Implicit attitudes toward risk: the construction and validation of the measurement method

Methods that are typically used to examine individual differences in risk attitudes (e.g. lotteries, dilemmas, questionnaires) require participants to explicitly declare their willingness to take risk. Therefore, they may be biased by the need for self-presentation or situational characteristics such as time pressure and cognitive constraints that lead to more spontaneous and automatic processing of risk-related information. The aim of this study was to construct an indirect measure of risk attitudes that is free of these methodological limitations. The method based on the Implicit Association Test shows high internal reliability and satisfactory stability over time. It correlates moderately with different explicit measures of risk attitudes that are related to sensation seeking. Finally, it is characterized by a high predictive power. Adding the implicit measure to the set of independent variables representing declarative evaluations of risk attitudes significantly improved the model predicting risky real-life behavior. We argue that the indirect assessment of risk attitudes presented in this paper may be used as an universal measure of people’s risk propensity that is free of biases related to self-presentation and situational factors.     

Public reactions to the 2013 Chinese H7N9 Influenza outbreak: perceptions of risk,stakeholders, and protective actions

This study used the Protective Action Decision model to guide a survey of 762 respondents in Anhui province, China during the 2013 Avian Influenza A (H7N9) outbreak. The data suggest that three types of psychological variables – risk perceptions, protective action perceptions, and stakeholder perceptions – influenced people’s behavioral expectations of adopting protective actions. In addition, the effects of demographic variables on behavioral expectations were quite variable, with some being unrelated to behavioral expectations, some being related but unmediated by the psychological variables, and others being either partially or completely mediated by the psychological variables. These results can help public health officials to communicate more effectively when encouraging people to protect themselves during an influenza emergency.