Content-Type: text/html No exceptions to the rule: The ubiquity of journalism norms throughout 29 years of environmental movement coverage No exceptions to the rule: The ubiquity of journalism norms throughout 29 years of environmental movement coverage No exceptions to the rule: The ubiquity of journalism norms throughout 29 years of environmental movement coverage A submission to the Association for Education in Journalism & Mass Communication Miami, Florida Linda Jean Kensicki Assistant Professor School of Journalism & Mass Communication University of Minnesota 304 Murphy Hall 206 Church Street S.E. Minneapolis, MN 55455, USA phone: 612 625 9388 fax: 612 626 8251 [log in to unmask] Abstract No exceptions to the rule: The ubiquity of journalism norms throughout 29 years of environmental movement coverage Research has suggested that the present media merger frenzy will result in one-dimensional content due to a reduced number of media outlets and pervasive cross-ownership. This research examined 1,180 articles about environmental pollution over 29 years from four very different newspapers. It was found that content was overwhelmingly skewed to be more relevant to those in upper socioeconomic classes regardless of socioeconomic readership, geographic location, specific issue or time. Heavily weighted coverage that could not have been found through random chance alone was attributed to pervasive journalistic norms. Introduction One of the principle points of opposition against the present media merger frenzy, is the fear that a monolithic media will create content that is one-dimensional. This prediction of what could appear in the future (or what has already befallen media), is based on a precept that past media offered a multiplicity of voices because of the sheer number of media outlets available. While there is limited research examining content from divergent media outlets over a protracted period of time on one specific issue, the widespread assumption is that smaller, alternative presses often cover issues very differently than mainstream outlets. Indeed, other limited studies have found this to be true in terms of specific journalistic practices, such as sources used. This research examines 29 years of media coverage concerning one pressing and enduring political group - the environmental movement. Further, this research examines content from a more conceptual approach of relevance within each gr oup examined. Content about the environmental movement and the issue of pollution were evaluated from 1,180 articles within four very different socioeconomic and geographic newspapers. This analysis aimed to uncover if there were any tangible differences between these media that would reflect each outlet's unique socioeconomic readership or geographic-specific issues. The four media selected for this study were 1) smaller newspapers that had readership from lower-socieoconomic groups in Los Angeles, California; 2) the Los Angeles Times, which had a significantly higher socioeconomic readership; 3) smaller newspapers that had readership from lower-socieoconomic groups in New York, New York; and 4) the New York Times, which had a significantly higher socioeconomic readership. This research examined the internal influences that have been found to have some level of influence on media content. Issues such as journalist values, journalist routines, organizational structures and economic forces were evaluated as possible contributors to thematic threads within media content from similar media sources. However, the stated purpose of alternative press and non-mainstream media has long been to offer a perspective often not found in standard media fare. Thus, it would appear unlikely that media outlets, which target vastly different readers than those typically found receiving mainstream press, would experience similar levels of internal influences. Disparity in content between mainstream and alternative media outlets would be expected if divergent newspapers did, in fact, aim to serve the unique constituencies of their readership. Before continuing on to an evaluation of internal influences on media content in general, this research will briefly review important historical events and natural disasters that impacted the environmental movement. This review is necessary to determine the types of issues that were prevalent during this period - and thus worthy of media coverage. The analysis of media content coincides with the third wave of environmentalism, roughly from 1972 until the present. The Environmental Movement The third wave of environmentalism began roughly three decades ago and at the turn of the twenty-first century, the label 'environmentalist' is one that almost all Americans feel comfortable with. What first brought environmental concerns to the press was unquestionably Rachel Carson's 1962 publication Silent Spring (Dunlap & Mertig, 1992; Schoenfeld, Meier & Griffin, 1979; Trefethen, 1975). Carson's Silent Spring tied together pesticides with mother's breast milk and exposed the complexities of environmental destruction with pragmatic implications on daily life. After Carson's publication, other outside factors gave rise to an unprecedented level of media coverage and public interest in environmental issues. The two most pronounced events were the first Earth Day in 1970 and the moon landing in 1969. In 1970, what many scholars credit as the official beginning of the modern environmental movement, twenty million people participated in national Earth Day celebrations (Dunlap & Gale, 1972). Other scholars suggest that the breathtaking view of earth from the moon launched widespread interest in environmental causes (Roth, 1978; Schoenfeld, Meier & Griffin, 1979). After Earth Day and the landing on the moon, the Arab Oil embargo of 1973 and 1974 stunned a machine-dependent United States and contributed to media's renewed interest in environmental issues . However, media interest and public opinion dropped off in the next few years. Many credit the pro-environmental Carter Administration of the late seventies with the sharp decrease in environmental activity, as activists may have believed that pressing issues were being handled by the government (Manes, 1990). A few years later, and less than one full year before Ronald Reagan took office, the World Climate Conference agreed to examine global warming more closely (Caldwell, 1992). The basic scientific understanding of global warming - that carbon dioxide emissions from fossil fuels (oil, coal etc), methane from agriculture and CFC's found in ozone depleting materials block infrared radiation from escaping the earth - was gaining prominence. Directly relevant to all life on earth, this trapped radiation was found to later increase surface heat on this planet, which then raises temperatures and sea levels (Wilson, 2000). As these revelations gained more credence, they also increased the amount of environmental coverage. In addition, President Reagan's recurrent themes of environmental deregulation for increased energy production may have led to increases in environmental reporting and environmental action (Portney, 1984). Those previously concerned about environmental issues but complacent due to their belief that government was handling the issue during the Carter administration, were suddenly frightened by the implications of a deregulatory government. This public interest in environmental issues stemming from fear of Reagan's deregulation was compounded by the non-enforcement of environmental policies by Anne Gorsuch of the Environmental Protection Agency and James Watt of the Department of the Interior (Dunlap & Mertig, 1992; Cutter et al. 1991; Dunlap, 1987; Milbrath 1984). In addition, Gillroy and Shapiro (1986) write that several worldwide incidents managed to keep environmental issues at the forefront of the public's mind during the eighties and nineties as well. Issues surrounding contaminated groundwater; air pollution; nuclear waste