Chapter 5Risk to Individuals:Perception and RealityNote. The lecture is mostly based on the sources stated at the end of the firstsection – I don’t have anything original to say.In everyday language risk means “s ome th i ng undesirable that may hap-pen in the fut u re ” and thi s lecture deals with risks to us as individuals. Oneprominent context concerns the risk of dying from different causes. Pedant i -cally, dying is a certainty rather than a possibility, but of course one meansdying prematurely. Aside from the fact that we’re all aware of such risks,this context has particular features, not necessari l y present for other types ofrisk. First, it’s a yes/no event. Second, we have very good data on ages and(proximat e) causes of death. Third, there is a large industry (life insurance)associated with this particular risk. Fourth, news media, television in par-ticular, devote a lot of time to risks to life and health, (often, as describedlater, disproportionate to the size of risk). It is easy to find populationstatistics suggest i n g that I (at age 59) have a 62% chance of living past age80, bu t much harder to assess the risk that I wil l face an u nd es i rab l y lowquality of life at age 80, which is a risk not sharing the four features above.So keep in mind that “risks of deat h” are in fact qu i t e atypical amongst allthe risks we face.There are many ways to categorize risks. Amongst behaviors we perceiveas voluntary, some (e.g. bungee jumping off a bridge) we perceive as riskywhile others (e.g. goi n g to the movies) we don’t. Other risks – e.g. catchinginfluenza or being injured in an automobile accident – are associated withwhat we perceive as ordinary everyday life rather than explicitly voluntaryactivities. So one top-level categorization is to distinguish between “volun-5354CHAPTER 5. RISK TO INDIVIDUALS: PERCEPTION AND REALITYtary” and “involuntary” risks, keeping i n mind there is no sharp division.The point is that it generally makes sense to compare two voluntary risksor two involuntary risks, but often does not make sense to compare risksacross the two types.In class in 2011 I asked my stu d entsIf your roommate s aid “I am going to ]ldots” would the thoughtrisky immediately come to mind?The data shows the perce ntage saying risky.5%: learn to target sh oot at a gun club25%: learn to ride a horse25%: take cross-country road trip with pare nts40%: walk across c am pus alone at midnight50%: learn downhill skiing70%: try onli n e dating80%: buy motor bike for commute95%: try ecst as y.xxx comment on this data.To compare risks quantitatively we need some “medium of exchange”. Ifthe risks are purely financial, then we can use money; if the threat is purelya risk of death (rather than injury) we can use the notion of micromort inthe next section . Otherwise it is in pr act i ce difficult to measure “amount ofharm” on a numerical sc al e .To think quantitatively about involuntary risks, one st ar t in g point is toseparate the chance of a ty p ic al person suffering noticable harm, and theamount of harm caused i f they do suffer. Le t me call these likelihood andseverity. If we are able and willing to measure “amount of harm” on anumerical scale, then we can take “severity” as the mean amount of harmcaused when the ris k event occurs.In some settings it is conventional to make a technical quantitative def-inition of risk as the product(probability of harm) × (ex pec t ed amount of harm, if harmed)which is conceptually equi valent to considering the total harm to the wholepopulation. But for a descriptive view of risks as seen by an individual, itseems better to separate t h e two components. In particular, this allows oneto compare risks via a scatter diagram of (likelihood, seve r ity). However,5.1. SOME DATA ON RISKS 55if the amount of harm is very variable then the exp ec t ati on is not such auseful summary statistic.Section 5.1 gives some data about risks. In class I refer to the 2002Ropeik - Gray book Risk. A practical guide for deciding what’s really safeand what’s really dangerous in the world around you. This has 6 -8 pagechapters on each of 48 specific risks, written in a fixed format and sum-marized by “likelihood” and “consequences” of each risk. A natural courseproject is to seek data and to write a report on another risk in the samestyle.Section 5.2, based on the 2010 Ropeik book How Risky is it Really?,describes psychol ogi cal factors which can make a risk seem more threateningor less threatening than it really is.Section 5.3 concerns presentation of statistical data to the public, basedon an article on the Understanding Uncertainty site. And section 5.4 in-troduces the economic and public policy aspects of risk, starting from anarticle by Cameron on the unwise choice of phrase statistical value of life.5.1 Some data on risksIn an article1describing his “Physics for Future Presidents” course, twicevoted “best class on campus” by Berkeley undergraduates, Richard MullerwritesUnlike physics students, liberal arts students do not mind learn-ing numbers and facts. They are empowered when they knowthings, such as what really happened at Chernobyl, how manypeople died of cancer at Hiroshima, what spy satellites can real l ydo (and not do), what Moore’s Law is ( most students have neve rheard of it), and what the differences are between MRI, CAT,and PET scans. I tell them that whatever their point of view,knowledge will help them. They will be able t o win argumentswith their friends and parents. They seem to be particul arl yhappy about the latter.A useful way of comparing risks of death is via a micromort (W),definedas a 1 in a million chance of death. Here are some typical values from (W)for activities that increase the risk by one micromort, and their associatedcause of death:1In www.amac ad . o rg / pdfs/SLACweb.pdf56CHAPTER 5. RISK TO INDIVIDUALS: PERCEPTION AND REALITYsmoking 1.4 cigarettes (cancer, heart disease)spending 1 hour in a coal mine (black lung disease)living 2 days in New York or Boston (air pollution)living 2 months in Denver (cancer from cosmic radiati on )drinking Miami water for 1 year (can ce r from chloroform)living 150 years within 20 miles of a nuclear power plant (cancer fromradiation)eating 100 charcoal-broiled steaks (cancer from benzopyrene)traveling 6 miles (10 km) by motorbike (accide nt)traveling 230 miles (370 km) by car (acci d ent)flying 1000