Risk Appraisal Definition
The term risk appraisal refers to an evaluation of the chances that a future event may occur. Similar terms include risk assessment, risk perception, perceived likelihood, and perception of vulnerability. One might appraise the risk of globally significant events (e.g., terrorism, natural disasters) as well as those that are personally relevant (e.g., losing a child, developing a disease). People’s beliefs about risk influence a wide variety of decisions and behaviors in many life domains including careers, relationships, and health. As examples, college students are relatively less likely to choose careers in which the chances of getting a job are low, and women are unlikely to get mammographies if they do not feel at risk for breast cancer. If risk appraisals are incorrect, they may lead to faulty decision making and counterproductive behavior, necessitating the development of strategies to correct these appraisals. Laypeople evaluate risk very differently than experts do, and such differences can have public policy implications.
Risk Appraisal Measurement
A common method of assessing perceptions of risk is to ask individuals whether an event will or will not happen, or have them estimate the chances it will happen on a percentage (0%-100%) scale. One problem with the latter approach is that people often have difficulty thinking about risk in numerical terms and think about probabilities differently than experts do. For example, when people estimate that an event has a 50% chance of happening, what they really mean is that it might or might not happen and they are not sure. Several other numerical measures attempt to minimize this problem, such as a magnifier scale in which the lower end of a percentage scale (between 0% and 1%) is divided into smaller units to encourage respondents to use this part of the scale for rare events. Another approach is to have individuals make relative judgments, such as how the risk of one event compares with that of another (e.g., having cancer vs. having heart disease) or how one person’s risk compares with that of another person or persons. Still another is to use a verbal scale (ranging, for example, from “very unlikely” to “very likely”). Any of these measures can be designed to be conditional on some behavior (e.g., “If you continue to smoke, what are your chances of getting lung cancer?”). Affective risk appraisals such as worry or the feeling that something will happen can also be assessed.
The choice of measures can greatly influence the findings of a given study. For example, people’s verbal risk appraisals are more sensitive to new information than are their numerical risk appraisals, and people may be pessimistically biased regarding their absolute risk of an event yet optimistically biased about their risk relative to that of their peers. Affective perceptions of vulnerability (e.g., “feeling at risk”) are sometimes more predictive of behaviors such as vaccination than are cognitive perceptions of vulnerability.
Errors in Risk Appraisal
People often make errors when appraising their risk. They tend to overestimate small risks and underestimate large risks, and perceive positive events as more likely to occur than negative events. A disproportionate number of individuals consider their level of personal risk to be lower than that of their peers, a bias called unrealistic optimism. People also tend to overestimate the risk of outcomes that have a low probability of occurring and yet result in major consequences (e.g., nuclear reactor explosions, airplane accidents). These events are often marked by a feeling of dread, lack of control, and the potential for extremely negative outcomes. Indeed, people often believe that high risk initiatives have low benefit, whereas experts believe exactly the reverse (as in the case of nuclear energy).
People may be able to estimate the likelihood of a single event occurring (e.g., getting a job), but they often have trouble estimating the probability of a compound event (e.g., getting a job and being promoted). They also have trouble understanding how quickly risk accumulates; for example, smokers do not realize that their risk of lung cancer relative to that of nonsmokers gets substantially higher the more years they continue to smoke. People have trouble making decisions where the risk of one outcome increases and the risk of another decreases (as is the case for many health therapies), and they have trouble considering all possible outcomes when assessing likelihood. For example, when a sample of smokers was asked how many smokers out of 100 would die of lung cancer, the average response was 42, but when asked the same question with a longer list of possible diseases, the average response for lung cancer was much lower. People’s risk perceptions are often constructed on the spot based on the way these risk perceptions are measured and the information available to the respondent at that moment. This may explain why risk appraisals are often not as predictive of behavior as one might expect.
Influences on Risk Appraisal
Many of the errors contaminating people’s risk appraisals can be explained by a set of basic psychological phenomena. Amos Tversky and Daniel Kahneman found that people rely on a variety of heuristics, or rules of thumb, when assessing the likelihood of events. An example of such a heuristic is the availability heuristic, whereby people estimate the probability of an event occurring based on how easy it is to think of an instance of that event. People tend to remember (and thus overestimate) events that happened relative to events that did not happen and are likely to remember vivid events more than mundane events. For instance, people may overestimate the number of times that disclosing personal information on a date backfired, yet suppress memory of many times that it did not. Highly publicized events such as airplane crashes raise risk appraisals for air travel, despite the fact that thousands more die in less publicized car accidents. People also engage in egocentrism, which means that thoughts about the self are more prominent than are thoughts about other people. As a result, when comparing their risk with that of others, people rely on the abundance of information they have about themselves, leading to errors like unrealistic optimism.
People think differently about frequencies than they do about proportions. For example, upon hearing that 10 individuals (of 100) were infected with a communicable disease, people worry more than if they hear that 10% of this group was infected, despite the statistics being equivalent. People seem to focus exclusively on the 10 people who might have been infected. People also fail to acknowledge the actual chances of an event when assessing their own (or another person’s) risk. For example, when determining whether a young woman is anorexic, people might compare her build and symptoms with that of other anorexic women and pay little attention to the actual prevalence of anorexia (which is lower than that of other health problems with similar symptoms). Such a bias results from a heuristic called the representativeness heuristic, which is used to make judgments of similarity.
People tend to judge risks higher when sad, and lower when angry or happy. Their motivation to believe that good things will happen to them and bad things will not often color their risk perceptions in a self-enhancing direction. As a result, they may be resistant to information designed to increase their risk perceptions, which explains why many health promotion campaigns are unsuccessful. People high in self-esteem, extraversion, dispositional optimism, or sensation-seeking tend to estimate their personal risks as relatively lower, whereas those who are depressed tend to be more pessimistic (though some evidence suggests that depressed people are more realistic about their risk than are nondepressed people). Contrary to popular belief, adults and adolescents do not vary greatly in their appraisals of the risk of various activities, and even older adults have been found to be unrealistically optimistic about their chances of experiencing negative events relative to peers.
Improving Risk Appraisal
Given the many errors in the way people appraise risk, and given the association of risk appraisals with important decisions and behaviors, it is important to devise ways to improve the accuracy of risk appraisals. The increasing use of computer aids and the Internet will be helpful in this regard. Several Web sites provide personalized risk information to individuals about their chances of getting a disease, and decision aids are now available to patients who need to appraise the risk of competing options (e.g., watchful waiting vs. surgery) when making important medical decisions. These decision aids collect information about patients’ values (such as whether years of life matter more than quality of life) to help them understand their risk and to make informed decisions. Evidence also indicates that training in the social sciences, and psychology in particular, improves the ability to reason and make probabilistic judgments, which is necessary to appraise risk accurately. If made available to a wider audience, similar training may reduce errors in risk appraisal on a wider scale.
- Slovic, P. (1987). Perception of risk. Science, 236, 280-285.
- Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185, 1124-1131.
- Weinstein, N. D. (2003). Exploring the links between risk perceptions and preventive health behavior. In J. Suls & K. A. Wallston (Eds.), Social psychological foundations of health and illness (pp. 22-53). Malden, MA: Blackwell.