disposal; controversy at the Environmental Protection Agency; oil spills such as the Exxon Valdez disaster; the catastrophe in Bhopal, India; wild life degradation and the discovery of asbestos poisoning continued to push coverage of environmental issues. Indeed, accidents may have led to a continuance of coverage during the last forty years simply because there have been so many of them. During the first five years of the eighties alone, there were 7,000 accidents involving toxic chemicals (Diamond, 1985). The massive accidents in India and Ukraine are particularly important due to their sheer scope in devastation. Bhopal, India suffered through a catastrophic chemical leakage of methyl isocyanate (MIC) gasses from an American pesticide factory owned by Union Carbide in December of 1984. Since the original incident over 20,000 people died and over 500,000 people have been injured (Bhopal, 2001). Only two years later in April of 1986, the number four reactor in Chernobyl, Ukraine exploded. Thirty people died immediately while 15,000 people died soon afterwards. Since that time nearly 3.5 million people suffered illnesses because of radioactive contamination (Chernobyl, 2001). These events were soon followed by an accident within the United States borders during March of 1989. An Exxon ship leaked millions of tons of oil into the Alaskan waters, which killed countless animals that lived off or in the ocean. The nation watched hours of videotape showing several species struggling for their last breath in a sea of oil. This accident clearly raised the consciousness of Americans to the environmental dangers that can remain in an economy and government dependent on oil and machinery. These continuous national and global issues pushed environmental causes into the media spotlight fairly consistently over the period of examination. Undoubtedly, there were many local issues that unique to the area, depending upon geography. For example, in the case of this content analysis, it could be assumed that Los Angeles papers would report on automobile pollution much more than newspapers within New York City, where the overwhelming majority of the population depend on mass transit. Yet, regardless of issues with local prominence, some scholars have suggested that there are particular internal influences on media content that shape every news story. These influences have been found to differ in their degree depending on the specific type of media outlet. Internal Influences on Media Content Journalist Values As Reese (1990) found in his seminal piece discussing a socialist at the Wall Street Journal, journalists must work within an occupational 'ideology'. When a reporter deviates from this norm, she or he will face an inevitable reprimand or 'repair' from within the system. However, Reese (1990) makes the argument that deviations from a journalist's occupational ideology are extremely rare due to the force that these principles have in daily routines. Tuchman (1978) has noted that through the framework of deviance, media select certain issues only if they are congruent with news values and routines. Thus, Tuchman asserts that it is not the news topic itself that makes a subject newsworthy but its relationship to the values and routines of media. This media construction is the result of ideological, organizational, professional and individual conflicts with the news organization (Tuchman, 1978). By taking on a social constructionist perspective, it is important to examine what exactly political movements are deviating from when they generate coverage. It is within this theoretical vein that the values of journalists and the positions of social movements become so important. Gans (1979) found that a journalist's personal values and beliefs affect how she or he frame stories and what types of stories they report. He believed that journalists hold certain values that cannot be extracted from their writing and that these values are taught to journalists through his or her education, coworkers and superiors. Journalist values were said to be ethnocentricism, altruistic democracy, responsible capitalism, small-town pastoralism, individualism, moderate positions, adherence to social order, and national leadership (Gans, 1979). These values can have powerful effects on the news that journalists create. Ideally, a journalist will not bring his or her own values into the newsroom. Yet, according to Gans, these values are fostered and encouraged within media. While the presence or non-presence of journalistic values is still heavily contested, still other factors are present which strongly influence media content. Shoemaker and Reese (1991) purport that values do not influence media as much as journalist routines, organizational structures and economic forces. Journalist Routines Routines are the patterned, repeated practices of a larger group in which a person knowingly or unknowingly participates. For the journalist, routines make up their daily pattern of collecting information. The routines journalists use "form the immediate context, both within and through which these individuals do their job" (Shoemaker and Reese, 1991, 105). In essence, journalists act as gatekeepers of information to the public. They make decisions every day that influence the information we receive. Their routines of news sources and news gathering dictate - to a large extent - what a they will choose to include in a report and what they will choose to omit. As Berkowitz (1992) suggests, the media agenda is often constructed by powerful news sources. These consistent procedures of whom to contact in news gathering consequently define what is expected in the profession and what constitutes news. As in any occupation there are expectations of job performance and behavior. If, for example, a textbook publisher suddenly brought auto-mechanic tools to their job, an unspoken occupational routine will have been broken. These subtle, and not so subtle, nuances of job expectations are routinely rewarded and punished. If a reporter masters routine methods of newsgathering, such as asking the 'right' questions or finding the 'right' angle, they are praised for their professionalism (Tuchman, 1978). These routines stem from standards of importance inherent within the news business, according to Shoemaker and Reese (1991). News norms are the constructs that define and distill the information which audience members want to know about. A reporter's routines are formed after they understand what is 'important' within the news itself. Shoemaker and Reese (1991) found that a story's prominence, human interest, level of controversy, unusualness, timeliness and proximity dictate the importance of a news story. These are the textbook standards of newsworthiness are taught to journalists while they are in college and are obviously subjective in nature. Each journalist must decide for their own selves if a story fits within these guidelines. The prominence of a story is measured by the impact it has on people's lives. If a story is judged to have a high degree of impact, it is viewed as more important than other stories. Human-interest stories are valued because of the public's desired level of escapism, according to Shoemaker and Reese. In addition, controversy and conflict is important because "conflict is inherently more interesting than harmony" (Shoemaker and Reese, 1991, 91). The unusual is important because it breaks the everyday norms of society and thus is more interesting to the reader. Timeliness is a factor in news decision making because individuals want to know what is happening now in the world - not what is projected to happen in the future. Finally, the proximity of a news event is important because the closer an event or issue is to the audience, the more impact it has on their lives. If a story does not meet one of these criterion for newsworthiness, it is simply not reported because it is ruled unimportant. However, beyond the values media workers possess and the