EXPERTS AND THE PUBLIC

Part two

EXPERTS AND THE PUBLIC
Although claims that expert judgment is more veridical than the public’s are not examined here, the majority of the research indicates the public determine risks differently from experts (for a minority view, see Rowe & Wright, 2001). In addition, Sjöberg (1999b) claimed people are not that misinformed about all risks and he cites some studies showing convergence of expert and public opinion (Wyler, Masuda & Holmes 1968) which was specific to illnesses. Nonetheless, these findings are in the minority.

The vast majority of studies document differences between experts and the public in risk perception (Slovic, Fischhoff, & Lichtenstein 1980, Slovic 1987, Kraus, et al 1993, & Slovic et al 1995). For example, we know the public ranks some risks higher, such as chemical products (Kraus, Malmfors, & Slovic 1992 and Slovic et al 1995), radioactive waste disposal (Kletz 1996), and spray cans (Slovic 1987). On the other hand, the public ranks some risks lower than experts, such as X-rays (Slovic, Fischhoff, & Lichtenstein 1979 & Slovic 1987), downhill skiing (Savadori, Rumiati, & Bonini 1998), and bicycles (Slovic 1987).

While Drottz-Sjöberg and Sjöberg (1991) argue differences may exist before scientists receive their education (scientists self-select themselves out of the public), they admit socialization of values, conformity pressures, and familiarity may still be at work.

In general, experts seem to pay more attention to probability while the public is concerned about consequences. Overall, public risk judgments are less closely related to fatalities than those made by experts. Hyperbolically, Renn went as far as claiming probability plays hardly any role at all (2004).

Of course, hazard experts disagree among themselves as well. Kahan, Slovic, Braman & Gastil 2006, argue “…disagreements among risk experts are distributed in patterns that cannot plausibly be linked either to access to information or capacity to understand it” (p, 1093). They claim cultural worldviews, such as political ideology and institutional affiliation may account bias expert judgment as well (Slovic 1995, p. 662). As such, it has been argued that experts may screen arguments to protect their existing beliefs.

The primary rationale seems to be that experts rationalize hazards against dosage and exposure. The public does not. For example, “[t]he public would have more of an all or none view of toxicity… [T]hey appear to equate even small exposures to toxic or carcinogenic chemical with almost certain harm” (Kraus, Malmfors & Slovic 1992, pp. 217 & 228) despite well-documented hormesis effects to some chemicals. MacGregor, Slovic and Malmfors reported “…people reserve the term exposure for substantial contact or contact sufficient to cause cancer” (p. 653).

When Weinstein (1987) studied public sensitivity to chemicals in food the statement: When some chemical is discovered in food, I don’t want to hear statistics, I just want to know if it’s dangerous or not elicited strong agreement from 62% and moderate agreement from 21.6% of the respondents. While the contagion effect between product lines might not be defensible (Berube unpublished manuscript), there is clearly a contagion or cascade effect in terms of food. “[E]ven a minute amount of a toxic substance in one’s food will be seen as imparting toxicity to the food; any amount of carcinogenic substance will impart carcinogenicity, etc.” (Kraus, Malmfors & Slovic, 1992, p. 229). As well put elsewhere, when a young child drops a lollipop on the floor, the brief contact with dirt causes the parent to throw it away rather than washing it off and returning it to the child. Evidence like this lead MacGregor, Slovic and Malmfors to conclude “…somewhat subtle changes in how the concept of exposure is conceptualized and communicated evokes very different inferences about its meaning” (1999, p. 652) and offers the risk communicator opportunities to ponder.

Additional sources for disagreement between experts include the open ended nature of scientific claims. Science is almost never definitive. In addition, knowledge building often involving legalistic and technocratic debates over findings and this may be disadvantageous to public groups by increasing confusion, engendering panic, etc. Altogether, this is interpreted by the public as disagreement which increases uncertainty (Kajanne & Pirttilä-Backman 1999) and uncertainty impacts trust (see below).

