A Study on Framing Effects in Risk Elicitation
2005 (English)In: Proceedings of the International Conference on Computational Intelligence for Modelling Control and Automation (CIMCA), Vienna, 2005, 2005, 689-694 p.Conference paper (Refereed)
Decision analysis tools are an effective way of structuring complex decision situations. However, their failure to incorporate reliable methods for elicitation is a shortcoming that needs to be dealt with. Since different elicitation methods have shown to yield different results, it is important to more thoroughly emphasize on aspects that can reduce biased results. The development of methods that explicitly recognize framing problems and aim to reduce these effects are needed. This study deals with framing problems within elicitation and how to reduce discrepancies between normative and descriptive behaviour in elicited risk data. The results indicate that the extra transitional state in one of the presentation formats, here referred to as Trade for, generated data that deviated more from normative rules when participants experienced gain prospects. On the other hand, for loss prospects the format more in line with normative rules depended on the presentation order of probabilities.
Place, publisher, year, edition, pages
2005. 689-694 p.
Research subject Computer and Systems Sciences
IdentifiersURN: urn:nbn:se:su:diva-38030DOI: 10.1109/CIMCA.2005.1631344ISBN: 0-7695-2504-0OAI: oai:DiVA.org:su-38030DiVA: diva2:305863
The International Conference on Computational Intelligence for Modelling Control and Automation (CIMCA), November 28 - November 30, 2005, Vienna, Austria