routines they follow in their daily patterns, the organizational structures and economic forces of media have a profound impact as well. Organizational Structures Shoemaker and Reese found that "the personal attitudes and values of news media owners may be reflected not only in editorials and columns, but also in news and features" (Shoemaker & Reese, 1991, 223). The researchers concluded that elites of news organizations not only have the power to hire and fire those that they feel do not contribute to the organization but also that they shape content through subtle suggestions within the newsroom itself. The organizational structures of media greatly influence news content, according to Gitlin (1980). For example, Ben Bagdikian (1990) notes that an editor first must decide what is newsworthy, the reporter then must decide what is worthy of notice at the scene and then thirdly, the editor must evaluate how to frame and place the resulting story. Thus, it is important to question what standards of newsworthiness an elite editor might have from a socioeconomic standpoint. In addition, as communication industries face tighter economic constraints, the pressures of 'producing' greatly affect the culture of the newsroom. As in recent cases from the Boston Globe and the Miami Herald, columnists can simply fabricate stories under the pressures to produce. Feasibly, reporters could also slant stories to please a dominating owner who is slashing the staff budget. The immediacy of a paycheck is just as important to the journalist as it is to the construction worker. It is na‹ve to think that somehow a journalist will rise above the business tactics of the real world and objectively report the news when threatened with the loss of economic income. Thus, the ultimate control of any business lies in its owner (Shoemaker & Reese, 1991). This fact does not change in the communication business. The owners have the final choice on whether or not a story should be pursued; whether the publication should produce the story; what types of news-gatherers should be hired and who should be fired. Obviously, the owner's own political ideology can become important in these decisions. While there is some movement within the occupational ideology of journalism set forth by the owners, the individual reporter must always face some form of 'repair' when deviating from strict norms (Reese, 1990). Repair from the organization can come in three different forms: distancing any threat from the work of an offending reporter; again emphasizing prevailing routines that prevent outside forces from influencing content; or marginalizing the offending reporter and the message (Reese, 1990). For many modern social movements challenging corporate inte rests and business tactics, the ideology of media owners is often counter to their own. Economic Forces An increasing amount of companies are being bought by larger corporations and owned cohesively under one large umbrella. Bagdikian (2000, viii) found that "for the first time in U.S. history, the country's most widespread news, commentary, and daily entertainment are controlled by six firms that are among the world's largest corporations, two of them foreign." While concentrated ownership undoubtedly shrinks the range of information that is allowed, cross-ownership in communications has allowed for greater potential for conflicts of interest (Shoemaker & Reese, 1991). In cross-ownership, as the possibility for conflict increases, the amount of 'acceptable' news decreases accordingly. For example, General Electric, which owns various companies in the businesses of plastics, appliances, aircraft engines, medical services, insurance, financial services, transportation and turbine machines for nuclear reactors, has joint ownership in NBC. Consequently, NBC owns television and radio stations, cable stations such as Bravo, American Movie Classics, CNBC and Court TV. Furthermore, NBC owns NBC Network News, which produces Dateline NBC, Nightside, Meet the Press and The Today Show (Miller, 1996). It does not require a suspension of logic to conclude that those on Dateline NBC may not be as prone to cover an insurance fraud of a General Electric subsidiary if they value their own employment. Similar examples of cross-ownership exist at the other two major networks - CBS and ABC. Surely, a reporter who finds corruption in a subsidiary company could face obstacles in producing a damaging piece of news. Internal cannibalizations cannot be endorsed by the elites in the business because it would mean a substantial monetary loss to a large portion of their ownings. Another example of economic forces that shape news content comes from advertisers. If a television show or a newspaper publication receives consistent subsidy from advertisers, they could logically consider several other options before exposing a damaging attack on that advertiser or reveal an ideal that an advertiser may not support. In addition, the target of a media publication may come under scrutiny from advertisers. Chomsky and Herman (1988) suggest that advertisers will not be likely to support media content that targets audiences with decreased buying power. Thus, in many cases, advertisers can dictate the audience of news as well as the content. These economic factors - corporate sponsorship, media ownership and advertisers - can have dramatic influences on news content. Shoemaker and Reese (1991, 167) stated in their book, Mediating the Message: Theories of Influences on Mass Media Content, that "the greater the physical distance of the owners from the community being served, the more community interests may take a backseat to corporate and economic factors." Without an undivided interest in the surrounding community, media organizations can become indebted to corporate subsidies. This obligation can translate into altered media content. While these internal influences are said to have a powerful point in the argument against media convergence, there are not consistent studies that examine media content over time from very different media outlets, to see if the content is actually different according to media readership. This research attempts to fill that void. Methodology Content Selection Two cities were selected for examination because each city had at least two newspapers and the two newspapers in each city had the largest differences in household income readership levels across the country. This factor was significant as it monitored what those in lower socioeconomic classes were actually reading across the United States. The two cities that meet this criteria were Los Angeles, California and New York, New York. The fact that the New York Times ended up being included in this study undoubtedly gives a more accurate summation of what all Americans were reading about air pollution due to the powerful influence that the New York Times has on other papers throughout the country (Dreier, 1982; Gans, 1979; Reese & Danielian, 1989; Shoemaker & Reese, 1991). Furthermore, both of these cities have been consistently ranked within the top ten air polluted cities over the last forty years, according to Environmental Defense (2000). Newspapers in particular were chosen due to previous research findings that suggest it is newspapers, not television, that set the public agenda (Epstein, 1978). As McCombs (1978) wrote, newspapers tend to have a stronger agenda setting effect on the population due to their powerful influence in creating, organizing, and sustaining the public agenda. Within each city, the newspaper with the highest socioeconomic readership was first selected. The result was that the New York Times (daily circulation of 1,086,293) with average readers having a household income of $66,700 per year and the Los Angeles Times (daily circulation of 1,078,186) with average readers having a household income of $50,000 per year (McClintic, 1998) were chosen for inclusion in the study. Within the Los Angeles area, the San Bernardino Sun, with an