Associate Professor / Doctorate Program in Sociology and Political Science / UFMG
Adjunct Faculty Associate / Institute For Social Research/ University of Michigan
Prepared for presentation at the Open Meeting of the Global Environmental Change Research Community
Rio de Janeiro, 6-8 October, 2001 The assertion that the measurement and analysis of public opinion has become a central feature of liberal-pluralist democracies is hardly disputable nowadays. Opinion polls are increasingly used by political parties, newspapers, radio, TV and business organizations to profile issues and influence decision-making. On a similar fashion, survey methodology (the academic, more theoretical and analytical oriented counterpart of commercial public opinion polling) has sought to play a role in the production of scientific knowledge, as well as in the planning and realization of social change.
In the last decades survey methodology has sought to produce knowledge relevant for diversified areas of prevailing social and political issues such as voting, social stratification, racial relations or crime. One could have expected then, that in the eighties, when global environmental problems became a central issue in social life and in social science, survey research would be called upon to produce scientifically based knowledge relevant for policy making. The challenge that has been posed is to achieve useful knowledge through description and understanding of the human role in causing global environmental change and the consequences of these changes for society (Simões, S. & Stycos J., 1995). This challenge has several major components.
The first one resides in the task of measuring the global dimension of environmental problems. A crucial issue we should bear in mind is that the environment is a multidimensional concept, and very often the local, the national and the global are presented to respondents as distinct dimensions, although it is rarely investigated how clear-cut, if at all, these dimensions are for the respondents.
Many of the issues which gave rise to questionnaire items in surveys of the mass public are either local environmental problems (which are the ones survey research is better equipped to measure, but are not necessarily directly related to biosphere changes) or are global questions (such as global warming, loss of biodiversity, depletion of the ozone layer) so far removed from the daily life of individuals that the questions might make no sense to large segments of the population, bringing into question the validity of the measures.
We could distinguish then two different approaches to the measurement of individual’s perception and behavior concerning global change. The first one directly asks the population about their individual perceptions of the salience, importance, causes and consequences of, for instance, global warming and loss of biodiversity. A second approach argues that, in thinking about the human dimensions of global environmental change, energy use is one of the solution/amelioration via changes in individual values and behavior. Both in terms of the resources consumed or transformed to produce energy and in terms of the pollution or degradation, which results from energy production, this is a major factor. In global environmental terms, one is talking, for instance, about global warming that results from increases in carbon dioxide levels in the atmosphere or biodiversity loss, which is related to forest burning, among other changes (Simões, S. & Hogan, D., 1997).
According to this second approach, one should be measuring attitudes (transformations in values relative to consumption patterns, support for public policy related to individual and public transportation); consumption behavior (energy conservation through more efficient or less use of fossil fuels; re-using and recycling materials, which reduces the energy necessary for production; or transportation) and political behavior related to those issues. This last rationale was only more recently developed by the Global Environmental Survey, but in most surveys we can find a combination of direct abstract questioning about global environmental problems and various attitudinal/behavioral items related to energy use.
The second challenge we face is to understand the growing “greening of the world” - the widespread concern with environmental issues throughout the world and across the various social segments in the last decades. The question being posed is: what is being measured? Overstatement of concern or deep-rooted attitudes and behavioral change? In fact, in this chapter we are concerned with both how broad environmental concern among Brazilians is, and how deep it is. In considering the breadth of concern we look into whether it is felt by a majority or minority, and to what degree, and among whom/ what social segments). In examining the depth of the Brazilian mass public’ views, we address questions such as the implication of environmental concern in relation to behavior. This chapter also seeks to present and discuss Brazilian’s environmental attitudes and behavior from an international comparative approach. In order to do that, we will draw on data from various national and international surveys (see section below).
Are we all green? Values and attitudes versus behavior In this section we will briefly introduce some of the major findings on Brazilian’s concerns and behaviors, with a special focus on the inconsistencies one can find between these two dimensions of environmentalism.
