Lay perceptions of earthquake hazard information

Dave Krantz, Katherine Thompson

People encounter probabilities frequently in their everyday lives, but the vast majority of them ignore probabilities or do not understand them fully, a tendency which often leads them to make decisions based on faulty or incomplete information.  Often these decisions are not vital, e.g., choosing whether to bring an umbrella based on a 30% probability of rain.  But some probabilistic events, like earthquakes, carry with them very serious possible consequences. It therefore is important that the scientists who study such significant, probabilistic natural hazards have experience with or knowledge about earthquakes; whether or not they are generally more risk-seeking; and whether or not they are naturally adept at understanding numbers.  We will use these results to help inform the scientists who create earthquake hazard maps (and other public-information campaigns) about the best ways to clearly communicate information to the public and better understand the way their reports are used by the general public.  The current format of the earthquake hazard maps that are available to Californians relies heavily on probabilities, and we therefore hypothesize that the general public does not understand these maps—and therefore their own personal risk—in the same way as the scientists who created them.  This study intends to measure the differences between layperson and expert interpretations of California earthquake hazard maps, and test whether some alternative presentations of hazard information might communicate earthquake risk to the public more clearly and with less room for misinterpretation.

It also explores alternate ways of presenting hazard data, to determine which presentation format most effectively translates information from scientists to public. Participants both from California and from elsewhere in the United States are included, to determine whether familiarity — either with the experience of an earthquake, or with the geography of the forecast area — affects people’s ability to interpret an earthquake hazards map. We hope that the comparisons between the interpretations by scientific experts and by different groups of laypeople will both enhance theoretical understanding of factors that affect information transmission and assist bodies such as the WGCEP in their laudable attempts to help people prepare themselves and their communities for possible natural hazards.

 Major Findings

• Results from first responders and emergency managers replicated previous findings: It appears that Prospect Theory’s overweighting of small probabilities and underweighting of large probabilities occurs even in the domain of earthquake probabilities—subjects rated the chance of an earthquake “this week” as much higher than it is in reality (<1%), and significantly underweighted the chance of an earthquake over the next 100 years (>99%), for example.
• Lay participants showed systematic errors when interpreting either type of earthquake hazard map, in both cases underestimating the probability of an earthquake in a given region. Experts slightly (though not significantly) underestimated that same probability for the newer map, but correctly identified the probability for the older map.
• It appears that the amount and magnitude of lay participants’ error is partially moderated by their naïve estimates of earthquake probability, and their familiarity with prior earthquakes in California.

Broader Impacts

• Scientists at the US Geological Survey have considered this work when deciding how to create new earthquake hazard maps. These preliminary results have been presented in meetings of USGS seismologists and communication officers, and the ensuring conversation has led to increased collaboration between CRED and the USGS.

CRED2 Award (2010-2015): Funding was provided under the cooperative agreement NSF SES-0951516 awarded to the Center for Research on Environmental Decisions.