average household income of $36,400 (McClintic, 1998) was sampled. In New York City, the publication examined to represent content targeted to lower socioeconomic classes was the New York Daily News, with an average household income of $42,200 (McClintic, 1998) and a daily circulation of 730,761 readers. These newspapers were selected due to their relative difference in socioeconomic readership from the first two newspapers chosen - the New York Times and the Los Angeles Times. The difference in household income between the two primary New York papers was $24,500 and the difference between the two Los Angeles papers was $13,600. As an additional component to this research, other newspapers with low socioeconomic readership were sampled within a general pool. In the Los Angeles area, listings included the Herald Examiner and the Los Angeles Daily News (daily circ. 200,387), as well as several black press newspapers, such as the Los Angeles Herald Dispatch, the Los Angeles Sentinel, the Southeast Wave Star and the Southside Journal. There is not conclusive socioeconomic data available for all of these newspapers. However, the newspaper with the highest socioeconomic readership of this group is the Los Angeles Daily News (average household income of $44,500), whose readers still have incomes far below those who read the Los Angeles Times (McClintic, 1998). The other general lower socioeconomic publications within the New York City area included the New York Post as well as black press newspapers titled the New York Voice, the New York Beacon and the Amsterdam News. Using other low-income newspapers for this study was necessary for several reasons. Primarily, in many years there were simply not enough articles for examination. This could have been due to an actual dearth in content, but was more likely a breach in indexing the data. The only source of indexes for lower 'prestige' papers was NewsBank. However, this source was highly selective - so much so that in certain years, the preponderance of articles available were from one state in particular if a newsworthy event happened in that state. However, even if this disparity is a result of minimal coverage, nothing in agenda setting is dependent upon absolute numbers of frequency. What remains essential is that these issues were being covered in papers targeting lower socioeconomic groups, which this sample proves correct. Articles from all of the newspapers, both upper and lower socioeconomically targeted, were retrieved if the term 'air pollution' was in the headline or lead paragraph. This was done to see if there was any linkage made by the media from pollution to the environmental organizations that fight to alleviate the problem. The issue of air pollution specifically was chosen due to the finding of several studies that pollution affects those in lower socioeconomic classes at a highly disproportionate rate (Bryant & Mohai, 1992; Bullard, 1994; Lee, 1992; Schwab, 1994; White, 1998). Thus, one would expect to find a large amount of coverage applicable to those in lower socioeconomic classes. Further, the issue of air pollution in specific, was chosen as it was a consistent theme throughout the 29-year period examined. Articles were sampled from 1972 until 2000. The initial date was selected due to the fact that indexes for large newspapers, outside of the New York Times, generally began that year. In addition, other research has found that coverage of environmental issues and the environmental movement was sparse during the sixties and grew exponentially after the first Earth Day in 1970 (Schoenfeld, Meier & Griffin, 1979). Thus, examining content in the sixties would not have added much supplementary information to the study. Two coders examined all the newspaper articles. Content from newspapers targeting lower socioeconomic classes were first collected to ascertain the number of total articles available during this period. This collection yielded an average of 12 articles per year over the 29-year period. Due to the large amount of newspaper content available from the New York Times and the Los Angeles Times, these newspapers were randomly sampled in proportion to the amount found from lower socioeconomic newspapers. Thus, twelve articles from the New York Times and the Los Angeles Times were randomly selected, through every nth article listed, on a yearly basis. This resulted in an evaluation of 1,392 articles (696 articles from newspapers with a lower socioeconomic readership divided evenly between papers in New York and Los Angeles and 696 articles from newspapers with an upper socioeconomic readership divided evenly between the New York Times and the Los Angeles Times). After removing editorially based pieces, the final tally of articles included for study was 1,180. From this number, 331 were from the New York Times and 349 were from the Los Angeles Times, for a total of 680 articles from an upper socioeconomic readership. The remaining 500 articles were from a lower socioeconomic newspaper (256 articles came from New York papers and 244 from Los Angeles papers). More specifically, 99 articles were from the New York Daily News, 40 from the New York Post, 56 from other New York papers and 61 from other lower socioeconomic New Jersey papers. Finally, 80 articles were from the Los Angeles Herald Examiner, 34 from the San Bernardino Sun, 112 from the Los Angeles Daily News, and 18 from other lower socioeconomic California papers. The articles from general newspapers in New York, New Jersey and California with lower socioeconomic readers, were only included if they were addressing issues of New York City and Los Angeles respectively and were rel atively close in geographic proximity. Operationalization The issue of air pollution was comprised of three main attributes: cause, effect and responsibility. Each attribute was then divided into several frames: government, industry, social movement, individual and natural. Coding was not restricted to only one frame per attribute. Rather, up to five frames per each attribute were possible (although, extremely unlikely). Coding in this manner revealed whom the media suggested as the principal agent causing pollution; what exactly the effects have been (and on whom); and who was responsible for rectifying the situation. This proved especially important for an analysis of different socioeconomic groups. Lower socioeconomic classes have a distrust of industry and government (Howell & Fagan, 1988). The operationalization of each frame within its respective issue attribute was constructed as follows: Causal Attribute Government: government policy/activity/regulations as cause of environmental problems Industry: industrial activity/products as cause of environmental problems Social Movement: movement activity as cause of environmental problems Individual: individual activity as cause of environmental problems Natural: natural biological processes as cause of environmental problems Effect Attribute Government: environmental problem affecting government budget Industry: environmental problem affecting industry prices, profitability, marketplace Social Movement: environmental problem affecting movement participation, effectiveness Individual: environmental problem affecting health, financial status, land ownership Natural: environmental problem affecting earth Responsibility Attribute Government: government activity to propose/rectify environmental problems Industry: industry activity to propose/rectify environmental problems Social Movement: movement activity to propose/rectify environmental problems Individual: individual activity to propose/rectify environmental problems Natural: rectifying environmental problems naturally Questions of relevance in media content exposed differences in socioeconomic-specific responses to environmental concerns. Specific methodological steps were taken to help reveal this phenomenon. For example, if the "individual" frame was chosen within any attribute then the precise reason