FACTORS INFLUENCING THE PUBLIC'S VIEW Perceptions of Risk and Benefit Studies of risk perception reveal a gap between lay people's judgment of nuclear hazards and the risks estimated by technical risk assessments. In a 1979 study by Decision Research two small groups of informed people in Eugene, Oreg. (college students and members of the League of Women Voters) were asked to compare the benefits and risks of a variety of activities, ranging from smoking to vaccinations to swimming. The benefits of nuclear power were viewed as negligible, and the risks were judged to be almost as great as motor vehicle accidents, which claim about 50,000 lives each year (68). Although the two groups estimated that the number of deaths from nuclear power in an average year would be fewer than the number of deaths from any of the other activities or technologies, they used a very high multiplying factor to indicate how many deaths would occur in a "particularly disastrous' year. Almost 40 percent of the respondents estimated more than 10,000 fatalities would occur within 1 year, and more than 25 percent guessed there would be 100,000 fatalities. Many of the respondents expected such a disaster within their lifetimes, while the Sandia study suggests that there is only one U.S. reactor site-Salem, N.J.-at which an accident might cause as many as 100,000 “early” fatalities and estimates that these consequences have only a 10-6 or 0.000001 chance of occurring at that site. That analysis suggests that the average American (who does not live near that site) has an even lower probability of being killed within a year of a reactor accident. In general, it appears that public perceptions of the possible consequences of an accident correspond somewhat with the findings of the most recent technical studies, but that the probability of such consequences is greatly overestimated. Data from the Netherlands confirm the public's perception of nuclear power as uniquely hazardous. Over 700 adults of varying ages, living at varying distances from industrial activities were asked to judge the "riskiness" of a wide range of activities in 1978 and 1979. Nuclear power was judged to be more risky than most of the other activities and technologies, including drunk driving, transporting chlorine by freight train, and working as a big-city policeman (74). Several factors appear to enter into people's views of nuclear power as particularly risky. First, respondents in both the Netherlands and Oregon were concerned about the size of a potential accident and the lack of individual control in preventing an accident. In the Oregon study, nuclear risks also were seen as "unknown to the public and to science" and as particularly severe and dreaded. Both the Oregon study and opinion surveys show that about 40 percent of the American public believe that a nuclear plant can explode like an atomic bomb, even though such an explosion is physically impossible. Familiarity also played a role in people's judgments. In the Oregon study, nuclear risks were perceived as greater because they were unfamiliar. Another factor entering into risk perceptions was people's difficulty in assessing the probability of a reactor accident. People's opinions about nuclear power and other "hazardous" activities and technologies are not determined by perceptions of risk alone. The perceived benefits offered by a technology must be weighed against the perceived risk in determining how acceptable the technology is. Most activities, including development of nuclear power, are undertaken initially in order to achieve benefits, not avoid losses, and for many activities the expected benefits far outweigh the potential losses. In the case of nuclear power, perceptions of benefit may have played an important role in the trend of public opinion. During the 1950's and 1960's, when electricity demand was growing rapidly, the development of nuclear energy was promoted as a means to meet future demand and there was little apparent opposition to the technology. As electricity demand slowed in the 1970's and 1980's some people may have seen nuclear power as less vital to economic growth, so that concerns about risk became more promi nent in their assessments of the technology. Analysis of recent survey data indicates that judgments of "beneficiality" currently have a strong influence on Americans' acceptance of nuclear power. After safety, the second most important factor in support for nuclear power appears to be the belief that nuclear powerplants are necessary to reduce American reliance on foreign oil (57). In both the Netherlands and the United States, people living near to nuclear plants have been more receptive to the technology. However, while the Netherlands study appeared to indicate a resigned acceptance of the risks of nuclear power, some surveys in the United States indicate that those living nearby are more aware of the benefits. For example, a majority of people living near Portland General Electric's (PGE's) Trojan Nuclear Station continued to approve of the plant following the accident at TMI in 1979, while customers throughout the entire PGE service territory were ambivalent (7). The primary reason cited for local support of the plant was that it produced needed power. Similarly, residents of the town closest to Maine Yankee nuclear station, who benefit from the jobs and taxes provided by the plant, continued to support the reactor through two statewide referendum votes in 1980 and 1982 which would have shut the plant down. Defeat of the two referendum votes appears to be based primarily on the perception of Maine voters that Maine Yankee provides needed lowcost electricity (see Case Studies). Despite these favorable local attitudes toward some nuclear plants, opinion polls at other plants show that local support shifted to majority opposition following the accident at Three Mile Island (24). Analysis of survey data at one host community suggests that Federal safety standards are now seen as being too weak. It appears that national events which increase perceived risk can offset local perceptions of benefit. Psychological Factors The apparent gap between technical studies of nuclear power risks and people's perceptions of those risks has led some observers to suggest