmental policy toolbox and to show how they can be used to greatest effect.
210 CHANGING BEHAVIOR IN HOUSEHOLDS AND COMMUNITIES The potential of communication and diffusion can be quantified only in relation to particular situations. In favorable situations for communication and diffusion alone, such as the 1970s energy crises, well-conducted communication programs have reduced household resource consumption by 10 to 20 percent for short periods, beyond what could be achieved by conventional information dis- semination (Stern, 1992~. In situations appropriate for combining communica- tion and diffusion with other policies, the instruments have even greater poten- tial. For example, an integrated residential energy-efficiency program in Hood River, Oregon, in the 1980s achieved nearly complete adoption of recommended energy-efficiency improvements throughout the community, a result never ap- proached by other programs, even when very strong financial incentives were offered (Hirst, 1987~. NOTES 1 Communication and diffusion are also at the heart of commercial advertising, much but not all of which runs counter to the goals of proenvironmental social marketing. This tension between environmental policy goals and those expressed in commercial "countermarketing" is discussed by Lutzenhiser (this volume, Chapter 3). Countermarketing may focus on specific behaviors (usually purchases); it also may promote general values and attitudes that support a range of environmentally consumptive behaviors. 2 Communication can change institutional and other contexts indirectly by influencing societal ways of thinking. The classic example is the effect of books like Silent Spring (Carson, 1962) on the U.S. environmental movement and public support for environmental regulation in the 1960s and early 1970s. 3 Chapters 6 and 7 report on research and practice in the public health and disaster prepared- ness communities, which operate from a philosophy very friendly to social marketing. They deal with hazards that are widely accepted as important and for which there is broad public support for using government to influence people to act in ways that promote both personal and social interests. Social consensus is harder to find in environmental policy. Consequently, practitioners of environ- mental "risk communication" often operate on a philosophy that favors providing balanced informa- tion that people can use to make informed decisions (National Research Council, 1989). Energy conservation, recycling, and "green" purchasing are among environmental policy goals for which various communication philosophies may operate in different communities or countries and for which the community philosophy may change with the times. 4 A great many environmentally significant behaviors can be classified as environmental activ- ism, nonactivist behaviors in the public sphere (e.g., contributing to organizations that work on environ- mental issues, attending public meetings, expressing opinions about environmental policies), or private- sphere environmentalism (e.g., purchasing "green" products, composting household waste, maintaining automobile engines to reduce pollution) (Stern, 2000). Environmental citizenship consists mainly of the second class of behaviors; only the last class of behaviors directly affects environmental quality. REFERENCES Carson, R. 1962 Silent Spring. Boston: Houghton-Mifflin.
PAUL C. STERN Gardner, G.T., and P.C. Stern 211 1996 Environmental Problems and Human Behavior. Needham Heights, MA: Allyn and Bacon. Hirst, E. Rogers, E. 1995 The Diffusion of Innovations. New York: The Free Press. Rogers, R. 1983 Cognitive and physiological processes in fear appeals and attitude change: A revised theory of protection motivation. In Social Psychology: A Sourcebook, J. Cacioppo and R. Petty, eds. New York: Guilford Press. Stern, P.C. 1992 What psychology knows about energy conservation. American Psychologist 47:1224- 1232. 1999 Information, incentives, and proenvironmental consumer behavior. Journal of Consum- er Policy 22:461-478. 2000 Toward a coherent theory of environmentally significant behavior. Journal of Social Issues 56:407-424. Stern, P.C., E. Aronson, J.M. Darley, D.H. Hill, E. Hirst, W. Kempton, and T.J. Wilbanks 1986 The effectiveness of incentives for residential energy conservation. Evaluation Review 10: 147- 176. 1987 Cooperation and Community Conservation. Report 1300Y 1987. Portland, OR: Bon- neville Power Administration.. Kaufmann-Hayoz, R., C. Battig, S. Bruppacher, R. Defila, A. Di Giolio, U. Friederich, M. Garbely, H. Gutscher, C. Jaggi, M. Jegen, A. Muller, and N. North 2001 A typology of instruments for the promotion of sustainable development. In Changing ThingsMoving People: Strategies for Promoting Sustainable Development at the Local Level, R. Kaufmann-Hayoz and H. Gutscher, eds. Basel, Switz.: Birkhauser. McKenzie-Mohr, D., and W. Smith 1999 Fostering Sustainable Behavior: An Introduction to Community-Based Social Market- ing. Gabriola Island, British Columbia, Can.: New Society Publishers. Morgan, M.G., B. Fischhoff, A. Bostrom, and C.J. Atman 2002 Risk Communication: A Mental Model Approach. New York: Cambridge University Press. National Research Council 1984 Energy Use: The Human Dimension. Committee on the Behavioral and Social Aspects of Energy Consumption and Production. P.C. Stern and E. Aronson, eds. New York: W.H. Freeman. 1989 Improving Risk Communication. Committee on Risk Perception and Communication. Washington, DC: National Academy Press. Weinstein, N. and P. Sandman 1992 A model of the precaution adoption process: Evidence from home radon testing. Health Psychology 1 1: 170- 180. Werner, C.M., and D. Adams 2002 Changing homeowners' behaviors involving toxic household chemicals: A psychologi cal, multilevel approach. Analyses of Social Issues and Public Policy 1:1-32.