Judging Risks

With the disaster in Japan, people are wary of the dangers of nuclear power. But is public attitude for energy safety reflecting actual risks? The PEW Research Center released a poll last Monday finding a record low level of support for increased use of nuclear power.

Opinion about expanding the use of nuclear power has fluctuated in recent years. However, the current measure matches a previous low in support for increased nuclear power recorded in September 2005 (39% favor, 53% oppose).

The latest national survey by the Pew Research Center for the People & the Press, conducted March 17-20 among 1,004 adults, finds little recent change in opinions about other energy policies – with one notable exception. With the recent surge in gas prices, support for increased offshore oil and gas drilling continues to rebound.

If people want to reduce energy use, great. If they think shifting energy production to other sources like coal and oil is safer, that’s a problem. Via Andrew Sullivan, I came across this excellent illustration:

Per unit of energy produced, nuclear power is safer than oil or coal by an order of magnitude. For “every person killed by nuclear power generation, 4,000 die due to coal, adjusted for the same amount of power produced.” Of course, there are differences from country to country and you can look at the study itself to find out more relevant details. But even so, the truth isn’t close to reflected in public opinion. The chart’s creator, Seth Godin, tries to figure out why:

Vivid is not the same as true. It’s far easier to amplify sudden and horrible outcomes than it is to talk about the slow, grinding reality of day to day strife. That’s just human nature.

I think that any time reality doesn’t match your expectations, it means that marketing was involved. Perhaps it was advertising, or perhaps deliberate story telling by an industry. Or perhaps it was just the stories we tell one another in our daily lives. It’s sort of amazing, even to me, how much marketing colors the way we see the world–our reaction (either way) to this chart is proof of it.

I’d agree that marketing and presentation plays a huge role in public perception. I am, after all, a communications director for my job.

But I want to stress the importance of human cognitive biases. As Godin points out, vivid visual disasters will have a disproportionate impact on people. But I think immediacy is one of the most important factors. When there’s an oil spill or a nuclear meltdown, people see the direct dangers of the power source. But much of the danger of coal comes from its emissions and longer-term damage. It’s more difficult for us to make the connection between a cause and effect when there’s more time between them.

But then, that’s why we do science and collect data. The challenge is getting us to internalize the scientific findings rather than just their inaccurate generalizations. Charts like these can help.

11 Responses to Judging Risks

  1. Cory Albrecht says:

    If you search on PubMed for the health effects of living near to fossil fuel-burning electrical plants and nuclear reactors you’ll find that the data for nuclear plants is ambiguous at best, while living near coal-fire plants is obviously detrimental.

    The one study that everyone always pointed to for health effects of leaving near a nuclear reactor was shown to have faulty data. It turned out that postal addresses were used instead of actual location of residence, and once that was corrected for, with the original subjects, the supposedly cancer rates faded into the background noise.

    But, as you imply, just try and point that out to people.

  2. “But I want to stress the importance of human cognitive biases.”

    As you should. Humans are not psychologically adept at risk assessment. Factors that might play into the coal vs. nuclear debate:

    – Natural disasters (such as mine collapses) are less emotionally evocative than those that are caused by human action;

    – Familiar technologies are perceived as less risky than unfamiliar ones;

    – Chronic risks (like those faced by miners) are less emotionally evocative than catastrophic risks (such as that of a nuclear meltdown).

    Communications people and experts can also skew the assessment in their favor by appearing trustworthy and responsive to the concerns of the public. (Who wouldn’t trust a face like Jesse’s?)

  3. David Levin says:

    While I don’t know enough to say for sure, I would guess that your death rates are misleading. Nuclear accidents are extremely rare events; estimating accurately both the small probabilities of catastrophes and the large costs (economic and human) is near impossible. Indeed, in the roughly 50 years of nuclear power, how many *serious* accidents can we count? (I think it is perhaps 3-4, but I don’t know for sure.) And we haven’t seen the worst possible accident, although I suppose Chernobyl comes close. There just isn’t enough information to accurately estimate the probabilities, and hence calculate expected values. So the true dangers of nuclear power are not easily quantified.