that there is little thought involved in the public's view of nuclear power. Instead, they argue, people react to nuclear energy on a purely emotional basis. For example, psychiatrist Robert DuPont argues that public concern about nuclear power is a "phobia" resulting from irrational psychological factors (17). Geographer Roger Kasperson cites frequently voiced concerns about genetic damage and cancer as evidence of the "emotional roots" of opposition to nuclear power, and psychiatrists Philip Pahner and Roger Lifton have suggested that fears of radiation from nuclear weapons have been "displaced" or "extended'' to nuclear power (30,36,55). Although there can be little doubt that emotional factors enter into the public's assessment of nuclear energy, further analysis of the public's view of risk indicates that the reasoning behind these opinions is more rational than first appears. First, while people may be inaccurate in their assessments of the probability of a catastrophic nuclear accident, they do not appear to overestimate the seriousness of such a catastrophe. Both proponents and opponents of the technology have an equally negative view of the deaths, illnesses and environmental damage that would result from a reactor accident (66). People who are concerned about nuclear safety do not view a radiation-induced death from a nuclear plant accident as significantly worse than a death from other causes, and they do not perceive genetic effects or other non-fatal consequences of such an accident as worse than death. The central area of disagreement between the experts and the concerned public lies in the area of greatest uncertainty even among the experts: the probability and impacts of a major accident (21). Secondly, lay people appear to rely on somewhat logical internal "rules of thumb" in assessing the magnitude of various risks. As shown in figures 40 and 41 people's assessments of the risks associated with various diseases and technologies correlate fairly well with statistical estimates of the risks. While the relative riskiness of the various activities was judged somewhat accurately, respondents in the Decision Research study tended to overestimate the risks of low-frequency events. According to the research team, this error results from people's assumption that an event is likely to recur in the future if past instances are easy Figure 40.-Relationship Between Judged Frequency and the Actual Number of Deaths 1,000,000 Estimated number of deaths per year NOTE: Respondents were told that about 50,000 people per year die from motor vehicle accidents. If judged and actual frequencies were equal, the data would fall on the straight line. The points and the curve fitted to them represent the averaged responses of a large number of lay people. While people were approximately accurate, their judgments were systematically distorted. To give an idea of the degree of agreement among subjects, vertical bars are drawn to depict the 25th and 75th percentile of individual judgment for botulism, diabetes, and all accidents. Fifty percent of all judgments fall between these limits. The range of responses for the other 37 causes of death was similar. SOURCE: Slovic, et al. (68). Reproduced by permission of P. Slovic. to recall. Moreover, the "availability" of an event in people's memories may be distorted by a recent disaster or vivid film. Because life is too short to actually experience all the hazards shown in figures 40 and 41, people tend to focus on dramatic and well-publicized risks and hence to overestimate their probability of occurrence (68). These findings help explain the increased public opposition to nuclear power reported in opinion polls over the past decade. Nuclear power's historic connections with the vivid, imaginable dangers of nuclear war lead people to associate the technology with catastrophe. Accidents such as the fire at Browns Ferry and the near-meltdown NOTE: Each point represents the average responses of the participants. The broken lines are the straight lines that best fit the points. If judged and technically estimated frequencies were equal, the data would fall on the solid line. The experts' risk judgments are seen to be more closely asociated with technical estimates of annual fatality rates than are the lay judgments. SOURCE: Slovic, et al., (68). Reproduced by permission of P. Slovic. In May 1979, 2 months following the Three Mile Island accident, there was a demonstration of about 200,000 people on the U.S. Capitol Grounds. Speakers urged the U.S. Congress to curtail further nuclear construction at TMI have added to the image of disaster in people's minds. The publicity surrounding these events, movies and books such as "The China Syndrome" and We Almost Lost Detroit, and the estimates of deaths in various safety studies have further enhanced the "availability" of nuclear power hazards in people's minds, creating a false "memory" of a disaster that has never occurred at a commercial nuclear reactor. Public education about nuclear safety systems, by identifying the various hazards those systems are designed to guard against, may only serve to increase the perceived risks of the technology. The Decision Research analysts also compared lay judgments of risks with the judgments of nationally known professionals in risk assessment (see fig. 41). The judgments of experts were much closer than those of the lay people to statistical ly calculated estimates of annual fatalities associated with various risky technologies and activities (21). However, while the experts knew more facts, their risk assessments also were found to be greatly influenced by personal judgment. Thus, expert studies also are subject to errors, including overconfidence in results and failure to consider the ways in which humans can affect technological systems. This latter problem was demonstrated clearly during the TMI accident, which was caused in part by human error. Because technical experts, like the general public, face limitations in evaluating the risks posed by nuclear power, it appears that there may be no single right approach to managing the technology. Instead, it is most appropriate to involve the public in order to bring more perspectives and knowledge to bear on the problem. There |