Via I Love Charts: Risk Perception and Actual Hazards


This seems very relevant to this crowd, though I’ll qualify my posting it here to show that I think it’s a poorly designed chart.

The chart attempts to show the disconnect between actual risk and perceived risk. One place it falls down is in conflating personal and societal risk. Cancer is an very large personal risk. Millions of people will be diagnosed and die due to cancer every year. But is it really four times the risk of an asteroid impact? On a personal level, no, because four people don’t die from cancer for every one that dies from asteroid impact. That said, on a societal level, the destruction wrought by an asteroid impact (a big one) is such that it may be comparable to the effects of cancer. And would you really argue that the threat of an asteroid impact is ten times that of airliner crashes?

Couching all of that, I appreciate what the chart is trying to accomplish. By pointing out the disconnect between what people are afraid of and what is actually threatening to them, it can help realign priorities. Focus should be shifted to prepare for and mitigate against that which can cause real harm (read: not asteroid impacts).

Risk perception disconnects are something that is well-known, and relatively well-studied. The CDC’s Crisis and Emergency Risk Communication course teaches us the following reasons why some things are feared beyond their possible threat:

>- Voluntary risks are more readily accepted than imposed risks.

>- Risks controlled by the individual or community are more readily accepted than risks outside the individual’s or community’s control.

>- Familiar risks are more readily accepted than unfamiliar risks.

>- Risks generated by nature are better tolerated than risks generated by man or institution.

>- Reversible risk is better tolerated than risk perceived to be irreversible.

>- Statistical risks for populations are better tolerated than risks represented by individuals.

>- Illnesses, injuries, and deaths spread over time at a predictable rate are better tolerated than illnesses, injuries, and deaths grouped by time and location.

>- Risks that do not single out a group, population, or individual are better tolerated than risks that are perceived to be targeted.

>- Risks generated by a trusted institution are better tolerated than risks generated by a mistrusted institution.

>- Risks that affected adults are better tolerated than risks that affect children.

>- Risks with a well-understood potential benefit and the reduction of well-understood harm are better tolerated than risks with little or no perceived benefit or reduction of harm.

Helpful tips from a not wholly helpful chart.