for this choice was noted. This process helped reveal if particular individual frames were relevant to the lower socioeconomic individual receiving the media message. These individual frames were coded according to relevance along socioeconomic divisions. For example, effects on personal health presumably had equal relevance across all socioeconomic classes while land ownership should have more relevance with higher socioeconomic classes more than lower socioeconomic classes. While a case could be made that health effects on lower socioeconomic groups were more pronounced due to a lack of medical care, the effects themselves before treatment, were deemed equal. In addition, while health effects of pollution may manifest themselves differently in divergent classes (i.e. diabetes in African Americans versus heart disease in middle class whites), the presence of health effects itself was determined to be equal for the purposes of this study. For individual frames, coders were given the following choices: personal health, individual auto use, disregard for non-automotive transportation such as bicycle use and walking, individual population control, smoking, improving household efficiency (heating appliances, burning wood, coal, oil or gas, water piping, foam insulation, building materials, etc.), household gardening (native plants & trees), minimizing consumable consumption, boycotting polluting businesses and other. If other was chosen, then the precise reason was noted. Within these options, personal health was viewed as applicable to all socioeconomic levels. Smoking was viewed as slightly more relevant to those in lower socioeconomic classes due to the strong inverse relationship between smoking and income in adults worldwide although clearly, there are those in all classes are smokers (Beech, Droker, Pree-Cary, Harris, 1997; World Health Organization, 1997). Individual auto use, disregard for non-automotive tr ansportation, minimizing consumable consumption and boycotting polluting businesses was constructed as slightly more applicable and relevant to upper socioeconomic classes, due to the intrinsic availability of material options as socioeconomic status increases. Finally, improved household efficiency, household gardening, and individual population control were viewed as more relevant to upper socioeconomic classes. Household efficiency and home gardening were included in this category due to the relatively large amount of money needed for a down payment on a home and the sustained funds needed for mortgage payments. Population control was included as an issue more relevant to upper socioeconomic classes because recent research has shown that as individual incomes and education rise, birth rates drop significantly (CNN, 1999; Population Council, 2001). In an effort to ascertain political apathy towards the issue of pollution across all socioeconomic classes, coders were asked to discern what the text suggested was the likelihood of solving pollution. Coders were given five choices along a Likert scale: extremely unlikely, unlikely, not mentioned, likely, extremely likely. Only direct mentions concerning the likelihood of solving pollution were coded. The question of personal apathy was also fundamental in searching the text for mentions of national and/or local environmental organizations. It was presumed that by connecting political organizations with the cause itself, a stronger connection to individual and local activism could be created. If an environmental organization was mentioned in an article about pollution, then apparent tactics of the movement (corporate sponsorship, journalist liaisons, political associations, organized events, etc.) were noted to find if there was a correlation between tactics used and acquiescence with journalist values, routines, organizational structures and economic forces in media. If coverage of a social movement was coded as negative, this position was further explicated through movement qualities noted earlier (limited opposition, reduced internal struggle, etc.) that are more relevant to those in upper socioeconomic classes. Further, negative coverage was categorized through media framing techniques (trivialization, polarization, internal dissention, etc.). This helped ascertain if coverage of these groups differed across media organizations, which was viewed as a reflection of perceived deviance within media institutions. This was important to discover if these factors helped construct a perception of environmentalism in this country that could have influenced participation within the movement. Direct mentions of the term "activist" and "environmentalist" were coded within article content. This was completed to gain a stronger understanding of the level of individual responsibility for environmental pollution conferred by media. In addition, any articles suggesting pollution as one rooted in a struggle for civil rights were coded in this study as well as mentions of socioeconomic factors in environmental coverage. A connection between civil rights and air pollution was determined if race, individual identity or civil rights organizations were mentioned in relation to environmental destruction. Connecting air pollution to socioeconomic factors was only possible if income, specific economically divided geographic locations or education levels were mentioned. As previous research has shown, connecting social activist causes with civil rights issues of inequality generally increases involvement from lower socioeconomic classes. Again, under the guide of agenda setting, civil rights and socioeconomic factors were seen as relevant arguments for those in the lower socioeconomic strata. Coders were also asked to determine the level of scientific jargon found in article content. Coders were given five choices: easily comprehended (never used scientific terminology); generally understood (occasional use of scientific terminology; some difficulty in comprehension (moderate use of scientific terminology; difficult to understand (frequent use of scientific terminology); or extremely difficult to comprehend (heavy use of scientific terminology). It was suggested that as the level of scientific language increased, the issue became more relevant to those in upper socioeconomic classes due to the conflation of education and income in the United States. Finally, coders evaluated the general focus of the article. Coders were provided eight options for this macro-level, thematic frame. Their choices were government regulation, political debate, judicial legislation, health hazard, scientific study, individual behavior, public protest or other. Results Frequencies The analysis of 1,180 articles between two coders generated a high 91.25 percent inter-coder reliability score for media attributes. Scott's Pi was computed at .60, representing the inter-coder agreement after chance has been removed. While still generating a far higher number than what would be expected by chance alone, the somewhat lower Pi score was as much a reflection of heavily weighted cases within the categorical variables (as was the case in this study) as it was a removal of chance from the inter-coder reliability percentage (Riffe, Lacy & Fico, 1998). For example, when one assumes an even dispersion of cases within the twenty variables coded for this analysis (i.e. a dichotomous variable would assume a .5 probability for each of the two cases), Scott's Pi increases to .88. As the unit of analysis, newspaper articles were dispersed fairly equally across upper and lower socioeconomic papers. This sample was purposively collected to examine differences in both geography and socioeconomic readership. Thus, the equal dispersion is a reflection of the sampling scheme. Twenty eight percent of articles came from the New York Times, which represented content targeting upper socioeconomic classes within the first market. The newspapers with lower socioeconomic readers in the New York area (New York Daily News, 8.4; New