Anticipating Risks in Science and Engineering Flashcards
Which of the following examples refer to subjective probabilities? Mark all correct options.
a) When someone says, since the opposing team’s key player is missing we will probably win today, it marks a subjective probability assessment.
b) The subjective probability of an event E can be assessed by considering what kind of bets an individual agent would be willing to participate in.
c) An example of a subjective probability assessment is when I calculate the probability that a fair (six-sided) dice will yield the value “three” when I throw it one time.
d) Subjective probability is the frequency with which event E occurs.
Answer: a), b)
Which of the following claims are correct regarding decision-making under certainty/uncertainty? Mark all correct options.
a) In a decision under certainty there is a small – practically non-existent – risk that the believed outcome does not occur.
b) If we know the probabilities of some possible outcomes, but lack knowledge about what all possible outcomes are, we are talking about decision-making under ignorance.
c) A necessary requirement for something to be a decision-making under risk is that the probability of each possible outcome is known.
d) Deep uncertainty is when one lacks knowledge about what all possible outcomes are.
Answer: c), d)
General Feedback
Remember: you distinguish the different levels of uncertainty in decision-making by assessing what is known about the possible outcomes and their probabilities.
Also, note that this question is about the theoretical requirements for different levels of uncertainty. In practice, we might consider some decision-making situation as a decision under ignorance if it is sufficiently similar to such a situation, but in theory, it is only a situation of ignorance if the requirements in the definition of “decision under ignorance” are fully met. Of course, the same goes for decision-making under deep uncertainty, risk and certainty.
Which of the following claims are true about the notion of the statistical expected value of an uncertain event? Mark all correct claims.
a) Depending on the context, the aim is to choose the action with the lowest expected value or the action with the highest expected value, when applying this notion of risk.
b) According to this notion, the actual risk is the event that will occur (even though we are uncertain what it might be).
c) The notion of the statistical expected value combines different quantitative notions of risks in order to give each random variable a numerical value that is supposed to represent the probability-weighted average of all its possible values.
d) To quantify the statistical expected values of actions involving heterogeneous outcomes never leads to issues of comparability.
Answer: a), c)
What is the difference between risk assessment and risk management? Mark all correct options.
a) Risk assessment is typically free from values, whereas risk management is typically laden with values.
b) Risk assessment concerns the identification and evidential justification of models of the risks, whereas risk management concerns the evaluation of possible implications from the available actions and their risks.
c) Political institutions typically should deal with risk management, and not with risk assessment.
d) Risk management is arguably the policy maker’s area of expertise, whereas risk assessment is arguably the scientist’s area of expertise.
Answer: a), b), c), d)
Consider a case where the statistical expected values between two or more actions are the same. What problem(s) does such a case present? Mark all correct options.
a) Such cases show that it’s not sufficient to define ‘risk’ as ‘statistical expected value’.
b) In such a case, the notion of the statistical expected value can’t by itself guide one’s decision.
c) Such cases show that it’s not necessary to define ‘risk’ as ‘statistical expected value’.
d) Such a case does not really present a problem because the statistical expected values of two risks can never be the same value.
Answer: a), b)
General Feedback
Important assumptions made in this scenario is that either (1) the outcomes are homogeneous, or (2) if the outcomes are heterogeneous then there is some reasonable way to compare and put values on the outcomes.
Also, be careful about the distinction between necessity and sufficiency. As a hint, not both of these options are correct, but it might not be that easy to find out which one is.
What does the St. Petersburg paradox show? That is, what conclusions could be drawn from the St. Petersburg paradox? Mark all correct options.
a) The paradox shows that the expected-value model of risk does not account for the possibility that the utility of something valuable (e.g. money) is diminishing in proportion to its magnitude.
b) The paradox shows that, depending on what its being quantified, people’s tendency to be risk averse/risk-loving relative to what is valued ought to be accounted for.
c) The paradox shows that the expected value of risky actions is sometimes different from what the expected value model says.
d) The paradox shows that, sometimes, the expected-value model is unusable without incorporating subjective evaluations of risk. In other words, the model’s external validity is debatable.
Answer: a), b), d)
Should psychological properties that clearly influence subjective evaluations be included in risk evaluation? Mark the correct option.
a) Yes, such properties need to be taken into account in risk evaluations.
b) Experts currently disagree on how to answer this question.
c) No, such properties are irrelevant for objective risk evaluations.
Answer: b)
What are some unwanted consequences if one were to apply the precautionary principle too strictly and without taking context-specific considerations into account? Mark all correct options.
a) It could risk creating too strict thresholds that in the end create paralysis among decision makers and proposing parties.
b) It would yield the opportunity for some decisions maker(s) to negate projects based on insufficient evidence, while the real reason is that the projects aren’t in line with the respective decision maker(s)’ personal interests.
c) It could risk neglecting important projects that indeed are backed up by evidence pointing to the absence of a risk or its relatively low probability, by demanding “proofs”.
Answer: a), b), c)
General Feedback
For the true option(s), note that the described consequences don’t follow by necessity. Rather, they can be viewed as consequences that are more likely to occur under a very strict application of the precautionary principle.
What is the point of Nozick’s critique, i.e. what does his critique show? Mark all correct options.
a) Nozick’s critique shows that, in risk management, a normative framework must take into account the managing person(s)’ virtues. In other words, is the managing person(s) acting with a “good heart”?
b) Nozick’s critique targets normative frameworks whose underlying principles rest on the view that violating someone’s right(s) is unconditionally morally wrong. He shows that such frameworks get into trouble when actions have to be chosen under risk, ignorance, and deep uncertainty.
c) Nozick’s critique shows that the consequences of exposing someone to uncertain risks are worse than exposing them to certain outcomes (even if such outcomes are negative).
Answer: b)
General Feedback
To escape Nozick’s critique, one must assume a deontological framework of ethics and its principles, since his critique is directed towards such a framework. In other words, it’s not justified to defend deontology based on considerations that normally belong to some other ethical framework.
