The Risk Format Effect On Risk Attitudes

Accumulated evidence showed that risk information format (e.g. numeric probability, numeric frequency, pie chart, bar chart) affected risk attitudes in medical decision making (Kelton, et al.,2010). However, few researches on risk format effect have been done in the economic decision field.As there were numerous formats representing risk information and the risk format effect on risk perception or risk attitudes depended on some other influential factors (e.g. the complexity of medical decision tasks and the magnitudes of the risks), researchers got inconsistent or even conflicting results on this effect. The present dissertation probed the risk format effect on risk attitudes in economic decision making. For simplicity, we adopted a classic, easy decision task as the experimental task and used only the pie chart and the numeric probability as the risk information formats and the50/50probability as the companion risk.Experiment1compared the risk format effect on risk attitudes between the pie chart and the numeric probability. The results indicated that, when the risk was shown with the numeric probability, one tended to be risk-aversive and risk-neutral when risk was shown with the pie chart.Experiment2and Experiment3examined the risk format effect in decision tasks under different psychological distances. Experiment2used the social distance as the psychological distance whereas Experiment3adopted the temporal distance as the psychological distance. Convergently, these two experiments showed that risk information format (numeric probability vs. pie chart) exerted the same effect on psychologically proximal and distal decision making. As is shown in Experiment1, Experiment2and Experiment3indicated deciding with numeric-probability risk tended to be more risk-aversive than deciding with pie-chart risk. The social distance in Experiment2did not modify one’s risk attitudes while the temporal distance in Experiment3did. In Experiment3, compared with temporally distal decision, temporally proximal decision induced more risk-aversive choices. However, the effects caused by the temporal distance and the risk information format did not interact.In conclusion, people tended to be more risk-aversive when the risk was put with numeric probability than with pie chart. Moreover, deciding for now is more risk-aversive than deciding for2months later.