York Post, 3.4; other, 9.9) comprised roughly twenty two percent of total articles. Twenty one percent of articles sampled came from lower socioeconomic newspapers within the Los Angeles area (Los Angeles Herald Examiner, 6.8; San Bernardino Sun, 2.9; Los Angeles Daily News, 9.5; other, 1.5). Finally, 29 percent of the articles reviewed for the content analysis came from the Los Angeles Times. All articles sampled were coded along three attribute dimensions: cause, effect and responsibility. Industry was found to be the overwhelming cause of pollution within the United States (72.9 percent). The standard error of this proportion was .009, suggesting that the 'industry' cause frame in the general media population could be as high as 73.8 percent or as low as 72 percent. The relatively small standard error of proportion (SE(p)) found throughout this study was a reflection of the large sample size and the lack of variability in case values (Riffe, Lacy & Fico, 1998). A majority of 59.1 percent (SE(p)=.0101) of content suggested 'neutral' effects of the air pollution problem. The effects dimension of air pollution produced the largest percentage of 'individual' frames with 31.3 percent. Only 7.7 percent of newspaper content framed the earth or other natural phenomenon as experiencing the effects of air pollution. Government was framed as the responsible agent for air pollution in 78.7 percent (SE(p)=.0084) of content. Of this small 'individual' percentage, frames were divided among 'auto use' (16.7 percent, SE(p)=.1522) and 'minimizing consumable goods' and 'boycotting businesses' (33.33 percent each, SE(p)=.1922). A near totality of content found absolutely no mention of any solutions for air pollution (95.4 percent, SE(p)=.0043). Ninety percent of articles (SE(p)=.0061) did not mention local environmental organizations while eighty three percent (SE(p)=.0077) of articles did not mention national environmental organizations. If national environmental movements were mentioned in content about air pollution, it was almost exclusively as a brief mention without any accompanying information. In fact, 99 percent of content (SE(p)=.0041) that mentioned a national environmental organization was coded as 'neutral.' Movements were found to 'mainstream their views' in 3.3. percent of coverage. Organized events (2.2 percent) and an 'expanded political terrain' (2.1 percent) comprised the bulk of remaining content. Ninety-nine percent of content (SE(p)=.0006) did not suggest any unique qualities of the environmental movement as a whole. Ninety eight percent of content (SE(p)=.0026) did not mention the term 'activist.' Further, eighty five percent (SE(p)=.0072) of content did not use the term 'environmentalist.' Civil rights and socioeconomic class factors were not mentioned in almost all of the content covering 29 years (99.8 percent, SE(p)=.0009 and 98.7 percent, SE(p)=.0023 respectively). Fifty-five percent of language (SE(p)=.0102) was coded as easy and 40.7 percent (SE(p)=.0101) as generally understandable. Finally, the general macro frame of newspaper articles over 29 years of air pollution coverage was found to be government regulation (52.1 percent, SE(p)=.0102). Scientific studies (11.1 percent), judicial legislation (6.7 percent) and other frames (16.9 percent) comprised the bulk of remaining air pollution macro-frames. Remaining content was found to promote general apathy: 95.4 percent of coverage did not mention any solution to pollution; 90 percent did not mention local environmental organizations; 83 percent did not mention national environmental organizations; 99 percent of coverage that did address environmental organizations did so in a neutral frame; 98 percent of coverage did not mention the term 'activist'; and 85 percent did not mention the term 'environmentalist'. In sum, coverage attributes were found to be overwhelmingly relevant to upper socioeconomic individuals in particular in all media outlets: 72.9 percent of coverage suggested industry as the cause of pollution; 58.4 percent of individual cause frames suggested 'auto use' as the individual cause of pollution; 59.1 percent found neutral effects of pollution; 78.7 percent framed government as the responsible agent; individual responsibility frames were divided between auto use, minimizing consumable goods and boycotting businesses as the individual responsibilities of pollution; the majority of environmental organizations mentioned in the content were entrenched in political system; 99.8 percent of coverage did not mention civil rights; 98.7 percent of coverage did not mention socioeconomic class factors; and 52.1 percent of articles presented the general macro frame as government regulation. Only two variables contradicted this overwhelming socioeconomic bias within media content. The first was the language difficulty measure, which found that 55 percent of content was easy to understand. The second was the individual effect attribute dimension of pollution that found 98.9 percent of effects within the attribute to be personal health. Associations An examination of the relationship between collapsed coded variables and two newspaper categories according to geography (New York and Los Angeles region), found six significant relationships with very weak strength between the two variables and no measurable effects (30 percent), one relationship was found significant with actual strength between the two variables (5 percent), eleven insignificant relationships between variables (55 percent), and two relationships that could not be measured due to skewed data uniformly across all newspapers (10 percent). The relationship between collapsed coded variables and two newspaper categories according to socioeconomic readership levels (high socioeconomic readership and low socioeconomic readership) found five significant relationships with very weak strength between the two variables and no measurable effects (25 percent), no relationships with actual strength between the two variables (0 percent), thirteen insignificant relationships between variables (65 percent), and two relationships that could not be measured due to uniformly skewed data across all newspapers sampled (10 percent). Thus, all relationships except one that found strength between aggregated demonstrated movement tactics and geographical area of publication, found either no relationship between variables, weak relationships between variables with no measurable effects or could not be computed due to heavily skewed data across all newspapers. Non-collapsed 2 x 2 tables found no measurable significant relationships and only six (30 percent) weak relationships with no measurable effects. Collapsed data across geography found the only strong and significant relationship (5 percent), and six (30 percent) weak relationships were found that demonstrated no measurable effects. Collapsed data across socioeconomically different newspapers found no measurable significant relationships and only five (25 percent) weak relationships with no measurable effects. Computation of significance in relationships with no measurable effects were attributed to the large sample size. Thus, if a relationship found si gnificance but showed only very weak effects on variables coded, it was concluded that no meaningful relationship existed. In comparing these overwhelming frequencies across newspapers, the data was translated into the following categories: Newspaper associations Significance Area Causal attribute dimension Weak Geography Effect attribute dimension None Responsibility attribute dimension Weak Geography Individual cause frames Weak Socioeconomic Readership Individual effect frames None Individual responsibility frames Not interpretable Solution likelihood Weak Geography & Socioeconomic Readership Local environmental org Weak Geography National environmental org Weak Socioeconomic Readership Movement tactic Strong Geography Type of coverage None Negative coverage frames Not interpretable Movement