Social amplification of risk--Burns et al 1993

Incorporating Structural Models into Research on the Social Amplification of Risk: Implications for Theory Construction and Decision Making

This study empirically tests the framework of social amplification of risk, which argues that "social and economic impacts of an adverse event are determined not only by the direct biological and physical consequences of the event, but by the interaction of powerful psychological, cultural, social, and institutional processes that amplify or attenuate public response to the event" (Burns et al., p612).

In this study, the authors examined the causal relationship between "risk signals," "media coverage," "public response," and some physical consequences of adverse events.

108 hazards were identified and evaluated by experts to determine their political and economic implications. Public responses were measured by the level of political involvement, worry, and risk-reducing action induced by the hazards. Although measured at the individual level, these public responses variables were then aggregated to represent macro-level preferences. Information about media coverage was also collected to reflect the number of stories, duration of coverage, and issue half-life.

The most important finding of this study rests with the fact that both "casualties" and "property damage," indicators of the consequence of hazards, showed only marginally significant effect on the societal impact of the issues. It is, instead, the media coverage of and public reactions to the issues that matter.

However, I have several questions after reading this article.
1. What is the subject of amplification? Is it the social and political impact of the issue or the perceived risk of the issue? For example, some people may not care as much about what the risks are going to do to other people, or the whole society, as to themselves. Employing an "objective" measure of the impact of hazards does not hit the target idea. It is also difficult for me to draw a linkage between public reaction measures with the "impact" variable formed by the opinions of the experts.

2. Based on my first question, the causal order of their model (with "societal impact" being the final dependent variable ) may be open for adjustment. Can public worry and willingness to participate politically serve as the final outcome and be predicted by the "perceived" societal impact of hazards?

3. Showing a strong relationship between, for example, media coverage and societal impact does not necessarily mean that these hazards were "amplified." To what baseline does this amplification or attenuation compare? I think this is the most critical issue that scholars in this line of research should address.