ABSTRACT
Close to three decades ago, American Psychologist published in its September 1996 issue three articles on risk communication. One of the three, by Monahan and Steadman (1996) on “Violent Storms and Violent People,” introduced, by way of analogy, weather forecasting to the center-stage of discourse on behavioral risk assessment. In their oft-cited paper, Monahan and Steadman began by querying why weather forecasters do such a good job. They cited the report of the National Research Council’s Committee on Risk Perception and Communication (Improving Risk Communication, 1989) as identifying five chief reasons for the success of weather forecasters: (1) frequent practice, (2) base rate information, (3) actuarial support, (4) availability of feedback, and (5) educational programs/training programs. Although the theme of their article is how behavioral risk prediction can benefit from lessons learned from meteorologists, the substance of the article explored risk communication. Monahan and Steadman note that the National Weather Service (NWS) went through three iterations, from deterministic predictions prior to 1965 (“It will/will not rain today”) to probabilistic pronouncements (“There is a 40% chance of rain today”) to categorical messages, reserved primarily for severe weather (e.g., tornado or hurricane “watches” and “warnings”). The NWS apparently still uses probabilistic statements when it comes to common weather events and categorical messages only for severe events. Why the NWS shifts to categorical messages for severe (rare) events lies in the NWS’s “skepticism about the competence of users—particularly the general public—to optimally process information about low probability events (Hughes, 1980)” (Monahan & Steadman, 1996, p. 934). A more generalized problem of innumeracy (a relative incapacity or difficulty with numbers) was described more recently by Scurich, Monahan, and John (2012). Scurich et al. (2012) noted, “almost half the population has trouble with mundane, simple numeric tasks, such as estimating a price per ounce at a grocery store” (p. 549).