qualities Not interpretable Movement quality frames None Activist mentions Weak Environmentalist mentions Weak Socioeconomic Readership Civil rights None Socioeconomic factors None Language difficulty Weak Geography Macro-frame Weak Socioeconomic Readership Thus, from the twenty statistically measurable relationships between coded variables and four newspaper categories (New York Times, New York low S.E.S., Los Angeles Times, Los Angeles low S.E.S), six were found significant but showed very weak strength in relationships and no measurable effects between variables (30 percent), no relationships were found that showed actual strength between variables (0 percent), two were found insignificant (6.6 percent), and twelve were not interpretable because of heavily weighted data - demonstrating extreme uniformity across papers (60 percent). Finally, the three central attribute dimensions that were coded (cause, effect and responsibility), were found to be extremely similar across newspapers (Table 1). For example, the largest difference between papers along any frame within an attribute dimension was 3.3 percent. Table 1 Synopsis of Attribute Dimensions Across Newspapers Attribute Dimensions NY Times NY (Low SES) LA Times LA (Low SES) Cause Neutral 4.7% 3.4% 5.9% 6.3% Government .8% 1.0% 1.2% .7% Industry 25.3% 24.9% 24.0% 22.5% Social Mvmt 0% 0% 0% 0% Individual 2.2% 3.7% .6% 2.3% Natural .1% .1% 1.4% 1.3% Effect Neutral 21.5% 18.4% 18.2% 20.6% Government .3% .5% .1% .1% Industry .4% .5% .1% 0% Social Mvmt 0% .1% 0% 0% Individual 8.4% 11.6% 11.0% 10.9% Natural 2.5% 2.0% 3.6% 1.5% Responsibility Neutral 4.8% 3.9% 7.1% 5.0% Government 27.2% 27.5% 24.6% 25.7% Industry .9% 1.5% .8% 1.6% Social Mvmt .2% .3% .1% 0% Individual 0% 0% 0% .2% Natural 0% 0% .4% .6% TOTAL 100% 100% 100% 100% Column N 1992 1536 2082 1470 Total N = 7080 Indeed, while this research does not allow for an examination of which news outlet actually set the agenda for the other, it clearly indicates that there was strong homogeneity across papers. Differences in total percents across dimensions between newspapers were extremely small. The largest difference between papers was only +/-.5 percent (Table 2). Table 2 Percentage of Differences in Attribute Dimensions Across Newspapers New York (Low SES) Los Angeles Times Los Angeles (Low SES) New York Times -0.1 0.5 0 NY (Low SES) 0.3 0.1 Los Angeles Times -0.2 Perhaps even more striking is the summation of data through a Spearman rank correlation coefficient (Rho). The correlation between papers is almost a perfect score of 1 (Table 3) , reflecting almost total agreement across geography and across socioeconomic readership. Table 3 Spearman's Rho of Attribute Dimensions Across Newspapers New York (Low SES) Los Angeles Times Los Angeles (Low SES) New York Times .9737 .9601 .9759 NY (Low SES) .9576 .9704 Los Angeles TImes .9771 Discussion No substantive longitudinal data tracking environmental participation among various socioeconomic groups exists. However, several different studies have found exclusively consistent upper socioeconomic membership throughout the last three decades. The conclusion from these varying studies is that the mainstream environmental movement continues to be overwhelmingly white and middle to upper class (Freudenberg & Steinsapir, 1992; Gottlieb, 1993; Morrison, 1973; Morrison & Dunlap, 1986; Tremblay and Dunlap, 1978). However, environmental concern from African Americans and other minority groups has been found to be as high or higher than European Americans. Mitchell (1979) found 64 percent of blacks and 63 percent of whites were sympathetic with the environmental movement. In 1982, the Roper Organization found that 44 percent of African Americans believed that environmental regulation should be expanded - this was in contrast to only 36 percent of white respondents. In addition, Mohai (1990) found that blacks are as concerned or even more concerned about environmental issues as their white counterparts. Finally, the National Opinion Research Center found that in every single year from 1973 until 1988 more blacks than whites believed that the nation spends 'too little money' on environmental protection (Mohai, 1990). During the eighties and nineties, cohesive longitudinal studies reexamined concern among various socioeconomic classes and concluded that while activists are "drawn disproportionately from the upper-middle class, concern tends to cut across all socioeconomic categories" (Mohai, 1985, 821). Thus, while several other factors may be at work in keeping participation numbers low, such as Maslow's (1954) hierarchy of needs, elites already involved in the movement (Bullard & Wright, 1989; Taylor, 1989), and a lack of attention to civil rights within the movement (Bullard, 1990), media representation of environmental issues as extremely monolithic may be an integral component of this complex problem. This representation is crucial to an environmental movement that has increasingly addressed issues pertinent to an extensive cross-section of society (Van Liere & Dunlap, 1980). In the broadest sense, those in the lower socioeconomic classes suffer the most from health problems that are caused or exacerbated by environmental problems (Eckholm, 1977). It could certainly be argued that those in the bottom socioeconomic strata presumably have the largest stake in environmental improvement in whatever form that takes. However, much of this information has remained untold in smaller, alternative presses purported to address the concerns of their readers. Heavily weighted coverage that could not have been found through random chance alone was documented through both the runs test and the binomial test. These tests indicated an overwhelming skew in the data among all dichotomous variables. The overwhelming frequency of certain frames was pervasive among all different types of newspapers. This finding suggested that coverage was in some way biased and statistically 'unfair.' All but one of the relationships between newspapers and coded variables found either no relationship, weak relationships between variables that showed no actual effects or data could not be computed due to heavily skewed data uniformly across all newspapers. It is important to reiterate here that the large sample size was certain to inflate even the weakest relationships that showed no real effect between variables. Indeed, regardless of socioeconomic readership, geographic location, specific issue or time, coverage concerning the environmental movement was relatively unchanging. Through the longitudinal data collected, claims as to the ubiquity of this conclusion were possible. Four different newspaper types, each with distinct socioeconomic readership and geographic location, all had remarkably similar local coverage. Locality is important in this framework because there were certainly singular issues that each paper was dealing with, but yet all newspapers showed a rather monolithic presentation of the environmental movement and of air pollution. In fact, the majority of dimensions coded had frames encompassing over ninety percent of coverage throughout 29 years. One of the principle points of opposition against the present media merger frenzy, is the fear that a monolithic media will create content that is one-dimensional. This prediction of what could appear in the future, is based on a possible misconception that past media offered a multiplicity of voices because of the sheer number of media outlets available. An overwhelming uniformity across geographic boundaries, across divergent issues and topics, and across socioeconomic levels of readership was found in this research. Whether the coverage was from a small alternative press, or the New York Times, the content was unchanging. This could only be explained by a pervasive strength in journalist norms, routines and values. In this case, the norms of journalism played a stronger role in defining environmental coverage than did the readership market, the