Natural disasters impact people and call for appropriate interventions. This year alone has brought the COVID-19 pandemic, raging wildfires across the western United States, and an unprecedented storm season in the Atlantic. How people respond depends largely on their perception of the risks to themselves, loved ones, and property. Scholarly research about risk perception can help decisionmakers better understand how individuals respond to catastrophic events and better implement policies and interventions that can reduce costs and improve outcomes.
Resources for the Future (RFF) and the Stanford Woods Institute for the Environment brought together two social scientists for a conversation about their latest research into risk and response to natural disasters. Their conversation provides insights for individuals, policymakers, and others who may want to know what kinds of actions will be most effective in the face of these environmental threats.
The researchers: Jon Krosnick—a Stanford professor, RFF university fellow, and expert in public opinion research—has recently released a national survey conducted in collaboration with RFF and the survey research firm ReconMR measuring American opinions about climate change and natural disasters. And Gabrielle Wong-Parodi, a Stanford Woods Institute Center fellow, is beginning to analyze data from her recent national survey, which examines the perceived risks of the dual threats of COVID-19 and wildfire.
An edited version of Krosnick and Wong-Parodi’s conversation appears below. They talk about risk perception and the need for strong leadership that provides clear, evidence-based recommendations to the public—a key consideration for a new Congress and the presidential administration in the coming years.
Risk Assessment Helps Motivate Response to Natural Disasters
Jon Krosnick: Gabrielle, I'm interested in hearing your thoughts on the perception of probabilities and risk, and the management of these perceptions.
For example, economists have been among the most vocal scholars to call for a greater understanding of the decisions that people make about how to manage their lives, and the role of their perceptions of probabilities in those decisions. How large of a role would you say their perceptions actually play?
Gabrielle Wong-Parodi: In terms of decisionmaking with respect to disasters, we know that perceptions matter, but they are not the only things that matter. For example, the kinds of decisions people make, and whether those decisions are an immediate reaction or a longer-term plan, all play into it. Risk perception is part of the equation, but a constellation of other things also are important.
For instance, social influence—what others around you are doing—also matters and influences your perceptions and behaviors. In a recent experiment, we found that not only does it matter what the people around you are doing in terms of preparing for hurricanes, but also what they are not doing, which sends a signal. As another example, one of the biggest predictors in preparing for hurricanes is whether or not an individual has prepared for a hurricane before. Plus, some individuals simply have a predisposition to be planners, so it's a complicated story.
People Must Respond to Compound Threats
JK: What led you to examine the COVID-19 crisis in tandem with the wildfire crisis in your latest study?
GWP: In an ongoing study I've been doing on the Gulf Coast, we’ve been following thousands of individuals in Texas and Florida since 2017 and have noticed that repeated exposures to hazards matter for people's decisionmaking, as well as their physical and mental health. Coping with repeated hazards also affects individuals’ ability to offer social support to others in their families and their communities.
By studying the risks of COVID-19 and wildfires together, I wanted to see how people are dealing with compound risks, and how that exposure to multiple risks may impact the decisions that people make. If these factors combine to influence people’s perception of risk and responses, then we hope to tease out why.
JK: What have you been seeing so far with your new study?
GWP: In a preliminary look at the data, some interesting findings have emerged. One of the most compelling—and perhaps troubling—results is that individuals who have been diagnosed with a respiratory illness, and therefore are at risk for exposure to wildfire smoke, are no more likely than anybody else to actually take precautionary measures such as wearing a mask during a smoke event. This might reflect a disconnect between the actual risk someone faces, and that person’s assessment of the risk they face, because generally speaking, people who think they are more at risk should have higher rates of wearing a mask.
JK: Evidence that perceived risk is not a primary driver of mask-wearing is quite interesting. Although that might seem irrational at first blush, perhaps it’s not so silly, since assessing risk is difficult, if not impossible, for most people without relevant expertise. Maybe people should simply wear masks regardless of perceived risk.
Perceived Risk Varies with the Context
JK: What other hypotheses do you hope to test with the survey data you’ve just collected?
GWP: One thing I mentioned earlier is social influence. We will examine response to social influence with regard to mask-wearing for both the pandemic and wildfires.
We would expect that individuals whose friends and family are likely to wear a mask will also adhere to mask-wearing. We also would anticipate that perceptions of the effectiveness of masks, for reducing the harm either from inhaling smoke or from being exposed to the virus, will be related to reported mask-wearing.
We also ask questions related to whether wildfires and the coronavirus were caused by or in part by human activities, and we will look at how those answers affect their perceptions of future risk.
JK: If survey respondents attribute COVID or the wildfires to human activity, what might change in their perception of those risks, versus if they believe these things simply have natural causes?
GWP: Interestingly, some recent research suggests that people who attribute global warming to humans tend to see global warming as more of a problem. Those individuals have higher perceived risk, and some initial evidence suggests that they may be more likely to take action to mitigate climate change or practice more pro-environmental behaviors.
People tend to think that if something is caused by human activities, then the consequences are more dire and human suffering will be greater; recent studies have suggested this. It will be interesting to see if people also make that link with wildfire smoke or COVID-19.