We will illustrate our arguments with data from three major international surveys that included Brazil (Health of the Planet-19921; World Values Surveys - 91 and 952; and the Global Environmental Survey - 19973) and four surveys conducted in Brazil (USAID Survey - 19914; What Brazilians Think about Ecology Survey – 19925; What Brazilians Think about the Environment, Development and Sustainability Survey – 19976; the Rio das Velhas Basin Survey - 19977). Given the choice of an international comparative approach for this chapter, I also draw on findings from an important international survey that did not include Brazil – the International Social Survey Program’s Research into Environmental attitudes and Perception Report8.
It should be noted that these surveys were conducted in two different points in time. Several of them have been carried out in the early nineties, especially in the period immediately prior or after the Rio Conference. Two of them (WBTEDS and Rio das Velhas Basin Survey) were conducted in the late nineties (1997) – this should allow us to look at convergence of findings, and continuities and discontinuities in the environmental attitudes and behavior among Brazilians.
A common survey finding in the last decade was the existence of a widespread concern with the environment. Earlier studies (e.g. Lipsey, 1977; Mc Evoy III, 1972) had showed that environmental concern was particularly prominent among higher-educated, higher-income, and younger people. However, more recent studies indicate that environmental concern is no longer a mainly middle-class issue, but transcends social-demographic borders (Ester et al., 1993; Nelissen, 1992; Olsen et al., 1992). Furthermore, it even transcends traditional North-South and East-West divisions (Dunlap et al., 1993; Ester and Mandemaker, 1994) – developing as well as developed nations share widespread environmental concern. This sounded to many as a rather puzzling finding since, as Dunlap (1993) pointed out, both conventional wisdom and theoretical knowledge would lead us to expect low levels of concern for the environment in the so called developing nations. Accordingly, Brazilian researchers (Crespo and Leitao, 1993, pp.235) were rather puzzled by their own 1992 survey findings:
"Such findings would support the view that Brazilians have a vanguard environmental awareness, being at the same time ready to stop the uncontrolled exploitation of natural resources as well as to make sacrifices to preserve nature. Such disposition would place Brazilians in the same level of environmental concern as the one of the advanced societies, the ones that have already met the demands set by the realm of necessity or modernity, being able to dedicate themselves to explore the open possibilities of the realm of freedom or post modernity”.
In this chapter I seek to look into these unexpected (implausible to some) findings on Brazilian environmentalism, by tackling both the breadth and the depth of environmentalism in Brazil. This section introduces the description and analysis of the breadth of environmental attitudes in Brazil.
A first way most surveys measure concern is by asking about the importance or the seriousness of environmental problems.
Survey research findings have repeatedly found that Brazilians consider the state of the environment to be a very serious issue. In the early nineties, nearly the majority (49%) rate environmental issues as "very serious" in the country (HPS). Likewise, measuring the importance of environmental problems in Brazil, the USIA Survey found that majorities of Brazilians (62%) considered them to be "very important". In the WBTES-92, the respondents were asked how much they were interested in the environment, question to which a majority (51%) replied by saying that they were "very interested". Similar findings held true for the HPS a majority in Brazil (53%) said they were personally a "great deal" concerned about the environment.
Looking at environmental quality,only 3% of Brazilians rated, overall, the quality of their environment as "very good" (USIA Survey). In its turn, in order to measure the respondent’s perception of environmental quality, the HPS distinguished among the community, national and world dimensions. Brazilians followed the trend applicable to most countries the more distant the environment being rated, the more negatively it was viewed. Moreover, a majority of Brazilians (76%) considered environmental problems "a great deal" of a threat to their own and theirfamily’s health (USIA Survey). In its turn, the HPS introduced a distinction between perception of effects to "respondents own health ten years ago; own health now and the health of their children and grandchildren over next 25 years." Brazilians followed the pattern of less concern in the past and growing concern about the future.
When looking at the above results one should bear in mind, however, that despite considering the environment an important or serious issue with implications for the family well-being, the environment has not appeared as a salient issue for Brazilians. This was the case in 1992 and in 1997 as well. In the HPS, only 2%, and in the USIA Survey 0 %, of Brazilian respondents volunteered environmental problems as the most important problem in the country. In the more recent surveys, the environment was still only spontaneously mentioned by less than 1%, in both the Rio das Velhas Survey and the WBTEDSS-97.