  4. JB says:

    Nuclear vs. Coal is akin to Plane Crash vs. Car Crash. It’s called representative bias… a common human error in judgement.

    The fact is nuclear power like plane rides are very much safe. However, because of the simple rarity of a traumatic event happening with nuclear power media and news stir up a frenzy with massive media coverage and cause hysterical panic among the common people. Papers want to sell, news shows want ratings… They don’t care about the 4,000 accidents caused by coal, it’s old news for them. They spend the majority of their years reporting on that. When something like this happens naturally the buzz is all about how unsafe nuclear power is. Just like when a plane crashes all of a sudden travel by plane is unsafe yet we get in our cars every day and there are big accidents every day from driving… this leads to the average person thinking it’s unsafe and a good opportunity for antinuclear protestors to push their agendas in a few people’s faces who don’t think critically about these things.

    Think about it people before you jump on the mass hysteria bandwagon.

    • JB says:

      Sorry I should have said “availability bias” in my last post not ‘representative’ forgive me. I always mix the two up.

      • Jesse Galef says:

        I think the comparison is a good one – we hear more per fatal plane crash than car crash (adjusted for mile traveled, etc). Wouldn’t it be great if in every report about plane crashes the news showed a chart like the one above, and showed the tiny ‘air travel risk’ increase ever-so-slightly compared to the automobile risk?

        At least with car accidents we can easily connect the driving to the death. With coal, it’s a long-term killer that affects a range of people whose connection to the plant (emissions through the air) is invisible to the naked eye.

      • Max says:

        Flying is safer per passenger mile, but driving is safer per passenger hour. That might explain why pilots have a higher rate of job-related deaths than professional drivers.

        Also, look up black swans and fat-tailed distributions. People tend to UNDERESTIMATE the risk of very rare but very consequential events, like the Japanese tsunami.

  5. J Myers says:

    Per unit of energy produced, nuclear power is safer than oil or coal by an order of magnitude. For “every person killed by nuclear power generation, 4,000 die due to coal, adjusted for the same amount of power produced.”

    If your graphic is to scale (and it’s not much use if it isn’t), it would seem that nuclear power is safer than coal by nearly three orders of magnitude (if 4,000x more die from coal than from nuclear power, and oil deaths appear to be roughly 1/4 of coal deaths, then oil deaths would be approximately 1,000x greater than nuclear deaths).

  6. Cory Albrecht says:

    When it comes to the danger/risks of different types of energy sources, I thought people might find this interesting.


    “This visualization compares the energy mix and number deaths related to each of the main sources of energy worldwide – coal, oil, natural gas, nuclear, hydro and biomass.”

  7. Jesper says:

    On the news here today was that Germany is going to completely stop using nuclear energy. In 2022 all of their nuclear power stations will be shut down. I haven’t followed the news in Germany closely, but since the disaster in Japan some unrest has arisen around nuclear energy, there were some protests and the government agrees with the people, so that’s what now lead to this radical decision.

    If you look at the data and these graphics, then based on the facts you have to conclude that nuclear energy is very safe. However, when something does go wrong with a nuclear power plant, it’s likely that it’s going to be a big disaster. It’s the same thing as flying in an airplane: there are a lot of people who are afraid of flying, even though it’s much safer per unit of distance travelled than driving in a car – and the people who are afraid of flying don’t even think twice about getting into a car.

    • Max says:

      The analysis was done by nextbigfuture.com, which is some kind of pro-industry, anti-regulation, Global Warming-denying blog. I don’t trust them.

      When you’re flying in an airliner, your chance of a fatal crash is the same whether there are 20 or 200 passengers on board, so it’s not really affected by the size of the disaster.

Leave a Reply to Cory AlbrechtCancel reply