Concern Level Versus Magnitude of Risks: A Consumers Union Survey. Cynthia Langlois and Edward Groth III, Consumers Union of the US, Inc., 101 Truman Avenue, Yonkers, NY 10703
Consumers Union, the publisher of Consumer Reports, conducted an in-house risk perception survey in the spring of 1995. About half the staff (n = 165), from all parts of the company, were randomly selected to participate. Their responses were anonymous. For 44 different risks, participants were asked to: indicate how concerned on a scale of one to five they were for themselves and their family; indicate how concerned they thought the American public as a whole should be; and estimate how many people die from that risk in one year. These perception data were compared to actual annual fatality statistics, where they were available. This comparison shows that the top ten risks by actual fatality were the same as the top ten risks by participants’ concern level, with one exception. We then broke both the concern data and the actual fatality data into three categories (high, medium and low) and compared them to each other. For the most part, participants were highly concerned about high fatality risks, moderately concerned about moderate fatality risks and slightly concerned about low fatality risks. In fact, when outliers are removed there is a correlation between concern level and actual fatalities. This comparison also shows that common "outrage factors" (i.e. uncontrollability, involuntariness, etc.) did not seem to play a major role when the participants were highly concerned, and when fatalities were high. When fatalities were lower, and perhaps less clear, outrage factors seemed to play more of a role in determining concern level, but still did not elevate any low-fatality risks into the high concern category. Although participants’ estimates of annual fatalities were very inaccurate, most individuals could reliably distinguish high fatality risks from low fatality ones. Clearly, although they do not know the number of annual fatalities, those surveyed are neither highly concerned about risks that cause few deaths, nor ambivalent about risks that cause many deaths.