In order to interpret the above findings we first need to bear in mind the distinction between importance and salience of issues. Furthermore, when the concept of the environment is presented in a generic and abstract form and compared to other broad issues such as health and education, it is not perceived as an important issue. When we change the frame of reference from the country to the city/local, and the environmental problems are presented in their more local and concrete dimensions, such as water pollution and sanitation, they appear as more relevant to respondents (this is the case in both the Rio das Velhas Basin Survey and WBTEDSS-97). If respondents are given the chance of spontaneously naming the most important environmental problem, national or local problems issues such as water pollution and sanitation appear as central.
In my view this can be seen as an indication of the need to further investigate the environment as a multidimensional concept and explore people’s views for different frames of reference (local, national, and global, for instance) and different dimensions (attitudes and behavior). The Rio das Velhas Basin Survey findings seem to indicate that people’s own definitions can be widely varied (“everything” for 17%; “life” for 13%; “quality of life” for 9%; “source of life” for 7%; “natural resources reserve” for 7%; “health” for 7%; noting that one-fourth of the respondents could not provide an answer). We would like to argue that in all cases, but especially in the case of developing countries (given higher levels of illiteracy and lack of information), it is necessary to make clear which dimension of this broad multidimensional concept people might have in mind when answering attitude items.
It is also worth mentioning that, as measured in the HPS, Brazil was among the developing nations with lowest levels of concern (together with Uruguay 3%, Philippines 2%, Nigeria, Poland and Hungary 1% each). Although we had sharp variations in the level of concern within each block of nations, it might be meaningful that only 4 developing nations (out of 12) had more than 10% volunteering environment as the most important problem, whereas only 3 developed nations (out of 12) had less than 10% doing so. But Brazil and Mexico, both developing nations, stood in polar positions one (Mexico - 29%) ranked among the ones where the environment is a salient issue and the other (Brazil) where it is very rarely volunteered. Economic problems appeared as "the most salient" issue in most societies; developed or developing. What is challenging here is to understand why the "environment" appears or not as a salient issue although the data shows that it was more likely to be volunteered in most of the developed countries, we are left with the problem of understanding why in societies such as Norway (2%) and Great Britain (3%) so few volunteered the environment as the most important problem, and why in developing countries like Mexico (29%), India (21%), Chile (20%) and Turkey (18%) it appeared as a salient issue. My belief is that trying to explain these somewhat unexpected findings we might gain insight into the other major factors conditioning salience of issue. As various survey researchers have repeatedly argued, we do need contextual data to make sense of these variations.
Among a variety ofenvironmental concerns, at the national level Brazilians are mostly concerned with water pollution (35% in the USIA Survey; 38% in the WBTE-92, 26% in the WBTEDS–97), deforestation (30% in the USIA Survey, 46% in the WBTE-92, 45% in the WBTEDS–97), and air pollution (29% in the USIA Survey; 18% in the WBTE-92, 12% in the WBTEDS–97). When it comes to the local dimension, sanitation and garbage collection are major issues in the earlier HPS as well as in the more recent WBTEDS-97 and Rio das Velhas Basin surveys. It stands out that those issues are mentioned spontaneoulsly in the WBTEDS-97 and appear as salient (together with water pollution) in the lists presented to respondents by HPS and the Rio das Velhas Basin Survey. It is also noteworthy that these findings converge with international findings (Health of the Planet, 1992; Environmental Monitor,1997), which shows that the mass public in the so-called developing countries choose water pollution as the most pressing problem, whereas the mass public in the developed countries are mostly concerned about air pollution.
