The Way We Think About Risk is Risky

Photo: Comstock

From the Soapbox Science blog on, here’s an interesting piece by risk consultant David Ropiek on the ways in which we perceive and react to risk. His basic thesis is that our interpretation of risk is almost always subjective rather than fact-based, which gets us into trouble:

We worry about some things more than the evidence warrants (vaccines, nuclear radiation, genetically modified food), and less about some threats than the evidence warns (climate change, obesity, using our mobiles when we drive). That produces what I have labeled the Perception Gap, the gap between our fears and the facts, which is a huge risk in and of itself.

The Perception Gap produces dangerous personal choices that hurt us and those around us (declining vaccination rates are fueling the resurgence of nearly eradicated diseases). It causes the profound health harms of chronic stress (for those who worry more than necessary). And it produces social policies that protect us more from what we’re afraid of than from what in fact threatens us the most (we spend more to protect ourselves from terrorism than heart disease)…which in effect raises our overall risk.

Mike B

Yawn, Bruce Schneier has been saying this for years.

Tom Maguire

Segment on Risk Telescoping. Part of a longer Science Channel program called Understanding Probability and Odds:

Hugh Culver

The perception gap explains so much of what drives the somewhat bizarre behavior that we see all around us - we worry more about how clean our car is than what it burns in fuel. I guess my question is: now what? How could we harness the power of the marketing geniuses on Maddison Ave. to redirect our attention to what really matters?


In many cases, such as vaccination, nuclear radiation, and climate change, it's not that the risk assessment is subjective. It is in fact as fact-based as any other risk assessment. It's just that interested parties have invested great amounts of time & effort into creating a set of false "facts", or disinformation, which some people accept as real and use to make their assessment. For instance, if the health claims made by the anti-nuke lobby were anywhere close to being true, opposition to nuclear power would be perfectly reasonable.

It's the old "garbage in, garbage out" principle at work. The problem isn't subjectivity, it's how to replace a world view formed from disinformation with one closer to reality.

Andrew Roberts

Perception of risk is subjective, it is because of the misinformation, fear mongering and false facts that lead to concern/fear/anxiety and worry. It is these feelings of "outrage" that lead us to look at the hazard in a completely different way to the "experts" view. Take a list of actual hazards to life, rank them in order. Then rank the hazards that upset people and compare the lists. What is the correlation? Well, in Wilson's work at Harvard it was 0.2. The hazard or risk that will kill you are not the same that upset people, when people are upset they think entirely differently about risk. The misinformation out there is the seed of outrage, it is a deliberate tactic of tech opponents, NGOs, activists to build fear/worry/anxiety so that people get upset and worry about things that are in fact, in real average mortaility/morbidity terms, low hazard. Same tactic can be applied by risk communicators trying to get people to take action over things that are in fact truly hazardous but of little concer; vaccinations, AIDS, smoking etc. This is precautionary advocacy used in a positive way, NGOs et al use precautionary advocacy to invoke fear about GM, nuclear power, pesticides etc. Monsanto, TepCo, Bayer will use outrage management in order to calm people, make them assess the hazard in an objective, calm way.



The BBC are conducting an experiment on risk that includes sections on our ability to estimate the riskiness of a given activity (I clearly suffer from the perception gap!) as well as sections on more conventional stuff, like choices between lotteries. Very interesting and well worth a look!

Andrew Roberts

The "perception gap" as you say is in fact one of Dr Vince Covello's four theories of Risk Communication, namely Risk Perception, and what underpins Risk Perception Theory is the fact that when people are stressed, their perceptions and decisions are influenced by a wide range of factors, technical facts often being the least important (worth less than 5%). In general, it is the fact that people are upset that makes them think something is dangerous. People can be upset for a variety of reasons, but generally it resolves around key risk perception factors: trust, benefit, control, fairness. Essentially these factors skew the perception of hazard, making some low hazard technologies (like GM crops) be perceived with dread, whilst genetically modified insulin for example can be accepted with relatively little concern (in this example the balance of benefits to risk is judged very much to be in favour of the user (diabetic) and hence GM insulin in now universally in use — contrast that with GM crops!). Your article is well received, risk communication is both fascinating and long overdue for further popular exploration.



If I hear one more climate change reference I am going to scream. Look at the facts; not the hype. I thought that's one of the tenants of Freakonomics. . .


In fairness, the the guy's a Harvard professor writing for; he probably has a contractual obligation (and certainly a social one) to at least feign belief in AGW. In any case, his broader point is valid.

"we spend more to protect ourselves from terrorism than heart disease"

This may descend into an argument over what is meant by "protect", but is this statement true?


It's also a "have you quit beating your wife?"-type question, since most of the effective protections against heart disease have negative costs (that is, you save money by doing them). Quit smoking? That's what, a couple of bucks per pack? Bike to work instead of driving the $50K SUV & filling its tank with $4/gal gas? Eat simple, healthy (and usually cheap) foods instead of expensive fat & salt-laden prepared ones? Even the daily aspirin is what, a couple of bucks per hundred?


The key to the article is in the last paragraph:

"That risk is inescapably subjective is disconcerting for those who place their faith in the ultimate power of Pure Cartesian...The challenge is to rationally let go of our irrational belief in the mythical God of Perfect Reason, and use what we know about the psychology of risk perception to more rationally manage the risks that arise when our subjective risk perception system gets things dangerously wrong."

I've dealt with so many scientists and engineers who are disciples of pure reason and who refuse to acknowledge that subjective risk perception is even valid, never mind being able to deal with it. Refusing to admit that the perception gap exists doesn't make it go away.


It is clear from the two sets comparisons in chapter 5, between a the risk of death in an airplane vs. car, and a child's risk of death in a pool vs. gun accident, that the authors have used very different strategies to analyze and present risk, in order to come to the bold and shocking conclusion that swimming pools are more dangerous than guns.
The statistics of the airplane vs. car comparison were based on a equation calculating the risk "per hour", and the pool vs. gun accident comparison were based on statistics of the total number of children who die in each type of accident. If the authors had used the same equation to compare the two different sets of risks, I think the results would have been different. Clearly, if we're going by the number of total deaths, in the car vs. plane comparison, like the authors did in the pool vs. gun comparison, the result would show that many more people in total die in car accidents than in plane accidents.
Can you think of a way to measure the findings of these two sets of risk research so that they really are comparable, taking into account the fact that nearly all American children have access to swimming in a pool at some point in life, and from the statistics I found, only about 35% of American children live in a home with a gun, and only about 30% of that 35% live in a home where the gun is stored loaded, and unlocked.
It seems that the research is presented in the way it is for the sole purpose of shocking readers.