There is much disagreement on if, why, and how experts and the public use different tools to perceive risk as well. For example, “the assertion that experts’ risk perception is driven by objective data and risk assessments and somehow more simple than that of the public is based on a small sample of experts studies by Slovic and colleagues (See Slovic, Fischhoff, & Lichtenstein 1979) in the end of the 70s.” Sjöberg added: “…the frequent assertion of simplistic structure in experts’ risk perception is an urban myth” (1999a, p. 8). The myth is due to the psychometric model which has come under attack (Sjöberg, L. (2000a). More research seems to be in order.

To add another level of complexity, “…merely mentioning the possible adverse consequences (no matter how rare) of some product or activity could enhance their perceived likelihood and make them appear more frightening” (Slovic 1986, p. 405) such as what occurred with high voltage lines and cellular telephones. Consequently, “many risk communications about chemical exposure may lead more often to confusion or heightened concerns, when it is actually intended to reduce concerns” (MacGregor, Slovic & Malmfors, p. 654). This becomes increasingly problematical as experts respond with more and improved risk assessment studies. Slovic (1986) warned merely mentioning possible adverse consequences could make them appear more frightening and even warned “…risk-assessment studies tend to increase perceived risk” (1993, p. 680) suggesting great care and this phenomenon needs to be taken in studying public opinions and attitudes and in designing risk messages for the public.

Our policy makers are not better equipped to determine public perception of risk at this time. “[W]hen politicians were asked to estimated that they believe was the public’s risk perception, they made gross errors” (Sjöberg 1999a, p.5). Oddly, this may be due to the input they got from active and concerned citizens rather than the public. Milbrath explained that those who were active and take part in the process, who are able and willing to give of their time and energy, are quite unrepresentative of the public at large (1981, p. 480). Unless great care is taken in risk perception research, it can be counterproductive when it makes a public view more salient, increasing its influence when the view is an unrepresentative generalization.

There is the additional problem associated with framing (Scheufele & Lewenstein, 2005). Framing refers to the idea that the way information is presented rather than the content itself can have an important impact on how audiences perceive the information. Modes of presentation can differ in terms of terminological choices, visual cues, or other factors (Scheufele, 1999). The "Frankenfood" label used during the GMO debate is good example of a frame that may directly impact risk perceptions among a public that does follow scientific rules of decision making.

Outside of nuclear energy, few studies have been undertaken dealing with a phenomenon like nanotechnology. Moreover, earlier risk studies suffered from a radiophobic bias (fear of things nuclear including bombs). Some recent studies in biotechnology and use of chemicals associated with food offer better guidance. For example, in 2004, Salvadori et al studied expert against public perceptions on biotechnology in Italy. In general, they found the experts significantly and systematically perceived less risk than the public. In addition, they noted higher perceived risks when the biotechnology involved food-related rather than medicine related applications. Most interesting was their conclusion that expert and non-expert differences may be affected by the nature of the hazard. They added “…public perception of risk…could be reduced by providing information about benefits….” Unfortunately, they also observed some perceptions, including those of experts, “…could be increased by providing information on harmful effects and negative consequences” (p. 1298) suggesting a complex dynamic is at work.

Slovic, Fischhoff and Lichtenstein supposed: “Attempts to characterize or compare risk, set safety standards, and make risk decisions will founder in conflict if policy makers insist, as they often have, on the narrow definition of risk as a conditional probability of dying” (1985, pp. 92-93). We need to understand better the processes involved rather than to institute experiments which may not be grounded in relevant research findings. We need to develop more refined techniques for representing uncertainty and data sets associated with traditional risk assessment models. “..[R]isk assessment [may have] been oversold because of the need to rationalize decisions about chemicals” (Neil, Malmfors & Slovic 1994, p. 201).

Finally, the window of opportunity remaining for risk communicators to engage the public is closing quickly as nanotechnology products are marched out onto the market. We know “…risk and benefit judgments of the hazards were found to be more strongly negatively correlated under time pressure” (Finucane 2001) and as such, it may behoove us to provide opportunities and methodologies to facilitate public engagement, sensibly and soon.