Similarly to the perception of problems in the national dimension, when it comes to the global dimension deforestation and water pollution are again viewed as the most important problems by Brazilians in 92 (WBTE-92) as well as in 97 (both in the Rio das Velhas Basin Survey and WBTEDS-97). Among the global environmental problems, nearly one third of Brazilians also single out loss of biodiversity and depletion of the ozone layer, in 92 as well as in 97 (WBTE-92 and WBTEDS-97). Smaller percentages mention global warming as one of the most important global problems (17% in the WBTE-92 and 19% in the WBTEDS-97 ). As the most important global problem, global warming is singled out by only 4% in the Rio das Velhas Basin Survey. In fact, most Brazilians say they have never heard about global warming (over 50%), or biodiversity (nearly 70%) (WBTMADS-97).
However, even when we move from the “cost-free” measures of environmental concern to the trade-offs, such as the dilemmabetween environmental protection vs economic development, majorities in Brazil adopt a pro-environmental position. This is a relatively consistent finding through time and various surveys (64% - USIA Survey, 71% - HPS, 47% - Rio das Velhas Survey). Furthermore, compared to residents of the high-income nations, Dunlap (1993) argues that Brazilians are as likely to give the environment priority over the economy. Even when confronted with the scenario of having to make personal sacrifices to protect the environment, a majority of Brazilians (53%) say they are willing to pay higher prices to protect the environment (HPS). Brazil is among the 17 nations (including low-income ones) out of 24 where majorities in 1992 indicated a willingness to pay higher prices for increased environmental protection (HPS).
How have the researchers “interpreted” such findings? Dunlap et al (1993 and 1995) sounded straightforward and confident in the conclusion that "that such higher proportions of citizens of so many nations, including many countries with low standards of living, express a willingness to pay higher prices if it will improve the environment is perhaps the strongest evidence we found of worldwide concern for environmental quality." (Dunlap, 1995)
In his turn, enthusiastically quoting the USIA Survey findings to support his analysis of worldwide concern for the environment, Worcester (1993, pp.14) sees asunsurprising that by margins of more than two to one, in developing countries such as Mexico (66% to 30%) and Brazil (64% to 24%) people would give priority to the protection of the environment over economic growth. This finding was unsurprising, in his view, under the circumstances - the fact that majorities of Brazilians (75%) and Mexicans (62%) believed environmental problems to be a great deal of a threat to their own and their family’s health.
The WBTE-92 researchers, however, were at first clearly awestricken by similar findings in their 1992 Brazilian survey. They found quite puzzling that a majority of Brazilians (57%) strongly disagreed with the statement "a comfortable life brought about by progress is more important than preserving nature". However, they believed that these amazing findings were taken even further in the answers to the question "Would you accept living with more pollution if this means more jobs? Do agree or disagree?” A majority (64%) strongly disagreed and only 17% strongly agreed. This question was replicated in the 1997 wave and, again, practically the same majority of Brazilians (65%) adopted a pro-environmental stance.
Such results and their interpretation have been met, however, by considerable skepticism among those who stress the pervasive huge gap between attitudes and behavior. We should therefore try to address the question posed earlier on: does environmental concern translate into pro-environmental action and support for public policies?
Moving on from the measurement of attitudes to the more methodologically challenging issue of behavior (or at least reported behavior), two dimensions of environmental behavior can be distinguished: consumer and political behavior. When looking at the consumer behavior dimension, the HPS found that avoiding environmentally harmful products was the most widely practiced of the three indicators of behaviors (the others were participation in organizations and vote), with respondents in 18 of the 24 nations reporting having done that in the past year. As Dunlap (1993, pp.33) pointed out, "not surprisingly, respondents in wealthy nations are most likely to report having done this, since they are most likely to have choices among products and to be able to afford choosing on more than price alone". Even that being the case, it is noteworthy that in the HPS Brazil appeared as the country with the smallest percentage of green consumers (26%), only matched by India (27%).
The WBTES-92 and 97 waves measured willingness to adopt behavior, which is rigorously an attitudinal measure rather than a behavioral one. In both waves, Brazilians tended to choose recycling (59% and 72%) and reducing energy and gas consumption (38% and 41%) as measures they would be willing to adopt. The mass public did not appear as willing topay more for food without chemical products (14% and 11%) or to buy energy efficient home appliances (5% in the 1997 wave).
