Modeling different behaviors in disclosing risk perception

Area of Science:

• Statistics
• Risk Analysis
• Psychometrics

Background:

• Ordered scales are commonly used to gauge public risk perception across various domains.
• Observed responses may not accurately reflect true risk awareness due to respondent uncertainty or errors.

Purpose of the Study:

• To propose a statistical model for ordinal risk perception data that accounts for response inaccuracies.
• To improve the estimation of true risk awareness by modeling different response behaviors.

Main Methods:

• Development of a statistical model incorporating three response behaviors: accurate, uncertain (random), and inaccurate (biased).
• Statistical inference techniques applied without assuming model correctness.
• Case studies involving geo-hydrological and work-related stress risks.

Main Results:

• The proposed model provides a more nuanced understanding of risk perception data.
• Demonstrated applicability in real-world scenarios of environmental and occupational risks.
• Improved accuracy in estimating true risk awareness compared to standard methods.

Conclusions:

• The novel model enhances the analysis of ordinal risk perception data by addressing response variability.
• This methodology offers a robust framework for assessing risk awareness in diverse fields.
• The findings have implications for public health, environmental policy, and occupational safety.