Disaster Training Needs Assessment: Understanding Households’ Perceptions on Stakeholder Roles and Protective Actions to Seismic Hazard in Sichuan, China

Research on debunking erroneous information suggests that programs for training people about how to respond during earthquakes are likely to be most effective if they explicitly address erroneous beliefs about the most appropriate protective actions. In addition, they should identify the characteristics of prospective information sources that are most likely to influence people's response actions. Accordingly, this study examined four seismic hazard response actions and their relationships with perceived stakeholder attributes, risk perceptions, and respondent characteristics. The results from 647 respondents showed that the most common response action intention was evacuation (4.77 out of a 1-5 scale, same as followings), followed by triangle of life (2.52), drop-cover-hold (2.47), and ignorance (1.66). Respondents rated ten stakeholder types on six characteristics—reliance, accessibility, amount of information, seismic hazard knowledge, trustworthiness, and responsibility for taking protective actions. In sum, respondents rated higher for responsibilities, trustworthiness, and expertise than for accessibilities and amount of information, indicating quality is more important than quantity. Correlation and regression analyses revealed that protective action was negatively correlated to stakeholders’ responsibility. Respondents who had higher concerns on the amount of information would be more likely to ignore the shaking, whereas respondents who favored easy-accessed information sources and believed stakeholders’ seismic hazard knowledge, would be more likely to take inappropriate protective action (e.g., triangle of life). Finally, immediate evacuation, which has been recognized as the easiest protective action, but is not recommended in an earthquake, was only correlated to risk perceptions, but not any of stakeholder attributes. The findings of this study suggest that risk communicators should consider the quality of information flow as well as individuals’ information reliance behaviors.