The other dimension of behavior, political behavior, wasmeasured, in 1992 by the HPS, by looking at participation in a group or organization that works to protect the environment. The proportion of Brazilian activists was unsurprising low- only 4%. In 1997 the WBTEDS, though, fifty percent of Brazilians say they are willing to join an environmental group, 55% willing to do volunteer work for NGOs, and a smaller percentage (26%) willing to make a financial contribution.
The Rio das Velhas Basin Survey helps to throw light into this issue by measuring both willingness and actual (reported) behavior. As one would have expected, the results suggest there is a substantial gap between willingness and actual behavior. Nearly 90% of the respondents declare they have never participated in environmental groups activities, educational campaigns, attended meetings with government officials or done volunteer work to clean for the environment. But willingness is relatively high for joining environmental groups, participating in meetings, doing volunteer work, and especially for taking part in educational campaigns.
The findings on willingness to join an environmental group should also be contrasted with knowing one. In the Rio das Velhas Basin Survey less than 1% spontaneously cited an NGO. Even when presented with a list of NGOs and governmental agencies in the WBTE-92, a large majority (82%) declared never having heard of any group or organization. Greenpeace and some other few NGOs reached 1%.
Still in the political dimension of behavior, as far as electoral politics is concerned, Brazil is among those countries with lowest percentages reporting "having voted or worked for candidates because of their position on environmental issues" (9% of Brazilians - HPS).
Since the Stockholm Conference in 1972, international agreements signed by nations worldwide have attributed fundamental importance to the communities in the definition of environmental public policies. Chapter 36 of the Agenda 21 is basically dedicated to the preparation of communities to take responsibility, individually and collectively, over development and environmental issues. In several of the international documents there are provisions and operational guidelines meant to place value not just in collective participation but also in the local spheres of government, closer to the environmental questions related to the citizen’s quality of life. Support for decentralization in environmental protection can be traced in the findings of the two waves of the WBTEDS survey (with high levels of adherence to the view that the municipal government and the individuals should have main responsibility for the solution of environmental problems).
On a similar fashion, when it comes to the perception of who should have the most important role in environmental protection, the mass public in the Rio das Velhas Basin Survey appears equally divided in relation to the federal government (24%), the local government (24%) and the individual citizen (24%).
The HPS found in 1992 that, worldwide, nationalgovernments were most likely to be seen as having primary responsibility for protecting the environment. However, Latin-American countries ranked highest among those nations whose citizens’ groups were more likely to be seen as having responsibility. This was the case for a majority in Brazil (60%) and highest percentage among all nations surveyed. For Brazilians, in 92, the government came second (26%) and business last and much lower (12%); the USIA Survey found that a plurality (43%) believed the government should have the primary responsibility, edging individual and citizen groups by a narrow margin (36%). Significantly fewer said business and industry should have principal responsibility (17%). Likewise, in the WBTES-92 and the WBTEDS-97 the government came first (51% and 43%, respectively), followed by "each one of us" (39% and 36%) and the municipal government (33% in both waves); again businessmen ranked very low (12% in both waves).
This support for decentralization of environmental policies and the reliance on the individual action seemed to some analysts to indicate a deeper level of pro-environmentalism. But is that the case? We found out that pro-environmentalism is broad (widespread concern), but how deep (behavior change) is it? One might think that before jumping to any conclusion we must not overlook some “inconsistencies” in our findings:
High levels of environmental concern, but poor knowledge of environmental issues
Environmental problems do not appear as salient in relation to other national issues, but majorities prioritize environmental protection in a trade-off with economic development
Willingness to join NGOs, but very poor knowledge of groups and very low reported activist behavior
Values and environmental attitudes do not correlate with environmental behavior (nor with willingness to adopt behavior) (see section below)
The proportion of people holding pro-environmental values is higher in the South-Southwest, but willingness to act is higher in the North-Northeast regions (see below).
This gap between values/attitudes and environmental behavior is not, however, a problem particular to Brazil. This happens to be one of the central issues analyzed in most international survey literature, as we shall see in the next section.