Individuals who have been diagnosed with a respiratory illness, and therefore are at risk for exposure to wildfire smoke, are no more likely than anybody else to actually take precautionary measures such as wearing a mask during a smoke event.Gabrielle Wong-Parodi
JK: Are you aware of any research that examines how people respond to imagined situations versus situations they have experienced personally?
GWP: Yes, we do know that imagined versus experienced situations matter. People’s unique experience of a situation matters, as well.
When I was doing interviews with individuals who were impacted by Superstorm Sandy in New Jersey, I spoke with a woman who had lost everything. It struck me when she said there was no way she could have imagined the unimaginable before the storm—she’d had no idea how bad it would be. She thought she did know, but she just did not and could not have. She told me that if there was some way to help people really imagine the worst, it would be so valuable.
If an individual has lived through a hurricane and nothing bad happened, their perception of the risk of the event decreases over time. This makes logical sense: if nothing bad happened to you, then you think the next time won’t be so bad, either, whereas individuals who do suffer severe loss or damage have higher perceived risk. In the case of COVID-19, it is not hypothetical—it is a real situation for everyone.
Right now, we’re pulling resources together to run a study that asks similar questions in California among people who are actually being affected by wildfires, because getting those insights would be important.
Political Party Affiliation Can Influence Disaster Preparedness
JK: Do you think wildfire preparedness is an arena that might be less politicized than others?
GWP: We do tend to see differences with respect to disaster preparedness across a variety of different hazards and people’s self-reported political party identification. With a quick glance at the data, it seems there is a difference between mask-wearing in response to wildfire smoke between people who identify as Democrats or Republicans—with Democrats more likely to wear one.
JK: Your survey interviewing was happening when mask-wearing had a political tinge to it. Do you think the politicization of mask-wearing regarding COVID-19 might have influenced mask-wearing for wildfires?
GWP: It would have been great if we could have asked the question about mask-wearing and wildfires last year. But yes, it seems like mask-wearing for wildfires has been changed by the conversation we've been having around COVID-19 and masks.
JK: Do people who identify as Republicans hold different views of disaster preparedness than do people who identify as Democrats?
GWP: Looking at some of my work on hurricane preparedness, I’ve found that preparedness does not always fall along party lines. In some of our studies, Democrats are more likely to see the federal government as responsible for helping to rebuild after hurricane events, and that the federal government should be responsible for helping people to prepare. Whereas people who identify as conservative are less likely to believe that.
We haven't seen real differences in terms of the level of preparedness across party lines—we’ve seen differences more along the lines of a sense of responsibility.
Empirical Data Should Inform Leadership and Public Policy
JK: With the wildfires, many people living near one another have been trying to protect themselves at the same time. This demand has caused a six-week waiting period for the delivery of masks that people purchased, for example. Have you observed—whether in the literature or from your own research and thinking—that people feel completely helpless when they are not able to access the resources they know could help them, or simply from being paralyzed by anxiety?
GWP: We have seen in many instances that people need to feel proactive and as though they are doing something to help themselves.
For example, individuals who were impacted by the Camp Fire in Paradise, California, were doing things like dipping rags into water and wearing them on their faces to protect against wildfire smoke. These were individuals who had to work outside, who didn't have the proper equipment and didn't know what to do in the absence of proper equipment—so they improvised. Doing something was a sort of coping mechanism and helped them feel some level of control.
GWP: This example indicates a need for strong leadership that issues clear messages, such as information about where to go and what to do in a disaster, or what kinds of alternatives will work when optimal resources (such as masks) are not available. The information given needs to be evidence based—not just any information or any consistent message, but evidence-based, scientifically backed messages providing people with options that are behaviorally realistic and resource sensitive for people.
Working with key policy constituents from the start of a research project to understand their needs is one way I try to ensure that my research improves policy. In our work, we often reach out or are in conversation with practitioners and policymakers as we are conducting our research, so that we can present that data in a way that can be most helpful and translate it into actionable knowledge.
JK: If people make decisions that put themselves in harm's way—such as by choosing to live in a fire-prone area, or to live far away from a fire station, or to build a house made of wood—and those individuals then become victims of a fire, our response as a country is not to just say, “Bad luck; your decision; live with the consequences.”
Insurance companies have to pay for a lost home, and the Federal Emergency Management Agency steps in. As a society, we value helping people who are hurt in those situations.
We’ve seen in our recent survey that large fractions of Americans—70 to 80 percent of people in a country that is now heavily divided—favor government action to minimize the risk of a fire. They support such things as hiring people to go into forests and clear burnable matter. They also support the government facilitating people getting insurance and paying for people who lose their houses. This means that the government—and therefore every taxpayer—has an interest in any single individual not underestimating their risk. We should all want individuals to take steps to accurately perceive that risk and minimize the risk, so we don't end up paying for it later.
The challenge for us as social scientists is to figure out how to help people calibrate their risk perceptions properly, to take appropriate preventive steps to minimize their risk, but not become so paralyzed by anxiety from risk that they can’t live a normal life. I think it's quite legitimate for public policy professionals to ask us what they should do, but we don't necessarily have the answers yet. We need more research to move us as a community of social scientists to a place where we have better advice for policymakers.