Judgments of risk and probability: the role of frequentistic information

✍ Scribed by Karl Halvor Teigen; Wibecke Brun; Rune Frydenlund

• Publisher: John Wiley and Sons
• Year: 1999
• Tongue: English
• Weight: 175 KB
• Volume: 12
• Category: Article
• ISSN: 0894-3257
• DOI: 10.1002/(sici)1099-0771(199906)12:2<123::aid-bdm313>3.0.co;2-5

✦ Synopsis

Risks and probabilities are often normatively de®ned in terms of the relative frequency of a target outcome. Psychological studies have shown that people are skilled in processing frequency information. Despite this, they often form discrepant estimates of probabilities and risks. The present paper traces this discrepancy to the perceived relevance of frequency information. In Study 1, student participants were asked to produce arguments for or against the riskiness of selected activities and substances. Less than 10% of the answers contained reference to harm or accident frequencies. In the next two studies, participants drew arrows between the concepts frequency, probability, and risk (Study 2), and winning frequency, winning probability, and quality of a successful soccer team (Study 3), arranged in a triad. In both diagrams, frequency was pictured as an eect rather than a cause. In Study 4, participants were to generate or choose explanations for verbal and numerical probability statements, and to describe what kind of evidence they would like to have to validate such statements. Here, frequentistic evidence was referred to or requested in 25±50% of the cases. It is concluded that the answers are more consistent with a dispositionist than a frequentist model of probabilities and risk.