geographic location or the specific issues addressed. Bibliography Bagdikian, B. (2000). The media monopoly, 6th edition. Boston, MA: Beacon Press. Bagdikian, B. (1990). The media monopoly, 3rd edition. Boston, MA: Beacon Press. Beech, Bettina M., Droker, S., Pree-Cary, J. & Scarinci, I. (2000). Smoking initiation among low-income African Americans: Qualitative assessment of contributing factors. Paper presented to the American Public Health Association: (November). Berkowitz, D. (1992). Who sets the media agenda? The ability of policymakers to determine news decisions. In J. D. Kennamer (Ed.), Public opinion, the press, and public policy (pp. 81-102). Westport, CT: Praeger. Bhopal (2001). Bhopal. http://www.bhopal.net. Bryant, B. & Mohai, P. (Eds.). (1992). Race and the incidence of environmental hazards: A time for discourse (pp.1-9). Boulder, CO: Westview Press. Bullard, R. D. (Ed.) (1994). Unequal protection: Environmental justice and communities of color. San Francisco, CA: Sierra Club Books. Bullard, R. D. (1990). Dumping in Dixie: Race, class, and environmental quality. Boulder, CO: Westview. Bullard, R. D. & Wright, B. H. (1989). The anatomy of black grassroots environmental activism: Case studies from the southern U.S. Paper presented at the annual meeting of the Mid-South Sociological Association, Baton Rouge. Caldwell, L. (1992). Globalizing environmentalism: Threshold of a new phase in international relations. In R. E. Dunlap & A. G. Mertig (Eds.), American Environmentalism: The U.S. Environmental Movement, 1970-1990 (pp. 63-76). New York, NY: Taylor & Francis. Chernobyl. (2001). Chernobyl. http://www.chernobyl.com. Chomsky, N. & Herman, E. S. (1988). Manufacturing consent: The political economy of the mass media. New York, NY: Pantheon Books. CNN (1999). Education, income tied to world population growth. http://www.cnn.com/US/9910/13/population.youth.surge/#r. Cutter, S., Renwick, H. & Renwick, W. (1991). Exploitation, conservation, preservation: A geographic perspective on natural resource use. New York, NY: John Wiley and Sons. Diamond, S. (1985). U.S. toxic mishaps in chemicals put at 6,928 in 5 years. The New York Times, Oct 3, A1, D22.Dillman, D. A. & Christenson, J. A. (1975). The public value for air pollution control. Cornell Journal of Social relations, 10, (Spring), 73-95. Dreier, P. (1982). The position of the press in the U.S. power structure. Social Problems, 29, 298-310. Dunlap, R. E. (1987). Polls, pollution, and the politics revisited: Public opinion on the environment in the Reagan era. Environment, 29, July/August, 6-11, 32-37. Dunlap, R. E. & Gale, R. P. (1972). Politics and ecology: A political profile of student eco-activists. Youth and Society, 3, 379-397. Dunlap, R. E. & Mertig, A. G. (1992). The evolution of the U.S. environmental movement from 1970 to 1990: An overview. In R. E. Dunlap & A. G. Mertig (Eds.), American environmentalism: The U.S. environmental movement, 1970-1990 (pp. 1-10). New York, NY: Taylor & Francis. Eckholm, E. P. (1977). The picture of health: Environmental programs. Society, 14, 63-66. Environmental Defense (2000). Pollution locator: Criteria air pollution - A national overview. http://www.scorecard.org/env-releases Freudenberg, N. & Steinsapir, C. (1992). Not in our backyards: The grassroots environmental movement. In R. E. Dunlap & A. G. Mertig (Eds.), American environmentalism: The U.S. environmental movement, 1970-1990 (pp. 27-37). New York, NY: Taylor & Francis. Gans, H. J. (1979). Deciding what's news. New York, NY: Pantheon Books. Gilroy, J. & Shapiro, R. (1986). The polls: environmental protection. Public Opinion Quarterly, 50, 270-279. Gitlin, T. (1980). The whole world is watching: Mass media in the making and unmaking of the new left. Berkley, CA: University of California Press. Gottlieb, R. (1993). Forcing the spring: The transformation of the American environmental movement. Washington, D.C.: Island Press. Howell, S. & Fagan, D. (1988). Race and trust in government: Testing the political reality model. Public Opinion Quarterly, 52, 343-350. Lee, C. (1992). Toxic waste and race in the United States. In B. Bryant & P. Mohai (Eds.), Race and the incidence of environmental hazards: A time for discourse (pp.10-27). Boulder, CO: Westview Press. Manes, C. (1990). Green rage: Radical environmentalism and the unmaking of civilization. Boston, MA: Little, Brown. Maslow, A. H. (1954). Motivation and personality. New York, NY: Harper & Row. Maslow, A. H. (1954). Motivation and personality. New York, NY: Harper & Row. McClintic, R. (Ed.). (1998). Marketer's guide to media: 1998-1999, volume 21. New York, NY: ASM Communications, Inc. Milbrath, L. (1984). Environmentalists: Vanguard for a new society. Albany: State University of New York Press. Miller, M. C. (1996). Free the media. The Nation, 262, 9-12. Mohai, P. (1985). Public concern and elite involvement in environmental-conservation issues. Social Science Quarterly, 66, 820-838. Morrison, D. E. (1973). The environmental movement: Conflict dynamics. Journal of Voluntary Action Research, 2, (Spring), 74-85. Morrison, D. E. & Dunlap, R. E. (1986). Environmentalism & elitism: A conceptual and empirical analysis. Environmental Management, 10, 581-589. Population Council (2001). Six billion and counting. http://popcouncil.org/ppdb/six_billion.html. Portney, P.R. (Ed.). (1984). Natural resources and the environment: The Reagan approach. Washington, DC: Urban Institute Press. Reese, R.. (1990). The news paradigm and the ideology of objectivity: A socialist at the Wall Street Journal. Critical Studies in Mass Communication, 7, 390-409. Reese, S. & Danielian, L. (1989). Intermedia influence and the drug issue: Converging on cocaine. In P. J. Shoemaker (Ed.), Communication campaigns about drugs: Government, media, public (pp.29-46). Hillsdale, NJ: Lawrence Erlbaum. Riffe, D., Lacy, S., & Fico, F. (1998). Analyzing media messages: Using quantitative content analysis in research. Mahwah, NJ: Lawrence Erlbaum Associates. Roth, C. (1978). Off the merry-go-round, on the escalator. In W. Stapp (Ed.). Environmental education perspectives and prospectives. (pp. 12-22). Columbus, OH: ERIC/SMEAC.Samdahl, D. M. & Robertson, R. (1989). Social determinants of environmental concern: Specification and test of the model. Environment and Behavior, 21, (1), 57-81. Schoenfeld, A C., Meier, R. F. & Griffin, R J. (1979). Constructing a social problem: The press and the environment. Social Problems, 27, 38-61. Schwab, J. (1994). Deeper shades of green: The rise of blue-collar and minority environmentalism in America. San Francisco, CA: Sierra Club Books. Shoemaker, P. & Reese, S. (1991). Mediating the message: Theories of influences on mass media content. White Plains, NY: Longman Publishing Group. Trefethen, J. (1975). An American crusade for wildlife. New York, NY: Boone and Crockett Club. Tremblay, K. R., Jr. & Dunlap, R. E. (1978). Rural-urban residence and concern with environmental quality: A replication and extension. Rural Sociology, 43, (Fall), 474-491. Tuchman, G. (1978). Making news: A study in the construction of reality, New York, NY: Free Press. Van Liere, K. D. & Dunlap, R. E. (1980). The social bases of environmental concern: A review of hypotheses, explanations and empirical evidence. Public Opinion Quarterly, 44, 181-197. World Health Organization (1997). Tobacco or health: A global status report. Paper presented to World Health Organization: Geneva, Switzerland. Wilson, K. (2000). Communicating climate change through the media: Predictions, politics and perceptions of risk. In S. Allan, B. Adam & C. Carter (Eds.), Environmental risks and the media (pp. 201-217). New York, NY: Routledge.World Health Organization (1997). Tobacco or health: A global status report. Paper presented to World Health Organization: Geneva, Switzerland. Wright, S. (1975). Explaining environmental commitment: The role of social class variables. Paper presented at the annual meeting of the Midwest Sociological Society, Chicago.